Authors: Jiankang Chen, Tong Zhang, Wei-Shi Zheng, Ruixuan Wang
Abstract: Out-of-distribution (OOD) detection is crucial in many real-world applications. However, intelligent models are often trained solely on in-distribution (ID) data, leading to overconfidence when misclassifying OOD data as ID classes. In this study, we propose a new learning framework which leverage simple Jigsaw-based fake OOD data and rich semantic embeddings (`anchors') from the ChatGPT description of ID knowledge to help guide the training of the image encoder. The learning framework can be flexibly combined with existing post-hoc approaches to OOD detection, and extensive empirical evaluations on multiple OOD detection benchmarks demonstrate that rich textual representation of ID knowledge and fake OOD knowledge can well help train a visual encoder for OOD detection. With the learning framework, new state-of-the-art performance was achieved on all the benchmarks. The code is available at \url{https://github.com/Cverchen/TagFog}.
Authors: Jiankang Chen, Ling Deng, Zhiyong Gan, Wei-Shi Zheng, Ruixuan Wang
Abstract: Out-of-Distribution (OOD) detection is crucial when deploying machine learning models in open-world applications. The core challenge in OOD detection is mitigating the model's overconfidence on OOD data. While recent methods using auxiliary outlier datasets or synthesizing outlier features have shown promising OOD detection performance, they are limited due to costly data collection or simplified assumptions. In this paper, we propose a novel OOD detection framework FodFoM that innovatively combines multiple foundation models to generate two types of challenging fake outlier images for classifier training. The first type is based on BLIP-2's image captioning capability, CLIP's vision-language knowledge, and Stable Diffusion's image generation ability. Jointly utilizing these foundation models constructs fake outlier images which are semantically similar to but different from in-distribution (ID) images. For the second type, GroundingDINO's object detection ability is utilized to help construct pure background images by blurring foreground ID objects in ID images. The proposed framework can be flexibly combined with multiple existing OOD detection methods. Extensive empirical evaluations show that image classifiers with the help of constructed fake images can more accurately differentiate real OOD images from ID ones. New state-of-the-art OOD detection performance is achieved on multiple benchmarks. The code is available at \url{https://github.com/Cverchen/ACMMM2024-FodFoM}.
Authors: Ebubechukwu Ike
Abstract: While DALL-E 3 has gained popularity for its ability to generate creative and complex images from textual descriptions, its application in the domain of style transfer remains slightly underexplored. This project investigates the integration of DALL-E 3 with traditional neural style transfer techniques to assess the impact of generated style images on the quality of the final output. DALL-E 3 was employed to generate style images based on the descriptions provided and combine these with the Magenta Arbitrary Image Stylization model. This integration is evaluated through metrics such as the Structural Similarity Index Measure (SSIM) and Peak Signal-to-Noise Ratio (PSNR), as well as processing time assessments. The findings reveal that DALL-E 3 significantly enhances the diversity and artistic quality of stylized images. Although this improvement comes with a slight increase in style transfer time, the data shows that this trade-off is worthwhile because the overall processing time with DALL-E 3 is about 2.5 seconds faster than traditional methods, making it both an efficient and visually superior option.
Authors: Shahran Rahman Alve
Abstract: This project aims to develop a robust video surveillance system, which can segment videos into smaller clips based on the detection of activities. It uses CCTV footage, for example, to record only major events-like the appearance of a person or a thief-so that storage is optimized and digital searches are easier. It utilizes the latest techniques in object detection and tracking, including Convolutional Neural Networks (CNNs) like YOLO, SSD, and Faster R-CNN, as well as Recurrent Neural Networks (RNNs) and Long Short-Term Memory networks (LSTMs), to achieve high accuracy in detection and capture temporal dependencies. The approach incorporates adaptive background modeling through Gaussian Mixture Models (GMM) and optical flow methods like Lucas-Kanade to detect motions. Multi-scale and contextual analysis are used to improve detection across different object sizes and environments. A hybrid motion segmentation strategy combines statistical and deep learning models to manage complex movements, while optimizations for real-time processing ensure efficient computation. Tracking methods, such as Kalman Filters and Siamese networks, are employed to maintain smooth tracking even in cases of occlusion. Detection is improved on various-sized objects for multiple scenarios by multi-scale and contextual analysis. Results demonstrate high precision and recall in detecting and tracking objects, with significant improvements in processing times and accuracy due to real-time optimizations and illumination-invariant features. The impact of this research lies in its potential to transform video surveillance, reducing storage requirements and enhancing security through reliable and efficient object detection and tracking.
Authors: Hidehisa Arai, Keishi Ishihara, Tsubasa Takahashi, Yu Yamaguchi
Abstract: World models have emerged as promising neural simulators for autonomous driving, with the potential to supplement scarce real-world data and enable closed-loop evaluations. However, current research primarily evaluates these models based on visual realism or downstream task performance, with limited focus on fidelity to specific action instructions - a crucial property for generating targeted simulation scenes. Although some studies address action fidelity, their evaluations rely on closed-source mechanisms, limiting reproducibility. To address this gap, we develop an open-access evaluation framework, ACT-Bench, for quantifying action fidelity, along with a baseline world model, Terra. Our benchmarking framework includes a large-scale dataset pairing short context videos from nuScenes with corresponding future trajectory data, which provides conditional input for generating future video frames and enables evaluation of action fidelity for executed motions. Furthermore, Terra is trained on multiple large-scale trajectory-annotated datasets to enhance action fidelity. Leveraging this framework, we demonstrate that the state-of-the-art model does not fully adhere to given instructions, while Terra achieves improved action fidelity. All components of our benchmark framework will be made publicly available to support future research.
Authors: Junno Yun, Mehmet Ak\c{c}akaya
Abstract: Infrared (IR) imaging is commonly used in various scenarios, including autonomous driving, fire safety and defense applications. Thus, semantic segmentation of such images is of great interest. However, this task faces several challenges, including data scarcity, differing contrast and input channel number compared to natural images, and emergence of classes not represented in databases in certain scenarios, such as defense applications. Few-shot segmentation (FSS) provides a framework to overcome these issues by segmenting query images using a few labeled support samples. However, existing FSS models for IR images require paired visible RGB images, which is a major limitation since acquiring such paired data is difficult or impossible in some applications. In this work, we develop new strategies for FSS of IR images by using generative modeling and fusion techniques. To this end, we propose to synthesize auxiliary data to provide additional channel information to complement the limited contrast in the IR images, as well as IR data synthesis for data augmentation. Here, the former helps the FSS model to better capture the relationship between the support and query sets, while the latter addresses the issue of data scarcity. Finally, to further improve the former aspect, we propose a novel fusion ensemble module for integrating the two different modalities. Our methods are evaluated on different IR datasets, and improve upon the state-of-the-art (SOTA) FSS models.
Authors: Sadegh Nadimi, Vasileios Angelidakis, Sadaf Maramizonouz, Chao Zhang
Abstract: We introduce OCULAR, an innovative hardware and software solution for three-dimensional dynamic image analysis of fine particles. Current state-of-the art instruments for dynamic image analysis are largely limited to two-dimensional imaging. However, extensive literature has demonstrated that relying on a single two-dimensional projection for particle characterisation can lead to inaccuracies in many applications. Existing three-dimensional imaging technologies, such as computed tomography, laser scanning, and orthophotography, are limited to static objects. These methods are often not statistically representative and come with significant post-processing requirements, as well as the need for specialised imaging and computing resources. OCULAR addresses these challenges by providing a cost-effective solution for imaging continuous particle streams using a synchronised array of optical cameras. Particle shape characterisation is achieved through the reconstruction of their three-dimensional surfaces. This paper details the OCULAR methodology, evaluates its reproducibility, and compares its results against X-ray micro computed tomography, highlighting its potential for efficient and reliable particle analysis.
Authors: Ashley S. Dale, Mei Qiu, Foo Bin Che, Thomas Bsaibes, Lauren Christopher, Paul Salama
Abstract: Much effort has been made to explain and improve the success of transfer-based attacks (TBA) on black-box computer vision models. This work provides the first attempt at a priori prediction of attack success by identifying the presence of vulnerable features within target models. Recent work by Chen and Liu (2024) proposed the manifold attack model, a unifying framework proposing that successful TBA exist in a common manifold space. Our work experimentally tests the common manifold space hypothesis by a new methodology: first, projecting feature vectors from surrogate and target feature extractors trained on ImageNet onto the same low-dimensional manifold; second, quantifying any observed structure similarities on the manifold; and finally, by relating these observed similarities to the success of the TBA. We find that shared feature representation moderately correlates with increased success of TBA (\r{ho}= 0.56). This method may be used to predict whether an attack will transfer without information of the model weights, training, architecture or details of the attack. The results confirm the presence of shared feature representations between two feature extractors of different sizes and complexities, and demonstrate the utility of datasets from different target domains as test signals for interpreting black-box feature representations.
Authors: Hidir Yesiltepe, Tuna Han Salih Meral, Connor Dunlop, Pinar Yanardag
Abstract: In this work, we propose the first motion transfer approach in diffusion transformer through Mixture of Score Guidance (MSG), a theoretically-grounded framework for motion transfer in diffusion models. Our key theoretical contribution lies in reformulating conditional score to decompose motion score and content score in diffusion models. By formulating motion transfer as a mixture of potential energies, MSG naturally preserves scene composition and enables creative scene transformations while maintaining the integrity of transferred motion patterns. This novel sampling operates directly on pre-trained video diffusion models without additional training or fine-tuning. Through extensive experiments, MSG demonstrates successful handling of diverse scenarios including single object, multiple objects, and cross-object motion transfer as well as complex camera motion transfer. Additionally, we introduce MotionBench, the first motion transfer dataset consisting of 200 source videos and 1000 transferred motions, covering single/multi-object transfers, and complex camera motions.
Authors: Himanshu Mittal, Suvramalya Basak, Anjali Gautam
Abstract: Violence detection in surveillance videos is a critical task for ensuring public safety. As a result, there is increasing need for efficient and lightweight systems for automatic detection of violent behaviours. In this work, we propose an effective method which leverages human skeleton key-points to capture inherent properties of violence, such as rapid movement of specific joints and their close proximity. At the heart of our method is our novel Dynamic Interaction Feature Extraction Module (DIFEM) which captures features such as velocity, and joint intersections, effectively capturing the dynamics of violent behavior. With the features extracted by our DIFEM, we use various classification algorithms such as Random Forest, Decision tree, AdaBoost and k-Nearest Neighbor. Our approach has substantially lesser amount of parameter expense than the existing state-of-the-art (SOTA) methods employing deep learning techniques. We perform extensive experiments on three standard violence recognition datasets, showing promising performance in all three datasets. Our proposed method surpasses several SOTA violence recognition methods.
Authors: Sindhu Boddu, Arindam Mukherjee, Arindrajit Seal
Abstract: This paper presents a robust approach for object detection in aerial imagery using the YOLOv5 model. We focus on identifying critical objects such as ambulances, car crashes, police vehicles, tow trucks, fire engines, overturned cars, and vehicles on fire. By leveraging a custom dataset, we outline the complete pipeline from data collection and annotation to model training and evaluation. Our results demonstrate that YOLOv5 effectively balances speed and accuracy, making it suitable for real-time emergency response applications. This work addresses key challenges in aerial imagery, including small object detection and complex backgrounds, and provides insights for future research in automated emergency response systems.
Authors: Lucas Jaffe, Avideh Zakhor
Abstract: In person search, we detect and rank matches to a query person image within a set of gallery scenes. Most person search models make use of a feature extraction backbone, followed by separate heads for detection and re-identification. While pre-training methods for vision backbones are well-established, pre-training additional modules for the person search task has not been previously examined. In this work, we present the first framework for end-to-end person search pre-training. Our framework splits person search into object-centric and query-centric methodologies, and we show that the query-centric framing is robust to label noise, and trainable using only weakly-labeled person bounding boxes. Further, we provide a novel model dubbed Swap Path Net (SPNet) which implements both query-centric and object-centric training objectives, and can swap between the two while using the same weights. Using SPNet, we show that query-centric pre-training, followed by object-centric fine-tuning, achieves state-of-the-art results on the standard PRW and CUHK-SYSU person search benchmarks, with 96.4% mAP on CUHK-SYSU and 61.2% mAP on PRW. In addition, we show that our method is more effective, efficient, and robust for person search pre-training than recent backbone-only pre-training alternatives.
Authors: Bohan Li, Jiazhe Guo, Hongsi Liu, Yingshuang Zou, Yikang Ding, Xiwu Chen, Hu Zhu, Feiyang Tan, Chi Zhang, Tiancai Wang, Shuchang Zhou, Li Zhang, Xiaojuan Qi, Hao Zhao, Mu Yang, Wenjun Zeng, Xin Jin
Abstract: Generating high-fidelity, controllable, and annotated training data is critical for autonomous driving. Existing methods typically generate a single data form directly from a coarse scene layout, which not only fails to output rich data forms required for diverse downstream tasks but also struggles to model the direct layout-to-data distribution. In this paper, we introduce UniScene, the first unified framework for generating three key data forms - semantic occupancy, video, and LiDAR - in driving scenes. UniScene employs a progressive generation process that decomposes the complex task of scene generation into two hierarchical steps: (a) first generating semantic occupancy from a customized scene layout as a meta scene representation rich in both semantic and geometric information, and then (b) conditioned on occupancy, generating video and LiDAR data, respectively, with two novel transfer strategies of Gaussian-based Joint Rendering and Prior-guided Sparse Modeling. This occupancy-centric approach reduces the generation burden, especially for intricate scenes, while providing detailed intermediate representations for the subsequent generation stages. Extensive experiments demonstrate that UniScene outperforms previous SOTAs in the occupancy, video, and LiDAR generation, which also indeed benefits downstream driving tasks.
Authors: Qihang Fang, Chengcheng Tang, Bugra Tekin, Yanchao Yang
Abstract: Recent advancements in models linking natural language with human motions have shown significant promise in motion generation and editing based on instructional text. Motivated by applications in sports coaching and motor skill learning, we investigate the inverse problem: generating corrective instructional text, leveraging motion editing and generation models. We introduce a novel approach that, given a user's current motion (source) and the desired motion (target), generates text instructions to guide the user towards achieving the target motion. We leverage large language models to generate corrective texts and utilize existing motion generation and editing frameworks to compile datasets of triplets (source motion, target motion, and corrective text). Using this data, we propose a new motion-language model for generating corrective instructions. We present both qualitative and quantitative results across a diverse range of applications that largely improve upon baselines. Our approach demonstrates its effectiveness in instructional scenarios, offering text-based guidance to correct and enhance user performance.
Authors: Ali K. AlShami, Ananya Kalita, Ryan Rabinowitz, Khang Lam, Rishabh Bezbarua, Terrance Boult, Jugal Kalita
Abstract: As the Computer Vision community rapidly develops and advances algorithms for autonomous driving systems, the goal of safer and more efficient autonomous transportation is becoming increasingly achievable. However, it is 2024, and we still do not have fully self-driving cars. One of the remaining core challenges lies in addressing the novelty problem, where self-driving systems still struggle to handle previously unseen situations on the open road. With our Challenge of Out-Of-Label (COOOL) benchmark, we introduce a novel dataset for hazard detection, offering versatile evaluation metrics applicable across various tasks, including novelty-adjacent domains such as Anomaly Detection, Open-Set Recognition, Open Vocabulary, and Domain Adaptation. COOOL comprises over 200 collections of dashcam-oriented videos, annotated by human labelers to identify objects of interest and potential driving hazards. It includes a diverse range of hazards and nuisance objects. Due to the dataset's size and data complexity, COOOL serves exclusively as an evaluation benchmark.
Authors: Zixian Ma, Jianguo Zhang, Zhiwei Liu, Jieyu Zhang, Juntao Tan, Manli Shu, Juan Carlos Niebles, Shelby Heinecke, Huan Wang, Caiming Xiong, Ranjay Krishna, Silvio Savarese
Abstract: While open-source multi-modal language models perform well on simple question answering tasks, they often fail on complex questions that require multiple capabilities, such as fine-grained recognition, visual grounding, and reasoning, and that demand multi-step solutions. We present TACO, a family of multi-modal large action models designed to improve performance on such complex, multi-step, and multi-modal tasks. During inference, TACO produces chains-of-thought-and-action (CoTA), executes intermediate steps by invoking external tools such as OCR, depth estimation and calculator, then integrates both the thoughts and action outputs to produce coherent responses. To train TACO, we create a large dataset of over 1M synthetic CoTA traces generated with GPT-4o and Python programs. We then experiment with various data filtering and mixing techniques and obtain a final subset of 293K high-quality CoTA examples. This dataset enables TACO to learn complex reasoning and action paths, surpassing existing models trained on instruction tuning data with only direct answers. Our model TACO outperforms the instruction-tuned baseline across 8 benchmarks, achieving a 3.6% improvement on average, with gains of up to 15% in MMVet tasks involving OCR, mathematical reasoning, and spatial reasoning. Training on high-quality CoTA traces sets a new standard for complex multi-modal reasoning, highlighting the need for structured, multi-step instruction tuning in advancing open-source mutli-modal models' capabilities.
Authors: Muhammad Umar Farooq, Awais Khan, Ijaz Ul Haq, Khalid Mahmood Malik
Abstract: Trust in social media is a growing concern due to its ability to influence significant societal changes. However, this space is increasingly compromised by various types of deepfake multimedia, which undermine the authenticity of shared content. Although substantial efforts have been made to address the challenge of deepfake content, existing detection techniques face a major limitation in generalization: they tend to perform well only on specific types of deepfakes they were trained on.This dependency on recognizing specific deepfake artifacts makes current methods vulnerable when applied to unseen or varied deepfakes, thereby compromising their performance in real-world applications such as social media platforms. To address the generalizability of deepfake detection, there is a need for a holistic approach that can capture a broader range of facial attributes and manipulations beyond isolated artifacts. To address this, we propose a novel deepfake detection framework featuring an effective feature descriptor that integrates Deep identity, Behavioral, and Geometric (DBaG) signatures, along with a classifier named DBaGNet. Specifically, the DBaGNet classifier utilizes the extracted DBaG signatures, leveraging a triplet loss objective to enhance generalized representation learning for improved classification. Specifically, the DBaGNet classifier utilizes the extracted DBaG signatures and applies a triplet loss objective to enhance generalized representation learning for improved classification. To test the effectiveness and generalizability of our proposed approach, we conduct extensive experiments using six benchmark deepfake datasets: WLDR, CelebDF, DFDC, FaceForensics++, DFD, and NVFAIR. Specifically, to ensure the effectiveness of our approach, we perform cross-dataset evaluations, and the results demonstrate significant performance gains over several state-of-the-art methods.
Authors: Abulikemu Abuduweili, Chenyang Yuan, Changliu Liu, Frank Permenter
Abstract: The denoising process of diffusion models can be interpreted as a projection of noisy samples onto the data manifold. Moreover, the noise level in these samples approximates their distance to the underlying manifold. Building on this insight, we propose a novel method to enhance sample generation by aligning the estimated noise level with the true distance of noisy samples to the manifold. Specifically, we introduce a noise level correction network, leveraging a pre-trained denoising network, to refine noise level estimates during the denoising process. Additionally, we extend this approach to various image restoration tasks by integrating task-specific constraints, including inpainting, deblurring, super-resolution, colorization, and compressed sensing. Experimental results demonstrate that our method significantly improves sample quality in both unconstrained and constrained generation scenarios. Notably, the proposed noise level correction framework is compatible with existing denoising schedulers (e.g., DDIM), offering additional performance improvements.
Authors: Xinjie Sun, Boxiong Wei, Yalong Jiang, Liquan Mao, Qi Zhao
Abstract: Thyroid nodule segmentation in ultrasound images is crucial for accurate diagnosis and treatment planning. However, existing methods face challenges in segmentation accuracy, interpretability, and generalization, which hinder their performance. This letter proposes a novel framework, CLIP-TNseg, to address these issues by integrating a multimodal large model with a neural network architecture. CLIP-TNseg consists of two main branches: the Coarse-grained Branch, which extracts high-level semantic features from a frozen CLIP model, and the Fine-grained Branch, which captures fine-grained features using U-Net style residual blocks. These features are fused and processed by the prediction head to generate precise segmentation maps. CLIP-TNseg leverages the Coarse-grained Branch to enhance semantic understanding through textual and high-level visual features, while the Fine-grained Branch refines spatial details, enabling precise and robust segmentation. Extensive experiments on public and our newly collected datasets demonstrate its competitive performance. Our code and the original dataset are available at https://github.com/jayxjsun/CLIP-TNseg.
Authors: Surya N Reddy, Vaibhav Kurrey, Mayank Nagar, Gagan Raj Gupta
Abstract: Proper use of personal protective equipment (PPE) can save the lives of industry workers and it is a widely used application of computer vision in the large manufacturing industries. However, most of the applications deployed generate a lot of false alarms (violations) because they tend to generalize the requirements of PPE across the industry and tasks. The key to resolving this issue is to understand the action being performed by the worker and customize the inference for the specific PPE requirements of that action. In this paper, we propose a system that employs activity recognition models to first understand the action being performed and then use object detection techniques to check for violations. This leads to a 23% improvement in the F1-score compared to the PPE-based approach on our test dataset of 109 videos.
Authors: Hao Cheng, Erjia Xiao, Jiayan Yang, Jiahang Cao, Qiang Zhang, Jize Zhang, Kaidi Xu, Jindong Gu, Renjing Xu
Abstract: Current image generation models can effortlessly produce high-quality, highly realistic images, but this also increases the risk of misuse. In various Text-to-Image or Image-to-Image tasks, attackers can generate a series of images containing inappropriate content by simply editing the language modality input. Currently, to prevent this security threat, the various guard or defense methods that are proposed also focus on defending the language modality. However, in practical applications, threats in the visual modality, particularly in tasks involving the editing of real-world images, pose greater security risks as they can easily infringe upon the rights of the image owner. Therefore, this paper uses a method named typographic attack to reveal that various image generation models also commonly face threats in the vision modality. Furthermore, we also evaluate the defense performance of various existing methods when facing threats in the vision modality and uncover their ineffectiveness. Finally, we propose the Vision Modal Threats in Image Generation Models (VMT-IGMs) dataset, which would serve as a baseline for evaluating the vision modality vulnerability of various image generation models.
Authors: Haosong Peng, Tianyu Qi, Yufeng Zhan, Hao Li, Yalun Dai, Yuanqing Xia
Abstract: With the advancement of computer vision, the recently emerged 3D Gaussian Splatting (3DGS) has increasingly become a popular scene reconstruction algorithm due to its outstanding performance. Distributed 3DGS can efficiently utilize edge devices to directly train on the collected images, thereby offloading computational demands and enhancing efficiency. However, traditional distributed frameworks often overlook computational and communication challenges in real-world environments, hindering large-scale deployment and potentially posing privacy risks. In this paper, we propose Radiant, a hierarchical 3DGS algorithm designed for large-scale scene reconstruction that considers system heterogeneity, enhancing the model performance and training efficiency. Via extensive empirical study, we find that it is crucial to partition the regions for each edge appropriately and allocate varying camera positions to each device for image collection and training. The core of Radiant is partitioning regions based on heterogeneous environment information and allocating workloads to each device accordingly. Furthermore, we provide a 3DGS model aggregation algorithm that enhances the quality and ensures the continuity of models' boundaries. Finally, we develop a testbed, and experiments demonstrate that Radiant improved reconstruction quality by up to 25.7\% and reduced up to 79.6\% end-to-end latency.
Authors: Ruida Zhang, Chengxi Li, Chenyangguang Zhang, Xingyu Liu, Haili Yuan, Yanyan Li, Xiangyang Ji, Gim Hee Lee
Abstract: Realistic scene reconstruction in driving scenarios poses significant challenges due to fast-moving objects. Most existing methods rely on labor-intensive manual labeling of object poses to reconstruct dynamic objects in canonical space and move them based on these poses during rendering. While some approaches attempt to use 3D object trackers to replace manual annotations, the limited generalization of 3D trackers -- caused by the scarcity of large-scale 3D datasets -- results in inferior reconstructions in real-world settings. In contrast, 2D foundation models demonstrate strong generalization capabilities. To eliminate the reliance on 3D trackers and enhance robustness across diverse environments, we propose a stable object tracking module by leveraging associations from 2D deep trackers within a 3D object fusion strategy. We address inevitable tracking errors by further introducing a motion learning strategy in an implicit feature space that autonomously corrects trajectory errors and recovers missed detections. Experimental results on Waymo-NOTR datasets show we achieve state-of-the-art performance. Our code will be made publicly available.
Authors: Jiacheng Jiang, Yuan Meng, Chen Tang, Han Yu, Qun Li, Zhi Wang, Wenwu Zhu
Abstract: Research on loss surface geometry, such as Sharpness-Aware Minimization (SAM), shows that flatter minima improve generalization. Recent studies further reveal that flatter minima can also reduce the domain generalization (DG) gap. However, existing flatness-based DG techniques predominantly operate within a full-precision training process, which is impractical for deployment on resource-constrained edge devices that typically rely on lower bit-width representations (e.g., 4 bits, 3 bits). Consequently, low-precision quantization-aware training is critical for optimizing these techniques in real-world applications. In this paper, we observe a significant degradation in performance when applying state-of-the-art DG-SAM methods to quantized models, suggesting that current approaches fail to preserve generalizability during the low-precision training process. To address this limitation, we propose a novel Gradient-Adaptive Quantization-Aware Training (GAQAT) framework for DG. Our approach begins by identifying the scale-gradient conflict problem in low-precision quantization, where the task loss and smoothness loss induce conflicting gradients for the scaling factors of quantizers, with certain layers exhibiting opposing gradient directions. This conflict renders the optimization of quantized weights highly unstable. To mitigate this, we further introduce a mechanism to quantify gradient inconsistencies and selectively freeze the gradients of scaling factors, thereby stabilizing the training process and enhancing out-of-domain generalization. Extensive experiments validate the effectiveness of the proposed GAQAT framework. On PACS, our 3-bit and 4-bit models outperform direct DG-QAT integration by up to 4.5%. On DomainNet, the 4-bit model achieves near-lossless performance compared to full precision, with improvements of 1.39% (4-bit) and 1.06% (3-bit) over the SOTA QAT baseline.
Authors: Gengze Zhou, Yicong Hong, Zun Wang, Chongyang Zhao, Mohit Bansal, Qi Wu
Abstract: The academic field of learning instruction-guided visual navigation can be generally categorized into high-level category-specific search and low-level language-guided navigation, depending on the granularity of language instruction, in which the former emphasizes the exploration process, while the latter concentrates on following detailed textual commands. Despite the differing focuses of these tasks, the underlying requirements of interpreting instructions, comprehending the surroundings, and inferring action decisions remain consistent. This paper consolidates diverse navigation tasks into a unified and generic framework -- we investigate the core difficulties of sharing general knowledge and exploiting task-specific capabilities in learning navigation and propose a novel State-Adaptive Mixture of Experts (SAME) model that effectively enables an agent to infer decisions based on different-granularity language and dynamic observations. Powered by SAME, we present a versatile agent capable of addressing seven navigation tasks simultaneously that outperforms or achieves highly comparable performance to task-specific agents.
Authors: Arturo Miguel Russell Bernal, Jane Cleland-Huang, Walter Scheirer
Abstract: The success of Emergency Response (ER) scenarios, such as search and rescue, is often dependent upon the prompt location of a lost or injured person. With the increasing use of small Unmanned Aerial Systems (sUAS) as "eyes in the sky" during ER scenarios, efficient detection of persons from aerial views plays a crucial role in achieving a successful mission outcome. Fatigue of human operators during prolonged ER missions, coupled with limited human resources, highlights the need for sUAS equipped with Computer Vision (CV) capabilities to aid in finding the person from aerial views. However, the performance of CV models onboard sUAS substantially degrades under real-life rigorous conditions of a typical ER scenario, where person search is hampered by occlusion and low target resolution. To address these challenges, we extracted images from the NOMAD dataset and performed a crowdsource experiment to collect behavioural measurements when humans were asked to "find the person in the picture". We exemplify the use of our behavioral dataset, Psych-ER, by using its human accuracy data to adapt the loss function of a detection model. We tested our loss adaptation on a RetinaNet model evaluated on NOMAD against increasing distance and occlusion, with our psychophysical loss adaptation showing improvements over the baseline at higher distances across different levels of occlusion, without degrading performance at closer distances. To the best of our knowledge, our work is the first human-guided approach to address the location task of a detection model, while addressing real-world challenges of aerial search and rescue. All datasets and code can be found at: https://github.com/ArtRuss/NOMAD.
Authors: Huajian Zeng, Maolin Gao, Daniel Cremers
Abstract: The interest in matching non-rigidly deformed shapes represented as raw point clouds is rising due to the proliferation of low-cost 3D sensors. Yet, the task is challenging since point clouds are irregular and there is a lack of intrinsic shape information. We propose to tackle these challenges by learning a new shape representation -- a per-point high dimensional embedding, in an embedding space where semantically similar points share similar embeddings. The learned embedding has multiple beneficial properties: it is aware of the underlying shape geometry and is robust to shape deformations and various shape artefacts, such as noise and partiality. Consequently, this embedding can be directly employed to retrieve high-quality dense correspondences through a simple nearest neighbor search in the embedding space. Extensive experiments demonstrate new state-of-the-art results and robustness in numerous challenging non-rigid shape matching benchmarks and show its great potential in other shape analysis tasks, such as segmentation.
Authors: Wenqing Wang, Yun Fu
Abstract: Text-to-3D generation is a valuable technology in virtual reality and digital content creation. While recent works have pushed the boundaries of text-to-3D generation, producing high-fidelity 3D objects with inefficient prompts and simulating their physics-grounded motion accurately still remain unsolved challenges. To address these challenges, we present an innovative framework that utilizes the Large Language Model (LLM)-refined prompts and diffusion priors-guided Gaussian Splatting (GS) for generating 3D models with accurate appearances and geometric structures. We also incorporate a continuum mechanics-based deformation map and color regularization to synthesize vivid physics-grounded motion for the generated 3D Gaussians, adhering to the conservation of mass and momentum. By integrating text-to-3D generation with physics-grounded motion synthesis, our framework renders photo-realistic 3D objects that exhibit physics-aware motion, accurately reflecting the behaviors of the objects under various forces and constraints across different materials. Extensive experiments demonstrate that our approach achieves high-quality 3D generations with realistic physics-grounded motion.
Authors: Andrea Avogaro, Luigi Capogrosso, Franco Fummi, Marco Cristani
Abstract: In the fast-fashion industry, overproduction and unsold inventory create significant environmental problems. Precise sales forecasts for unreleased items could drastically improve the efficiency and profits of industries. However, predicting the success of entirely new styles is difficult due to the absence of past data and ever-changing trends. Specifically, currently used deterministic models struggle with domain shifts when encountering items outside their training data. The recently proposed diffusion models address this issue using a continuous-time diffusion process. Specifically, these models enable us to predict the sales of new items, mitigating the domain shift challenges encountered by deterministic models. As a result, this paper proposes Dif4FF, a novel two-stage pipeline for New Fashion Product Performance Forecasting (NFPPF) that leverages the power of diffusion models conditioned on multimodal data related to specific clothes. Dif4FF first utilizes a multimodal score-based diffusion model to forecast multiple sales trajectories for various garments over time. The forecasts are refined using a powerful Graph Convolutional Network (GCN) architecture. By leveraging the GCN's capability to capture long-range dependencies within both the temporal and spatial data and seeking the optimal solution between these two dimensions, Dif4FF offers the most accurate and efficient forecasting system available in the literature for predicting the sales of new items. We tested Dif4FF on VISUELLE, the de facto standard for NFPPF, achieving new state-of-the-art results.
Authors: Diwen Wan, Yuxiang Wang, Ruijie Lu, Gang Zeng
Abstract: While novel view synthesis for dynamic scenes has made significant progress, capturing skeleton models of objects and re-posing them remains a challenging task. To tackle this problem, in this paper, we propose a novel approach to automatically discover the associated skeleton model for dynamic objects from videos without the need for object-specific templates. Our approach utilizes 3D Gaussian Splatting and superpoints to reconstruct dynamic objects. Treating superpoints as rigid parts, we can discover the underlying skeleton model through intuitive cues and optimize it using the kinematic model. Besides, an adaptive control strategy is applied to avoid the emergence of redundant superpoints. Extensive experiments demonstrate the effectiveness and efficiency of our method in obtaining re-posable 3D objects. Not only can our approach achieve excellent visual fidelity, but it also allows for the real-time rendering of high-resolution images.
Authors: Yuheng Xu, Taiping Zhang
Abstract: Traditional domain generalization methods often rely on domain alignment to reduce inter-domain distribution differences and learn domain-invariant representations. However, domain shifts are inherently difficult to eliminate, which limits model generalization. To address this, we propose an innovative framework that enhances data representation quality through probabilistic modeling and contrastive learning, reducing dependence on domain alignment and improving robustness under domain variations. Specifically, we combine deterministic features with uncertainty modeling to capture comprehensive feature distributions. Contrastive learning enforces distribution-level alignment by aligning the mean and covariance of feature distributions, enabling the model to dynamically adapt to domain variations and mitigate distribution shifts. Additionally, we design a frequency-domain-based structural enhancement strategy using discrete wavelet transforms to preserve critical structural details and reduce visual distortions caused by style variations. Experimental results demonstrate that the proposed framework significantly improves segmentation performance, providing a robust solution to domain generalization challenges in medical image segmentation.
Authors: Ye Wang, Yaxiong Wang, Guoshuai Zhao, Xueming Qian
Abstract: Continuous Generalized Category Discovery (C-GCD) aims to continually discover novel classes from unlabelled image sets while maintaining performance on old classes. In this paper, we propose a novel learning framework, dubbed Neighborhood Commonality-aware Evolution Network (NCENet) that conquers this task from the perspective of representation learning. Concretely, to learn discriminative representations for novel classes, a Neighborhood Commonality-aware Representation Learning (NCRL) is designed, which exploits local commonalities derived neighborhoods to guide the learning of representational differences between instances of different classes. To maintain the representation ability for old classes, a Bi-level Contrastive Knowledge Distillation (BCKD) module is designed, which leverages contrastive learning to perceive the learning and learned knowledge and conducts knowledge distillation. Extensive experiments conducted on CIFAR10, CIFAR100, and Tiny-ImageNet demonstrate the superior performance of NCENet compared to the previous state-of-the-art method. Particularly, in the last incremental learning session on CIFAR100, the clustering accuracy of NCENet outperforms the second-best method by a margin of 3.09\% on old classes and by a margin of 6.32\% on new classes. Our code will be publicly available at \href{https://github.com/xjtuYW/NCENet.git}{https://github.com/xjtuYW/NCENet.git}. \end{abstract}
URLs: https://github.com/xjtuYW/NCENet.git, https://github.com/xjtuYW/NCENet.git
Authors: Zehan Wang, Yuxuan Wei, Hui Yuan, Wei Zhang, Peng Li
Abstract: Three-dimensional (3D) point clouds are becoming more and more popular for representing 3D objects and scenes. Due to limited network bandwidth, efficient compression of 3D point clouds is crucial. To tackle this challenge, the Moving Picture Experts Group (MPEG) is actively developing the Geometry-based Point Cloud Compression (G-PCC) standard, incorporating innovative methods to optimize compression, such as the Region-Adaptive Hierarchical Transform (RAHT) nestled within a layer-by-layer octree-tree structure. Nevertheless, a notable problem still exists in RAHT, i.e., the proportion of zero residuals in the last few RAHT layers leads to unnecessary bitrate consumption. To address this problem, we propose an adaptive skip coding method for RAHT, which adaptively determines whether to encode the residuals of the last several layers or not, thereby improving the coding efficiency. In addition, we propose a rate-distortion cost calculation method associated with an adaptive Lagrange multiplier. Experimental results demonstrate that the proposed method achieves average Bj{\o}ntegaard rate improvements of -3.50%, -5.56%, and -4.18% for the Luma, Cb, and Cr components, respectively, on dynamic point clouds, when compared with the state-of-the-art G-PCC reference software under the common test conditions recommended by MPEG.
Authors: Lintong Du, Huazhen Liu, Yijia Zhang, ShuXin Liu, Yuan Qu, Zenghui Zhang, Jiamiao Yang
Abstract: Spatial phase unwrapping is a key technique for extracting phase information to obtain 3D morphology and other features. Modern industrial measurement scenarios demand high precision, large image sizes, and high speed. However, conventional methods struggle with noise resistance and processing speed. Current deep learning methods are limited by the receptive field size and sparse semantic information, making them ineffective for large size images. To address this issue, we propose a mutual self-distillation (MSD) mechanism and adaptive boosting ensemble segmenters to construct a universal multi-size phase unwrapping network (UMSPU). MSD performs hierarchical attention refinement and achieves cross-layer collaborative learning through bidirectional distillation, ensuring fine-grained semantic representation across image sizes. The adaptive boosting ensemble segmenters combine weak segmenters with different receptive fields into a strong one, ensuring stable segmentation across spatial frequencies. Experimental results show that UMSPU overcomes image size limitations, achieving high precision across image sizes ranging from 256*256 to 2048*2048 (an 8 times increase). It also outperforms existing methods in speed, robustness, and generalization. Its practicality is further validated in structured light imaging and InSAR. We believe that UMSPU offers a universal solution for phase unwrapping, with broad potential for industrial applications.
Authors: Saba Hesaraki, Morteza Akbari, Ramin Mousa
Abstract: Breast cancer stands as a prevalent cause of fatality among females on a global scale, with prompt detection playing a pivotal role in diminishing mortality rates. The utilization of ultrasound scans in the BUSI dataset for medical imagery pertaining to breast cancer has exhibited commendable segmentation outcomes through the application of UNet and UNet++ networks. Nevertheless, a notable drawback of these models resides in their inattention towards the temporal aspects embedded within the images. This research endeavors to enrich the UNet++ architecture by integrating LSTM layers and self-attention mechanisms to exploit temporal characteristics for segmentation purposes. Furthermore, the incorporation of a Multiscale Feature Extraction Module aims to grasp varied scale features within the UNet++. Through the amalgamation of our proposed methodology with data augmentation on the BUSI with GT dataset, an accuracy rate of 98.88%, specificity of 99.53%, precision of 95.34%, sensitivity of 91.20%, F1-score of 93.74, and Dice coefficient of 92.74% are achieved. These findings demonstrate competitiveness with cutting-edge techniques outlined in existing literature.
Authors: Itay Krispin-Avraham, Roy Orfaig, Ben-Zion Bobrovsky
Abstract: Object detection is a significant field in autonomous driving. Popular sensors for this task include cameras and LiDAR sensors. LiDAR sensors offer several advantages, such as insensitivity to light changes, like in a dark setting and the ability to provide 3D information in the form of point clouds, which include the ranges of objects. However, 3D detection methods, such as PointPillars, typically require high-power hardware. Additionally, most common spinning LiDARs are sparse and may not achieve the desired quality of object detection in front of the car. In this paper, we present the feasibility of performing real-time 3D object detection of cars using 3D point clouds from a LiDAR sensor, processed and deployed on a low-power Hailo-8 AI accelerator. The LiDAR sensor used in this study is the InnovizOne sensor, which captures objects in higher quality compared to spinning LiDAR techniques, especially for distant objects. We successfully achieved real-time inference at a rate of approximately 5Hz with a high accuracy of 0.91% F1 score, with only -0.2% degradation compared to running the same model on an NVIDIA GeForce RTX 2080 Ti. This work demonstrates that effective real-time 3D object detection can be achieved on low-cost, low-power hardware, representing a significant step towards more accessible autonomous driving technologies. The source code and the pre-trained models are available at https://github.com/AIROTAU/ PointPillarsHailoInnoviz/tree/main
Authors: Wenting Xu, Viorela Ila, Luping Zhou, Craig T. Jin
Abstract: The concept of function and affordance is a critical aspect of 3D scene understanding and supports task-oriented objectives. In this work, we develop a model that learns to structure and vary functional affordance across a 3D hierarchical scene graph representing the spatial organization of a scene. The varying functional affordance is designed to integrate with the varying spatial context of the graph. More specifically, we develop an algorithm that learns to construct a 3D hierarchical scene graph (3DHSG) that captures the spatial organization of the scene. Starting from segmented object point clouds and object semantic labels, we develop a 3DHSG with a top node that identifies the room label, child nodes that define local spatial regions inside the room with region-specific affordances, and grand-child nodes indicating object locations and object-specific affordances. To support this work, we create a custom 3DHSG dataset that provides ground truth data for local spatial regions with region-specific affordances and also object-specific affordances for each object. We employ a transformer-based model to learn the 3DHSG. We use a multi-task learning framework that learns both room classification and learns to define spatial regions within the room with region-specific affordances. Our work improves on the performance of state-of-the-art baseline models and shows one approach for applying transformer models to 3D scene understanding and the generation of 3DHSGs that capture the spatial organization of a room. The code and dataset are publicly available.
Authors: Mikolaj Czerkawski, Marcin Kluczek, J\k{e}drzej S. Bojanowski
Abstract: With the ever-increasing volumes of the Earth observation data present in the archives of large programmes such as Copernicus, there is a growing need for efficient vector representations of the underlying raw data. The approach of extracting feature representations from pretrained deep neural networks is a powerful approach that can provide semantic abstractions of the input data. However, the way this is done for imagery archives containing geospatial data has not yet been defined. In this work, an extension is proposed to an existing community project, Major TOM, focused on the provision and standardization of open and free AI-ready datasets for Earth observation. Furthermore, four global and dense embedding datasets are released openly and for free along with the publication of this manuscript, resulting in the most comprehensive global open dataset of geospatial visual embeddings in terms of covered Earth's surface.
Authors: Lasha Otarashvili, Tamilselvan Subramanian, Jason Holmberg, J. J. Levenson, Charles V. Stewart
Abstract: Recent work has established the ecological importance of developing algorithms for identifying animals individually from images. Typically, a separate algorithm is trained for each species, a natural step but one that creates significant barriers to wide-spread use: (1) each effort is expensive, requiring data collection, data curation, and model training, deployment, and maintenance, (2) there is little training data for many species, and (3) commonalities in appearance across species are not exploited. We propose an alternative approach focused on training multi-species individual identification (re-id) models. We construct a dataset that includes 49 species, 37K individual animals, and 225K images, using this data to train a single embedding network for all species. Our model employs an EfficientNetV2 backbone and a sub-center ArcFace loss function with dynamic margins. We evaluate the performance of this multispecies model in several ways. Most notably, we demonstrate that it consistently outperforms models trained separately on each species, achieving an average gain of 12.5% in top-1 accuracy. Furthermore, the model demonstrates strong zero-shot performance and fine-tuning capabilities for new species with limited training data, enabling effective curation of new species through both incremental addition of data to the training set and fine-tuning without the original data. Additionally, our model surpasses the recent MegaDescriptor on unseen species, averaging an 19.2% top-1 improvement per species and showing gains across all 33 species tested. The fully-featured code repository is publicly available on GitHub, and the feature extractor model can be accessed on HuggingFace for seamless integration with wildlife re-identification pipelines. The model is already in production use for 60+ species in a large-scale wildlife monitoring system.
Authors: Xiang Gao, Kai Lu
Abstract: The Segment Anything Model (SAM), originally built on a 2D Vision Transformer (ViT), excels at capturing global patterns in 2D natural images but struggles with 3D medical imaging modalities like CT and MRI. These modalities require capturing spatial information in volumetric space for tasks such as organ segmentation and tumor quantification. To address this challenge, we introduce RefSAM3D, which adapts SAM for 3D medical imaging by incorporating a 3D image adapter and cross-modal reference prompt generation. Our approach modifies the visual encoder to handle 3D inputs and enhances the mask decoder for direct 3D mask generation. We also integrate textual prompts to improve segmentation accuracy and consistency in complex anatomical scenarios. By employing a hierarchical attention mechanism, our model effectively captures and integrates information across different scales. Extensive evaluations on multiple medical imaging datasets demonstrate the superior performance of RefSAM3D over state-of-the-art methods. Our contributions advance the application of SAM in accurately segmenting complex anatomical structures in medical imaging.
Authors: Yusuke Hosoya, Masanori Suganuma, Takayuki Okatani
Abstract: Detecting objects occupying only small areas in an image is difficult, even for humans. Therefore, annotating small-size object instances is hard and thus costly. This study questions common sense by asking the following: is annotating small-size instances worth its cost? We restate it as the following verifiable question: can we detect small-size instances with a detector trained using training data free of small-size instances? We evaluate a method that upscales input images at test time and a method that downscales images at training time. The experiments conducted using the COCO dataset show the following. The first method, together with a remedy to narrow the domain gap between training and test inputs, achieves at least comparable performance to the baseline detector trained using complete training data. Although the method needs to apply the same detector twice to an input image with different scaling, we show that its distillation yields a single-path detector that performs equally well to the same baseline detector. These results point to the necessity of rethinking the annotation of training data for object detection.
Authors: Ming Tao, Bing-Kun Bao, Yaowei Wang, Changsheng Xu
Abstract: Large pretrained diffusion models have demonstrated impressive generation capabilities and have been adapted to various downstream tasks. However, unlike Large Language Models (LLMs) that can learn multiple tasks in a single model based on instructed data, diffusion models always require additional branches, task-specific training strategies, and losses for effective adaptation to different downstream tasks. This task-specific fine-tuning approach brings two drawbacks. 1) The task-specific additional networks create gaps between pretraining and fine-tuning which hinders the transfer of pretrained knowledge. 2) It necessitates careful additional network design, raising the barrier to learning and implementation, and making it less user-friendly. Thus, a question arises: Can we achieve a simple, efficient, and general approach to fine-tune diffusion models? To this end, we propose ONE-PIC. It enhances the inherited generative ability in the pretrained diffusion models without introducing additional modules. Specifically, we propose In-Visual-Context Tuning, which constructs task-specific training data by arranging source images and target images into a single image. This approach makes downstream fine-tuning closer to the pertaining, allowing our model to adapt more quickly to various downstream tasks. Moreover, we propose a Masking Strategy to unify different generative tasks. This strategy transforms various downstream fine-tuning tasks into predictions of the masked portions. The extensive experimental results demonstrate that our method is simple and efficient which streamlines the adaptation process and achieves excellent performance with lower costs. Code is available at https://github.com/tobran/ONE-PIC.
Authors: Gongfan Fang, Xinyin Ma, Xinchao Wang
Abstract: Transformer-based diffusion models have achieved significant advancements across a variety of generative tasks. However, producing high-quality outputs typically necessitates large transformer models, which result in substantial training and inference overhead. In this work, we investigate an alternative approach involving multiple experts for denoising, and introduce Remix-DiT, a novel method designed to enhance output quality at a low cost. The goal of Remix-DiT is to craft N diffusion experts for different denoising timesteps, yet without the need for expensive training of N independent models. To achieve this, Remix-DiT employs K basis models (where K < N) and utilizes learnable mixing coefficients to adaptively craft expert models. This design offers two significant advantages: first, although the total model size is increased, the model produced by the mixing operation shares the same architecture as a plain model, making the overall model as efficient as a standard diffusion transformer. Second, the learnable mixing adaptively allocates model capacity across timesteps, thereby effectively improving generation quality. Experiments conducted on the ImageNet dataset demonstrate that Remix-DiT achieves promising results compared to standard diffusion transformers and other multiple-expert methods. The code is available at https://github.com/VainF/Remix-DiT.
Authors: Abd Ur Rehman, Azka Rehman, Muhammad Usman, Abdullah Shahid, Sung-Min Gho, Aleum Lee, Tariq M. Khan, Imran Razzak
Abstract: Brain aging involves structural and functional changes and therefore serves as a key biomarker for brain health. Combining structural magnetic resonance imaging (sMRI) and functional magnetic resonance imaging (fMRI) has the potential to improve brain age estimation by leveraging complementary data. However, fMRI data, being noisier than sMRI, complicates multimodal fusion. Traditional fusion methods often introduce more noise than useful information, which can reduce accuracy compared to using sMRI alone. In this paper, we propose a novel multimodal framework for biological brain age estimation, utilizing a sex-aware adversarial variational autoencoder (SA-AVAE). Our framework integrates adversarial and variational learning to effectively disentangle the latent features from both modalities. Specifically, we decompose the latent space into modality-specific codes and shared codes to represent complementary and common information across modalities, respectively. To enhance the disentanglement, we introduce cross-reconstruction and shared-distinct distance ratio loss as regularization terms. Importantly, we incorporate sex information into the learned latent code, enabling the model to capture sex-specific aging patterns for brain age estimation via an integrated regressor module. We evaluate our model using the publicly available OpenBHB dataset, a comprehensive multi-site dataset for brain age estimation. The results from ablation studies and comparisons with state-of-the-art methods demonstrate that our framework outperforms existing approaches and shows significant robustness across various age groups, highlighting its potential for real-time clinical applications in the early detection of neurodegenerative diseases.
Authors: Gaurav Shrivastava, Abhinav Shrivastava
Abstract: Multi-step prediction models, such as diffusion and rectified flow models, have emerged as state-of-the-art solutions for generation tasks. However, these models exhibit higher latency in sampling new frames compared to single-step methods. This latency issue becomes a significant bottleneck when adapting such methods for video prediction tasks, given that a typical 60-second video comprises approximately 1.5K frames. In this paper, we propose a novel approach to modeling the multi-step process, aimed at alleviating latency constraints and facilitating the adaptation of such processes for video prediction tasks. Our approach not only reduces the number of sample steps required to predict the next frame but also minimizes computational demands by reducing the model size to one-third of the original size. We evaluate our method on standard video prediction datasets, including KTH, BAIR action robot, Human3.6M and UCF101, demonstrating its efficacy in achieving state-of-the-art performance on these benchmarks.
Authors: Xue Xian Zheng, M. M. Ur Rahma, Bilal Taha, Mudassir Masood, Dimitrios Hatzinakos, Tareq Al-Naffouri
Abstract: Camera-based photoplethysmography (PPG) obtained from smartphones has shown great promise for personalized healthcare and secure authentication. This paper presents a multimodal biometric system that integrates PPG signals extracted from videos with fingerprint data to enhance the accuracy of user verification. The system requires users to place their fingertip on the camera lens for a few seconds, allowing the capture and processing of unique biometric characteristics. Our approach employs a neural network with two structured state-space model (SSM) encoders to manage the distinct modalities. Fingerprint images are transformed into pixel sequences, and along with segmented PPG waveforms, they are input into the encoders. A cross-modal attention mechanism then extracts refined feature representations, and a distribution-oriented contrastive loss function aligns these features within a unified latent space. Experimental results demonstrate the system's superior performance across various evaluation metrics in both single-session and dual-session authentication scenarios.
Authors: Guy Levy, Nathan Liebmann
Abstract: Deepfakes are on the rise, with increased sophistication and prevalence allowing for high-profile social engineering attacks. Detecting them in the wild is therefore important as ever, giving rise to new approaches breaking benchmark records in this task. In line with previous work, we show that recently developed state-of-the-art detectors are susceptible to classical adversarial attacks, even in a highly-realistic black-box setting, putting their usability in question. We argue that crucial 'robust features' of deepfakes are in their higher semantics, and follow that with evidence that a detector based on a semantic embedding model is less susceptible to black-box perturbation attacks. We show that large visuo-lingual models like GPT-4o can perform zero-shot deepfake detection better than current state-of-the-art methods, and introduce a novel attack based on high-level semantic manipulation. Finally, we argue that hybridising low- and high-level detectors can improve adversarial robustness, based on their complementary strengths and weaknesses.
Authors: Xu Liu, Zhouhui Lian
Abstract: Remote Sensing Vision-Language Models (RS VLMs) have made much progress in the tasks of remote sensing (RS) image comprehension. While performing well in multi-modal reasoning and multi-turn conversations, the existing models lack pixel-level understanding and struggle with multi-image inputs. In this work, we propose RSUniVLM, a unified, end-to-end RS VLM designed for comprehensive vision understanding across multiple granularity, including image-level, region-level, and pixel-level tasks. RSUniVLM also performs effectively in multi-image analysis, with instances of change detection and change captioning. To enhance the model's ability to capture visual information at different levels without increasing model size, we design a novel architecture called Granularity-oriented Mixture of Experts to constraint the model to about 1 billion parameters. We also construct a large-scale RS instruction-following dataset based on a variety of existing datasets in both RS and general domain, encompassing various tasks such as object localization, visual question answering, and semantic segmentation. Substantial experiments have been conducted to validate the superiority of the proposed RSUniVLM up to state-of-the-art across various RS tasks. Code and model will be available at \href{https://github.com/xuliu-cyber/RSUniVLM}{here}.
Authors: Zihao Zhu, Hongbao Zhang, Guanzong Wu, Siwei Lyu, Baoyuan Wu
Abstract: Visual-textual inconsistency (VTI) evaluation plays a crucial role in cleansing vision-language data. Its main challenges stem from the high variety of image captioning datasets, where differences in content can create a range of inconsistencies (\eg, inconsistencies in scene, entities, entity attributes, entity numbers, entity interactions). Moreover, variations in caption length can introduce inconsistencies at different levels of granularity as well. To tackle these challenges, we design an adaptive evaluation framework, called Hierarchical and Multi-Grained Inconsistency Evaluation (HMGIE), which can provide multi-grained evaluations covering both accuracy and completeness for various image-caption pairs. Specifically, the HMGIE framework is implemented by three consecutive modules. Firstly, the semantic graph generation module converts the image caption to a semantic graph for building a structural representation of all involved semantic items. Then, the hierarchical inconsistency evaluation module provides a progressive evaluation procedure with a dynamic question-answer generation and evaluation strategy guided by the semantic graph, producing a hierarchical inconsistency evaluation graph (HIEG). Finally, the quantitative evaluation module calculates the accuracy and completeness scores based on the HIEG, followed by a natural language explanation about the detection results. Moreover, to verify the efficacy and flexibility of the proposed framework on handling different image captioning datasets, we construct MVTID, an image-caption dataset with diverse types and granularities of inconsistencies. Extensive experiments on MVTID and other benchmark datasets demonstrate the superior performance of the proposed HMGIE to current state-of-the-art methods.
Authors: Yan Zhang, Pengcheng Zheng, Chengxiao Zeng, Bin Xiao, Zhenghao Li, Xinbo Gao
Abstract: Remote Sensing (RS) image deblurring and Super-Resolution (SR) are common tasks in computer vision that aim at restoring RS image detail and spatial scale, respectively. However, real-world RS images often suffer from a complex combination of global low-resolution (LR) degeneration and local blurring degeneration. Although carefully designed deblurring and SR models perform well on these two tasks individually, a unified model that performs jointly RS image deblurring and super-resolution (JRSIDSR) task is still challenging due to the vital dilemma of reconstructing the global and local degeneration simultaneously. Additionally, existing methods struggle to capture the interrelationship between deblurring and SR processes, leading to suboptimal results. To tackle these issues, we give a unified theoretical analysis of RS images' spatial and blur degeneration processes and propose a dual-branch parallel network named AKMD-Net for the JRSIDSR task. AKMD-Net consists of two main branches: deblurring and super-resolution branches. In the deblurring branch, we design a pixel-adjustable kernel block (PAKB) to estimate the local and spatial-varying blur kernels. In the SR branch, a multi-domain attention block (MDAB) is proposed to capture the global contextual information enhanced with high-frequency details. Furthermore, we develop an adaptive feature fusion (AFF) module to model the contextual relationships between the deblurring and SR branches. Finally, we design an adaptive Wiener loss (AW Loss) to depress the prior noise in the reconstructed images. Extensive experiments demonstrate that the proposed AKMD-Net achieves state-of-the-art (SOTA) quantitative and qualitative performance on commonly used RS image datasets. The source code is publicly available at https://github.com/zpc456/AKMD-Net.
Authors: Saqib Javed, Ahmad Jarrar Khan, Corentin Dumery, Chen Zhao, Mathieu Salzmann
Abstract: Recent advancements in high-fidelity dynamic scene reconstruction have leveraged dynamic 3D Gaussians and 4D Gaussian Splatting for realistic scene representation. However, to make these methods viable for real-time applications such as AR/VR, gaming, and rendering on low-power devices, substantial reductions in memory usage and improvements in rendering efficiency are required. While many state-of-the-art methods prioritize lightweight implementations, they struggle in handling scenes with complex motions or long sequences. In this work, we introduce Temporally Compressed 3D Gaussian Splatting (TC3DGS), a novel technique designed specifically to effectively compress dynamic 3D Gaussian representations. TC3DGS selectively prunes Gaussians based on their temporal relevance and employs gradient-aware mixed-precision quantization to dynamically compress Gaussian parameters. It additionally relies on a variation of the Ramer-Douglas-Peucker algorithm in a post-processing step to further reduce storage by interpolating Gaussian trajectories across frames. Our experiments across multiple datasets demonstrate that TC3DGS achieves up to 67$\times$ compression with minimal or no degradation in visual quality.
Authors: Youssef Shoeb, Nazir Nayal, Azarm Nowzard, Fatma G\"uney, Hanno Gottschalk
Abstract: Detecting road obstacles is essential for autonomous vehicles to navigate dynamic and complex traffic environments safely. Current road obstacle detection methods typically assign a score to each pixel and apply a threshold to generate final predictions. However, selecting an appropriate threshold is challenging, and the per-pixel classification approach often leads to fragmented predictions with numerous false positives. In this work, we propose a novel method that leverages segment-level features from visual foundation models and likelihood ratios to predict road obstacles directly. By focusing on segments rather than individual pixels, our approach enhances detection accuracy, reduces false positives, and offers increased robustness to scene variability. We benchmark our approach against existing methods on the RoadObstacle and LostAndFound datasets, achieving state-of-the-art performance without needing a predefined threshold.
Authors: Sicong Tian, Haiyu Wu, Michael C. King, Kevin W. Bowyer
Abstract: One-to-many facial identification is documented to achieve high accuracy in the case where both the probe and the gallery are `mugshot quality' images. However, an increasing number of documented instances of wrongful arrest following one-to-many facial identification have raised questions about its accuracy. Probe images used in one-to-many facial identification are often cropped from frames of surveillance video and deviate from `mugshot quality' in various ways. This paper systematically explores how the accuracy of one-to-many facial identification is degraded by the person in the probe image choosing to wear dark sunglasses. We show that sunglasses degrade accuracy for mugshot-quality images by an amount similar to strong blur or noticeably lower resolution. Further, we demonstrate that the combination of sunglasses with blur or lower resolution results in even more pronounced loss in accuracy. These results have important implications for developing objective criteria to qualify a probe image for the level of accuracy to be expected if it used for one-to-many identification. To ameliorate the accuracy degradation caused by dark sunglasses, we show that it is possible to recover about 38% of the lost accuracy by synthetically adding sunglasses to all the gallery images, without model re-training. We also show that increasing the representation of wearing-sunglasses images in the training set can largely reduce the error rate. The image set assembled for this research will be made available to support replication and further research into this problem.
Authors: Ziyuan Qin, Dongjie Cheng, Haoyu Wang, Huahui Yi, Yuting Shao, Zhiyuan Fan, Kang Li, Qicheng Lao
Abstract: Contemporary Text-to-Image (T2I) models frequently depend on qualitative human evaluations to assess the consistency between synthesized images and the text prompts. There is a demand for quantitative and automatic evaluation tools, given that human evaluation lacks reproducibility. We believe that an effective T2I evaluation metric should accomplish the following: detect instances where the generated images do not align with the textual prompts, a discrepancy we define as the `hallucination problem' in T2I tasks; record the types and frequency of hallucination issues, aiding users in understanding the causes of errors; and provide a comprehensive and intuitive scoring that close to human standard. To achieve these objectives, we propose a method based on large language models (LLMs) for conducting question-answering with an extracted scene-graph and created a dataset with human-rated scores for generated images. From the methodology perspective, we combine knowledge-enhanced question-answering tasks with image evaluation tasks, making the evaluation metrics more controllable and easier to interpret. For the contribution on the dataset side, we generated 12,000 synthesized images based on 1,000 composited prompts using three advanced T2I models. Subsequently, we conduct human scoring on all synthesized images and prompt pairs to validate the accuracy and effectiveness of our method as an evaluation metric. All generated images and the human-labeled scores will be made publicly available in the future to facilitate ongoing research on this crucial issue. Extensive experiments show that our method aligns more closely with human scoring patterns than other evaluation metrics.
Authors: FNU Neha, Deepshikha Bhati, Deepak Kumar Shukla, Md Amiruzzaman
Abstract: Generative Adversarial Networks (GANs) have proven to be a powerful tool in generating artistic images, capable of mimicking the styles of renowned painters, such as Claude Monet. This paper introduces a tiered GAN model to progressively refine image quality through a multi-stage process, enhancing the generated images at each step. The model transforms random noise into detailed artistic representations, addressing common challenges such as instability in training, mode collapse, and output quality. This approach combines downsampling and convolutional techniques, enabling the generation of high-quality Monet-style artwork while optimizing computational efficiency. Experimental results demonstrate the architecture's ability to produce foundational artistic structures, though further refinements are necessary for achieving higher levels of realism and fidelity to Monet's style. Future work focuses on improving training methodologies and model complexity to bridge the gap between generated and true artistic images. Additionally, the limitations of traditional GANs in artistic generation are analyzed, and strategies to overcome these shortcomings are proposed.
Authors: Aditya Chinchure, Sahithya Ravi, Raymond Ng, Vered Shwartz, Boyang Li, Leonid Sigal
Abstract: The commonsense reasoning capabilities of vision-language models (VLMs), especially in abductive reasoning and defeasible reasoning, remain poorly understood. Most benchmarks focus on typical visual scenarios, making it difficult to discern whether model performance stems from keen perception and reasoning skills, or reliance on pure statistical recall. We argue that by focusing on atypical events in videos, clearer insights can be gained on the core capabilities of VLMs. Explaining and understanding such out-of-distribution events requires models to extend beyond basic pattern recognition and regurgitation of their prior knowledge. To this end, we introduce BlackSwanSuite, a benchmark for evaluating VLMs' ability to reason about unexpected events through abductive and defeasible tasks. Our tasks artificially limit the amount of visual information provided to models while questioning them about hidden unexpected events, or provide new visual information that could change an existing hypothesis about the event. We curate a comprehensive benchmark suite comprising over 3,800 MCQ, 4,900 generative and 6,700 yes/no tasks, spanning 1,655 videos. After extensively evaluating various state-of-the-art VLMs, including GPT-4o and Gemini 1.5 Pro, as well as open-source VLMs such as LLaVA-Video, we find significant performance gaps of up to 32% from humans on these tasks. Our findings reveal key limitations in current VLMs, emphasizing the need for enhanced model architectures and training strategies.
Authors: Ranjan Sapkota, Manoj Karkee
Abstract: In this study, we developed a customized instance segmentation model by integrating the Convolutional Block Attention Module (CBAM) with the YOLO11 architecture. This model, trained on a mixed dataset of dormant and canopy season apple orchard images, aimed to enhance the segmentation of tree trunks and branches under varying seasonal conditions throughout the year. The model was individually validated across dormant and canopy season images after training the YOLO11-CBAM on the mixed dataset collected over the two seasons. Additional testing of the model during pre-bloom, flower bloom, fruit thinning, and harvest season was performed. The highest recall and precision metrics were observed in the YOLO11x-seg-CBAM and YOLO11m-seg-CBAM respectively. Particularly, YOLO11m-seg with CBAM showed the highest precision of 0.83 as performed for the Trunk class in training, while without the CBAM, YOLO11m-seg achieved 0.80 precision score for the Trunk class. Likewise, for branch class, YOLO11m-seg with CBAM achieved the highest precision score value of 0.75 while without the CBAM, the YOLO11m-seg achieved a precision of 0.73. For dormant season validation, YOLO11x-seg exhibited the highest precision at 0.91. Canopy season validation highlighted YOLO11s-seg with superior precision across all classes, achieving 0.516 for Branch, and 0.64 for Trunk. The modeling approach, trained on two season datasets as dormant and canopy season images, demonstrated the potential of the YOLO11-CBAM integration to effectively detect and segment tree trunks and branches year-round across all seasonal variations. Keywords: YOLOv11, YOLOv11 Tree Detection, YOLOv11 Branch Detection and Segmentation, Machine Vision, Deep Learning, Machine Learning
Authors: Wenliang Zhong, Weizhi An, Feng Jiang, Hehuan Ma, Yuzhi Guo, Junzhou Huang
Abstract: Composed Image Retrieval (CIR) involves retrieving a target image based on a composed query of an image paired with text that specifies modifications or changes to the visual reference. CIR is inherently an instruction-following task, as the model needs to interpret and apply modifications to the image. In practice, due to the scarcity of annotated data in downstream tasks, Zero-Shot CIR (ZS-CIR) is desirable. While existing ZS-CIR models based on CLIP have shown promising results, their capability in interpreting and following modification instructions remains limited. Some research attempts to address this by incorporating Large Language Models (LLMs). However, these approaches still face challenges in effectively integrating multimodal information and instruction understanding. To tackle above challenges, we propose a novel embedding method utilizing an instruction-tuned Multimodal LLM (MLLM) to generate composed representation, which significantly enhance the instruction following capability for a comprehensive integration between images and instructions. Nevertheless, directly applying MLLMs introduces a new challenge since MLLMs are primarily designed for text generation rather than embedding extraction as required in CIR. To address this, we introduce a two-stage training strategy to efficiently learn a joint multimodal embedding space and further refining the ability to follow modification instructions by tuning the model in a triplet dataset similar to the CIR format. Extensive experiments on four public datasets: FashionIQ, CIRR, GeneCIS, and CIRCO demonstrates the superior performance of our model, outperforming state-of-the-art baselines by a significant margin. Codes are available at the GitHub repository.
Authors: Alex Rich, Noah Stier, Pradeep Sen, Tobias H\"ollerer
Abstract: The promise of unsupervised multi-view-stereo (MVS) is to leverage large unlabeled datasets, yet current methods underperform when training on difficult data, such as handheld smartphone videos of indoor scenes. Meanwhile, high-quality synthetic datasets are available but MVS networks trained on these datasets fail to generalize to real-world examples. To bridge this gap, we propose a semi-supervised learning framework that allows us to train on real and rendered images jointly, capturing structural priors from synthetic data while ensuring parity with the real-world domain. Central to our framework is a novel set of losses that leverages powerful existing monocular relative-depth estimators trained on the synthetic dataset, transferring the rich structure of this relative depth to the MVS predictions on unlabeled data. Inspired by perceptual image metrics, we compare the MVS and monocular predictions via a deep feature loss and a multi-scale statistical loss. Our full framework, which we call Prism, achieves large quantitative and qualitative improvements over current unsupervised and synthetic-supervised MVS networks. This is a best-case-scenario result, opening the door to using both unlabeled smartphone videos and photorealistic synthetic datasets for training MVS networks.
Authors: Qinchan (Wing), Li (Tina), Kenneth Chen (Tina), Changyue (Tina), Su, Qi Sun
Abstract: Diffusion models have shown unprecedented success in the task of text-to-image generation. While these models are capable of generating high-quality and realistic images, the complexity of sequential denoising has raised societal concerns regarding high computational demands and energy consumption. In response, various efforts have been made to improve inference efficiency. However, most of the existing efforts have taken a fixed approach with neural network simplification or text prompt optimization. Are the quality improvements from all denoising computations equally perceivable to humans? We observed that images from different text prompts may require different computational efforts given the desired content. The observation motivates us to present BudgetFusion, a novel model that suggests the most perceptually efficient number of diffusion steps before a diffusion model starts to generate an image. This is achieved by predicting multi-level perceptual metrics relative to diffusion steps. With the popular Stable Diffusion as an example, we conduct both numerical analyses and user studies. Our experiments show that BudgetFusion saves up to five seconds per prompt without compromising perceptual similarity. We hope this work can initiate efforts toward answering a core question: how much do humans perceptually gain from images created by a generative model, per watt of energy?
Authors: Jingxu Ng, Cheng Lv, Pu Zhao, Wei Niu, Juyi Lin, Yanzhi Wang
Abstract: Stable diffusion plays a crucial role in generating high-quality images. However, image generation is time-consuming and memory-intensive. To address this, stable-diffusion.cpp (Sdcpp) emerges as an efficient inference framework to accelerate the diffusion models. Although it is lightweight, the current implementation of ggml_conv_2d operator in Sdcpp is suboptimal, exhibiting both high inference latency and massive memory usage. To address this, in this work, we present an optimized version of Sdcpp leveraging the Winograd algorithm to accelerate 2D convolution operations, which is the primary bottleneck in the pipeline. By analyzing both dependent and independent computation graphs, we exploit the device's locality and parallelism to achieve substantial performance improvements. Our framework delivers correct end-to-end results across various stable diffusion models, including SDv1.4, v1.5, v2.1, SDXL, and SDXL-Turbo. Our evaluation results demonstrate a speedup up to 2.76x for individual convolutional layers and an inference speedup up to 4.79x for the overall image generation process, compared with the original Sdcpp. Homepage: https://github.com/SealAILab/stable-diffusion-cpp
Authors: Kaiwen Zha, Lijun Yu, Alireza Fathi, David A. Ross, Cordelia Schmid, Dina Katabi, Xiuye Gu
Abstract: Image tokenization, the process of transforming raw image pixels into a compact low-dimensional latent representation, has proven crucial for scalable and efficient image generation. However, mainstream image tokenization methods generally have limited compression rates, making high-resolution image generation computationally expensive. To address this challenge, we propose to leverage language for efficient image tokenization, and we call our method Text-Conditioned Image Tokenization (TexTok). TexTok is a simple yet effective tokenization framework that leverages language to provide high-level semantics. By conditioning the tokenization process on descriptive text captions, TexTok allows the tokenization process to focus on encoding fine-grained visual details into latent tokens, leading to enhanced reconstruction quality and higher compression rates. Compared to the conventional tokenizer without text conditioning, TexTok achieves average reconstruction FID improvements of 29.2% and 48.1% on ImageNet-256 and -512 benchmarks respectively, across varying numbers of tokens. These tokenization improvements consistently translate to 16.3% and 34.3% average improvements in generation FID. By simply replacing the tokenizer in Diffusion Transformer (DiT) with TexTok, our system can achieve a 93.5x inference speedup while still outperforming the original DiT using only 32 tokens on ImageNet-512. TexTok with a vanilla DiT generator achieves state-of-the-art FID scores of 1.46 and 1.62 on ImageNet-256 and -512 respectively. Furthermore, we demonstrate TexTok's superiority on the text-to-image generation task, effectively utilizing the off-the-shelf text captions in tokenization.
Authors: Shuzhao Xie, Jiahang Liu, Weixiang Zhang, Shijia Ge, Sicheng Pan, Chen Tang, Yunpeng Bai, Zhi Wang
Abstract: Effective compression technology is crucial for 3DGS to adapt to varying storage and transmission conditions. However, existing methods fail to address size constraints while maintaining optimal quality. In this paper, we introduce SizeGS, a framework that compresses 3DGS within a specified size budget while optimizing visual quality. We start with a size estimator to establish a clear relationship between file size and hyperparameters. Leveraging this estimator, we incorporate mixed precision quantization (MPQ) into 3DGS attributes, structuring MPQ in two hierarchical level -- inter-attribute and intra-attribute -- to optimize visual quality under the size constraint. At the inter-attribute level, we assign bit-widths to each attribute channel by formulating the combinatorial optimization as a 0-1 integer linear program, which can be efficiently solved. At the intra-attribute level, we divide each attribute channel into blocks of vectors, quantizing each vector based on the optimal bit-width derived at the inter-attribute level. Dynamic programming determines block lengths. Using the size estimator and MPQ, we develop a calibrated algorithm to identify optimal hyperparameters in just 10 minutes, achieving a 1.69$\times$ efficiency increase with quality comparable to state-of-the-art methods.
Authors: Leigang Qu, Haochuan Li, Wenjie Wang, Xiang Liu, Juncheng Li, Liqiang Nie, Tat-Seng Chua
Abstract: Large Multimodal Models (LMMs) have demonstrated impressive capabilities in multimodal understanding and generation, pushing forward advancements in text-to-image generation. However, achieving accurate text-image alignment for LMMs, particularly in compositional scenarios, remains challenging. Existing approaches, such as layout planning for multi-step generation and learning from human feedback or AI feedback, depend heavily on prompt engineering, costly human annotations, and continual upgrading, limiting flexibility and scalability. In this work, we introduce a model-agnostic iterative self-improvement framework (SILMM) that can enable LMMs to provide helpful and scalable self-feedback and optimize text-image alignment via Direct Preference Optimization (DPO). DPO can readily applied to LMMs that use discrete visual tokens as intermediate image representations; while it is less suitable for LMMs with continuous visual features, as obtaining generation probabilities is challenging. To adapt SILMM to LMMs with continuous features, we propose a diversity mechanism to obtain diverse representations and a kernel-based continuous DPO for alignment. Extensive experiments on three compositional text-to-image generation benchmarks validate the effectiveness and superiority of SILMM, showing improvements exceeding 30% on T2I-CompBench++ and around 20% on DPG-Bench.
Authors: Ao Wang, Fengyuan Sun, Hui Chen, Zijia Lin, Jungong Han, Guiguang Ding
Abstract: Multimodal Large Language Models (MLLMs) have recently demonstrated strong performance across a wide range of vision-language tasks, garnering significant attention in the computer vision. However, their efficient deployment remains a substantial challenge due to high computational costs and memory requirements. Recognizing the redundancy of information within the vision modality, recent studies have explored methods for compressing visual tokens in MLLMs to enhance efficiency in a training-free manner. Despite their effectiveness, existing methods like Fast rely on the attention between visual tokens and prompt text tokens as the importance indicator, overlooking the relevance to response text and thus introducing perception bias. In this paper, we demonstrate that in MLLMs, the [CLS] token in the visual encoder inherently knows which visual tokens are important for MLLMs. Building on this prior, we introduce a simple yet effective method for train-free visual token compression, called VTC-CLS. Firstly, it leverages the attention score of the [CLS] token on visual tokens as an importance indicator for pruning visual tokens. Besides, we also explore ensembling the importance scores derived by the [CLS] token from different layers to capture the key visual information more comprehensively. Extensive experiments demonstrate that our VTC-CLS achieves the state-of-the-art performance across various tasks compared with baseline methods. It also brings notably less computational costs in a training-free manner, highlighting its effectiveness and superiority. Code and models are available at \url{https://github.com/THU-MIG/VTC-CLS}.
Authors: Yuanbo Xiangli, Ruojin Cai, Hanyu Chen, Jeffrey Byrne, Noah Snavely
Abstract: Accurate 3D reconstruction is frequently hindered by visual aliasing, where visually similar but distinct surfaces (aka, doppelgangers), are incorrectly matched. These spurious matches distort the structure-from-motion (SfM) process, leading to misplaced model elements and reduced accuracy. Prior efforts addressed this with CNN classifiers trained on curated datasets, but these approaches struggle to generalize across diverse real-world scenes and can require extensive parameter tuning. In this work, we present Doppelgangers++, a method to enhance doppelganger detection and improve 3D reconstruction accuracy. Our contributions include a diversified training dataset that incorporates geo-tagged images from everyday scenes to expand robustness beyond landmark-based datasets. We further propose a Transformer-based classifier that leverages 3D-aware features from the MASt3R model, achieving superior precision and recall across both in-domain and out-of-domain tests. Doppelgangers++ integrates seamlessly into standard SfM and MASt3R-SfM pipelines, offering efficiency and adaptability across varied scenes. To evaluate SfM accuracy, we introduce an automated, geotag-based method for validating reconstructed models, eliminating the need for manual inspection. Through extensive experiments, we demonstrate that Doppelgangers++ significantly enhances pairwise visual disambiguation and improves 3D reconstruction quality in complex and diverse scenarios.
Authors: Tiancheng Li, Weijian Luo, Zhiyang Chen, Liyuan Ma, Guo-Jun Qi
Abstract: Proper guidance strategies are essential to get optimal generation results without re-training diffusion and flow-based text-to-image models. However, existing guidances either require specific training or strong inductive biases of neural network architectures, potentially limiting their applications. To address these issues, in this paper, we introduce Self-Guidance (SG), a strong diffusion guidance that neither needs specific training nor requires certain forms of neural network architectures. Different from previous approaches, the Self-Guidance calculates the guidance vectors by measuring the difference between the velocities of two successive diffusion timesteps. Therefore, SG can be readily applied for both conditional and unconditional models with flexible network architectures. We conduct intensive experiments on both text-to-image generation and text-to-video generations across flexible architectures including UNet-based models and diffusion transformer-based models. On current state-of-the-art diffusion models such as Stable Diffusion 3.5 and FLUX, SG significantly boosts the image generation performance in terms of FID, and Human Preference Scores. Moreover, we find that SG has a surprisingly positive effect on the generation of high-quality human bodies such as hands, faces, and arms, showing strong potential to overcome traditional challenges on human body generations with minimal effort. We will release our implementation of SG on SD 3.5 and FLUX models along with this paper.
Authors: Elay Dahan, Hedda Cohen Indelman, Angeles M. Perez-Agosto, Carmit Shiran, Gopal Avinash, Doron Shaked, Nati Daniel
Abstract: The use of synthetic images in medical imaging Artificial Intelligence (AI) solutions has been shown to be beneficial in addressing the limited availability of diverse, unbiased, and representative data. Despite the extensive use of synthetic image generation methods, controlling the semantics variability and context details remains challenging, limiting their effectiveness in producing diverse and representative medical image datasets. In this work, we introduce a scalable semantic and context-conditioned generative model, coined CSG (Context-Semantic Guidance). This dual conditioning approach allows for comprehensive control over both structure and appearance, advancing the synthesis of realistic and diverse ultrasound images. We demonstrate the ability of CSG to generate findings (pathological anomalies) in musculoskeletal (MSK) ultrasound images. Moreover, we test the quality of the synthetic images using a three-fold validation protocol. The results show that the synthetic images generated by CSG improve the performance of semantic segmentation models, exhibit enhanced similarity to real images compared to the baseline methods, and are undistinguishable from real images according to a Turing test. Furthermore, we demonstrate an extension of the CSG that allows enhancing the variability space of images by synthetically generating augmentations of anatomical geometries and textures.
Authors: Haoran Zhu, Chang Xu, Ruixiang Zhang, Fang Xu, Wen Yang, Haijian Zhang, Gui-Song Xia
Abstract: Tiny objects, with their limited spatial resolution, often resemble point-like distributions. As a result, bounding box prediction using point-level supervision emerges as a natural and cost-effective alternative to traditional box-level supervision. However, the small scale and lack of distinctive features of tiny objects make point annotations prone to noise, posing significant hurdles for model robustness. To tackle these challenges, we propose Point Teacher--the first end-to-end point-supervised method for robust tiny object detection in aerial images. To handle label noise from scale ambiguity and location shifts in point annotations, Point Teacher employs the teacher-student architecture and decouples the learning into a two-phase denoising process. In this framework, the teacher network progressively denoises the pseudo boxes derived from noisy point annotations, guiding the student network's learning. Specifically, in the first phase, random masking of image regions facilitates regression learning, enabling the teacher to transform noisy point annotations into coarse pseudo boxes. In the second phase, these coarse pseudo boxes are refined using dynamic multiple instance learning, which adaptively selects the most reliable instance from dynamically constructed proposal bags around the coarse pseudo boxes. Extensive experiments on three tiny object datasets (i.e., AI-TOD-v2, SODA-A, and TinyPerson) validate the proposed method's effectiveness and robustness against point location shifts. Notably, relying solely on point supervision, our Point Teacher already shows comparable performance with box-supervised learning methods. Codes and models will be made publicly available.
Authors: Yue Ma, Huantao Ren, Boyu Wang, Jingang Jin, Senem Velipasalar, Qinru Qiu
Abstract: Continual learning aims to update a model so that it can sequentially learn new tasks without forgetting previously acquired knowledge. Recent continual learning approaches often leverage the vision-language model CLIP for its high-dimensional feature space and cross-modality feature matching. Traditional CLIP-based classification methods identify the most similar text label for a test image by comparing their embeddings. However, these methods are sensitive to the quality of text phrases and less effective for classes lacking meaningful text labels. In this work, we rethink CLIP-based continual learning and introduce the concept of Label Vector Pool (LVP). LVP replaces text labels with training images as similarity references, eliminating the need for ideal text descriptions. We present three variations of LVP and evaluate their performance on class and domain incremental learning tasks. Leveraging CLIP's high dimensional feature space, LVP learning algorithms are task-order invariant. The new knowledge does not modify the old knowledge, hence, there is minimum forgetting. Different tasks can be learned independently and in parallel with low computational and memory demands. Experimental results show that proposed LVP-based methods outperform the current state-of-the-art baseline by a significant margin of 40.7%.
Authors: Shuwei Shi, Biao Gong, Xi Chen, Dandan Zheng, Shuai Tan, Zizheng Yang, Yuyuan Li, Jingwen He, Kecheng Zheng, Jingdong Chen, Ming Yang, Yinqiang Zheng
Abstract: The image-to-video (I2V) generation is conditioned on the static image, which has been enhanced recently by the motion intensity as an additional control signal. These motion-aware models are appealing to generate diverse motion patterns, yet there lacks a reliable motion estimator for training such models on large-scale video set in the wild. Traditional metrics, e.g., SSIM or optical flow, are hard to generalize to arbitrary videos, while, it is very tough for human annotators to label the abstract motion intensity neither. Furthermore, the motion intensity shall reveal both local object motion and global camera movement, which has not been studied before. This paper addresses the challenge with a new motion estimator, capable of measuring the decoupled motion intensities of objects and cameras in video. We leverage the contrastive learning on randomly paired videos and distinguish the video with greater motion intensity. Such a paradigm is friendly for annotation and easy to scale up to achieve stable performance on motion estimation. We then present a new I2V model, named MotionStone, developed with the decoupled motion estimator. Experimental results demonstrate the stability of the proposed motion estimator and the state-of-the-art performance of MotionStone on I2V generation. These advantages warrant the decoupled motion estimator to serve as a general plug-in enhancer for both data processing and video generation training.
Authors: Yugang Chang, Hongyu Chen, Fei Wang, Chengcheng Chen, Weiming Zeng
Abstract: This paper introduces the Maritime Ship Navigation Behavior Dataset (MID), designed to address challenges in ship detection within complex maritime environments using Oriented Bounding Boxes (OBB). MID contains 5,673 images with 135,884 finely annotated target instances, supporting both supervised and semi-supervised learning. It features diverse maritime scenarios such as ship encounters under varying weather, docking maneuvers, small target clustering, and partial occlusions, filling critical gaps in datasets like HRSID, SSDD, and NWPU-10. MID's images are sourced from high-definition video clips of real-world navigation across 43 water areas, with varied weather and lighting conditions (e.g., rain, fog). Manually curated annotations enhance the dataset's variety, ensuring its applicability to real-world demands in busy ports and dense maritime regions. This diversity equips models trained on MID to better handle complex, dynamic environments, supporting advancements in maritime situational awareness. To validate MID's utility, we evaluated 10 detection algorithms, providing an in-depth analysis of the dataset, detection results from various models, and a comparative study of baseline algorithms, with a focus on handling occlusions and dense target clusters. The results highlight MID's potential to drive innovation in intelligent maritime traffic monitoring and autonomous navigation systems. The dataset will be made publicly available at https://github.com/VirtualNew/MID\_DataSet.
Authors: Xuefeng Ni, Linshan Wu, Jiaxin Zhuang, Qiong Wang, Mingxiang Wu, Varut Vardhanabhuti, Lihai Zhang, Hanyu Gao, Hao Chen
Abstract: 3D medical image analysis is pivotal in numerous clinical applications. However, the scarcity of labeled data and limited generalization capabilities hinder the advancement of AI-empowered models. Radiology reports are easily accessible and can serve as weakly-supervised signals. However, large-scale vision-language pre-training (VLP) remains underexplored in 3D medical image analysis. Specifically, the insufficient investigation into multi-grained radiology semantics and their correlations across patients leads to underutilization of large-scale volume-report data. Considering intra-patient cross-modal semantic consistency and inter-patient semantic correlations, we propose a multi-task VLP method, MG-3D, pre-trained on large-scale data (47.1K), addressing the challenges by the following two aspects: 1) Establishing the correspondence between volume semantics and multi-grained medical knowledge of each patient with cross-modal global alignment and complementary modality-guided local reconstruction, ensuring intra-patient features of different modalities cohesively represent the same semantic content; 2) Correlating inter-patient visual semantics based on fine-grained report correlations across patients, and keeping sensitivity to global individual differences via contrastive learning, enhancing the discriminative feature representation. Furthermore, we delve into the scaling law to explore potential performance improvements. Comprehensive evaluations across nine uni- and cross-modal clinical tasks are carried out to assess model efficacy. Extensive experiments on both internal and external datasets demonstrate the superior transferability, scalability, and generalization of MG-3D, showcasing its potential in advancing feature representation for 3D medical image analysis. Code will be available: https://github.com/Xuefeng-Ni/MG-3D.
Authors: Leonid Antsfeld, Boris Chidlovskii
Abstract: We address the problem of 3D inconsistency of image inpainting based on diffusion models. We propose a generative model using image pairs that belong to the same scene. To achieve the 3D-consistent and semantically coherent inpainting, we modify the generative diffusion model by incorporating an alternative point of view of the scene into the denoising process. This creates an inductive bias that allows to recover 3D priors while training to denoise in 2D, without explicit 3D supervision. Training unconditional diffusion models with additional images as in-context guidance allows to harmonize the masked and non-masked regions while repainting and ensures the 3D consistency. We evaluate our method on one synthetic and three real-world datasets and show that it generates semantically coherent and 3D-consistent inpaintings and outperforms the state-of-art methods.
Authors: Donghang Lyu, Ruochen Gao, Marius Staring
Abstract: Medical image segmentation involves partitioning medical images into meaningful regions, with a focus on identifying anatomical structures or abnormalities. It has broad applications in healthcare, and deep learning methods have enabled significant advancements in automating this process. Recently, the introduction of the Segmentation Anything Model (SAM), the first foundation model for segmentation task, has prompted researchers to adapt it for the medical domain to improve performance across various tasks. However, SAM's large model size and high GPU requirements hinder its scalability and development in the medical domain. To address these challenges, research has increasingly focused on lightweight adaptations of SAM to reduce its parameter count, enabling training with limited GPU resources while maintaining competitive segmentation performance. In this work, we propose MCP-MedSAM, a powerful and lightweight medical SAM model designed to be trainable on a single GPU within one day while delivering superior segmentation performance. Our method was trained and evaluated using a large-scale challenge dataset\footnote{\url{https://www.codabench.org/competitions/1847}\label{comp}}, compared to top-ranking methods on the challenge leaderboard, MCP-MedSAM achieved superior performance while requiring only one day of training on a single GPU. The code is publicly available at \url{https://github.com/dong845/MCP-MedSAM}.
URLs: https://www.codabench.org/competitions/1847, https://github.com/dong845/MCP-MedSAM
Authors: Parthib Roy, Srinivasa Perisetla, Shashank Shriram, Harsha Krishnaswamy, Aryan Keskar, Ross Greer
Abstract: Human-interactive robotic systems, particularly autonomous vehicles (AVs), must effectively integrate human instructions into their motion planning. This paper introduces doScenes, a novel dataset designed to facilitate research on human-vehicle instruction interactions, focusing on short-term directives that directly influence vehicle motion. By annotating multimodal sensor data with natural language instructions and referentiality tags, doScenes bridges the gap between instruction and driving response, enabling context-aware and adaptive planning. Unlike existing datasets that focus on ranking or scene-level reasoning, doScenes emphasizes actionable directives tied to static and dynamic scene objects. This framework addresses limitations in prior research, such as reliance on simulated data or predefined action sets, by supporting nuanced and flexible responses in real-world scenarios. This work lays the foundation for developing learning strategies that seamlessly integrate human instructions into autonomous systems, advancing safe and effective human-vehicle collaboration for vision-language navigation. We make our data publicly available at https://www.github.com/rossgreer/doScenes
Authors: Jun Nie, Yonggang Zhang, Tongliang Liu, Yiu-ming Cheung, Bo Han, Xinmei Tian
Abstract: In this work, we propose a novel approach for detecting AI-generated images by leveraging predictive uncertainty to mitigate misuse and associated risks. The motivation arises from the fundamental assumption regarding the distributional discrepancy between natural and AI-generated images. The feasibility of distinguishing natural images from AI-generated ones is grounded in the distribution discrepancy between them. Predictive uncertainty offers an effective approach for capturing distribution shifts, thereby providing insights into detecting AI-generated images. Namely, as the distribution shift between training and testing data increases, model performance typically degrades, often accompanied by increased predictive uncertainty. Therefore, we propose to employ predictive uncertainty to reflect the discrepancies between AI-generated and natural images. In this context, the challenge lies in ensuring that the model has been trained over sufficient natural images to avoid the risk of determining the distribution of natural images as that of generated images. We propose to leverage large-scale pre-trained models to calculate the uncertainty as the score for detecting AI-generated images. This leads to a simple yet effective method for detecting AI-generated images using large-scale vision models: images that induce high uncertainty are identified as AI-generated. Comprehensive experiments across multiple benchmarks demonstrate the effectiveness of our method.
Authors: Yuanzhi Zhu, Hanshu Yan, Huan Yang, Kai Zhang, Junnan Li
Abstract: Generative models, particularly diffusion models, have made significant success in data synthesis across various modalities, including images, videos, and 3D assets. However, current diffusion models are computationally intensive, often requiring numerous sampling steps that limit their practical application, especially in video generation. This work introduces a novel framework for diffusion distillation and distribution matching that dramatically reduces the number of inference steps while maintaining-and potentially improving-generation quality. Our approach focuses on distilling pre-trained diffusion models into a more efficient few-step generator, specifically targeting video generation. By leveraging a combination of video GAN loss and a novel 2D score distribution matching loss, we demonstrate the potential to generate high-quality video frames with substantially fewer sampling steps. To be specific, the proposed method incorporates a denoising GAN discriminator to distil from the real data and a pre-trained image diffusion model to enhance the frame quality and the prompt-following capabilities. Experimental results using AnimateDiff as the teacher model showcase the method's effectiveness, achieving superior performance in just four sampling steps compared to existing techniques.
Authors: Jianing Zhang, Yuchao Zheng, Ziwei Li, Qionghai Dai, Xiaoyun Yuan
Abstract: Gaussian splatting has gained attention for its efficient representation and rendering of 3D scenes using continuous Gaussian primitives. However, it struggles with sparse-view inputs due to limited geometric and photometric information, causing ambiguities in depth, shape, and texture. we propose GBR: Generative Bundle Refinement, a method for high-fidelity Gaussian splatting and meshing using only 4-6 input views. GBR integrates a neural bundle adjustment module to enhance geometry accuracy and a generative depth refinement module to improve geometry fidelity. More specifically, the neural bundle adjustment module integrates a foundation network to produce initial 3D point maps and point matches from unposed images, followed by bundle adjustment optimization to improve multiview consistency and point cloud accuracy. The generative depth refinement module employs a diffusion-based strategy to enhance geometric details and fidelity while preserving the scale. Finally, for Gaussian splatting optimization, we propose a multimodal loss function incorporating depth and normal consistency, geometric regularization, and pseudo-view supervision, providing robust guidance under sparse-view conditions. Experiments on widely used datasets show that GBR significantly outperforms existing methods under sparse-view inputs. Additionally, GBR demonstrates the ability to reconstruct and render large-scale real-world scenes, such as the Pavilion of Prince Teng and the Great Wall, with remarkable details using only 6 views.
Authors: Xingyu Zheng, Xianglong Liu, Yichen Bian, Xudong Ma, Yulun Zhang, Jiakai Wang, Jinyang Guo, Haotong Qin
Abstract: Diffusion models (DMs) have been significantly developed and widely used in various applications due to their excellent generative qualities. However, the expensive computation and massive parameters of DMs hinder their practical use in resource-constrained scenarios. As one of the effective compression approaches, quantization allows DMs to achieve storage saving and inference acceleration by reducing bit-width while maintaining generation performance. However, as the most extreme quantization form, 1-bit binarization causes the generation performance of DMs to face severe degradation or even collapse. This paper proposes a novel method, namely BiDM, for fully binarizing weights and activations of DMs, pushing quantization to the 1-bit limit. From a temporal perspective, we introduce the Timestep-friendly Binary Structure (TBS), which uses learnable activation binarizers and cross-timestep feature connections to address the highly timestep-correlated activation features of DMs. From a spatial perspective, we propose Space Patched Distillation (SPD) to address the difficulty of matching binary features during distillation, focusing on the spatial locality of image generation tasks and noise estimation networks. As the first work to fully binarize DMs, the W1A1 BiDM on the LDM-4 model for LSUN-Bedrooms 256$\times$256 achieves a remarkable FID of 22.74, significantly outperforming the current state-of-the-art general binarization methods with an FID of 59.44 and invalid generative samples, and achieves up to excellent 28.0 times storage and 52.7 times OPs savings. The code is available at https://github.com/Xingyu-Zheng/BiDM .
Authors: Haoran Li, Yuli Tian, Yong Liao, Lin Wang, Yuyang Wang, Peng Yuan Zhou
Abstract: Distilling 3D representations from pretrained 2D diffusion models is essential for 3D creative applications across gaming, film, and interior design. Current SDS-based methods are hindered by inefficient information distillation from diffusion models, which prevents the creation of photorealistic 3D contents. Our research reevaluates the SDS approach by analyzing its fundamental nature as a basic image editing process that commonly results in over-saturation, over-smoothing and lack of rich content due to the poor-quality single-step denoising. To address these limitations, we propose GE3D (3D Generation by Editing). Each iteration of GE3D utilizes a 2D editing framework that combines a noising trajectory to preserve the information of the input image, alongside a text-guided denoising trajectory. We optimize the process by aligning the latents across both trajectories. This approach fully exploits pretrained diffusion models to distill multi-granularity information through multiple denoising steps, resulting in photorealistic 3D outputs. Both theoretical and experimental results confirm the effectiveness of our approach, which not only advances 3D generation technology but also establishes a novel connection between 3D generation and 2D editing. This could potentially inspire further research in the field. Code and demos are released at https://jahnsonblack.github.io/GE3D/.
Authors: Xiao Xu, Tianhao Niu, Yuxi Xie, Libo Qin, Wanxiang Che, Min-Yen Kan
Abstract: Multimodal Large Language Models (MLLMs) excel in vision--language tasks by pre-training solely on coarse-grained concept annotations (e.g., image captions). We hypothesize that integrating fine-grained concept annotations (e.g., object labels and object regions) will further improve performance, as both data granularities complement each other in terms of breadth and depth in concept representation. We introduce a new dataset featuring Multimodal Multi-Grained Concept annotations (MMGiC) for MLLMs. In constructing MMGiC, we explore the impact of different data recipes on multimodal comprehension and generation. Our analyses reveal that multi-grained concept annotations integrate and complement each other, under our structured template and a general MLLM framework. We clearly explore and demonstrate the potential of MMGiC to help MLLMs better locate and learn concepts, aligning vision and language at multiple granularities. We further validate our hypothesis by investigating the fair comparison and effective collaboration between MMGiC and image--caption data on 12 multimodal comprehension and generation benchmarks, e.g., their appropriate combination achieve 3.95% and 2.34% absolute improvements over image--caption data alone on POPE and SEED-Bench. Code, data and models will be available at https://github.com/LooperXX/MMGiC.
Authors: Jie Ning, Jiebao Sun, Shengzhu Shi, Zhichang Guo, Yao Li, Hongwei Li, Boying Wu
Abstract: Deep learning-based image denoising models demonstrate remarkable performance, but their lack of robustness analysis remains a significant concern. A major issue is that these models are susceptible to adversarial attacks, where small, carefully crafted perturbations to input data can cause them to fail. Surprisingly, perturbations specifically crafted for one model can easily transfer across various models, including CNNs, Transformers, unfolding models, and plug-and-play models, leading to failures in those models as well. Such high adversarial transferability is not observed in classification models. We analyze the possible underlying reasons behind the high adversarial transferability through a series of hypotheses and validation experiments. By characterizing the manifolds of Gaussian noise and adversarial perturbations using the concept of typical set and the asymptotic equipartition property, we prove that adversarial samples deviate slightly from the typical set of the original input distribution, causing the models to fail. Based on these insights, we propose a novel adversarial defense method: the Out-of-Distribution Typical Set Sampling Training strategy (TS). TS not only significantly enhances the model's robustness but also marginally improves denoising performance compared to the original model.
Authors: Qiao Feng, Yebin Liu, Yu-Kun Lai, Jingyu Yang, Kun Li
Abstract: We introduce FOF-X for real-time reconstruction of detailed human geometry from a single image. Balancing real-time speed against high-quality results is a persistent challenge, mainly due to the high computational demands of existing 3D representations. To address this, we propose Fourier Occupancy Field (FOF), an efficient 3D representation by learning the Fourier series. The core of FOF is to factorize a 3D occupancy field into a 2D vector field, retaining topology and spatial relationships within the 3D domain while facilitating compatibility with 2D convolutional neural networks. Such a representation bridges the gap between 3D and 2D domains, enabling the integration of human parametric models as priors and enhancing the reconstruction robustness. Based on FOF, we design a new reconstruction framework, FOF-X, to avoid the performance degradation caused by texture and lighting. This enables our real-time reconstruction system to better handle the domain gap between training images and real images. Additionally, in FOF-X, we enhance the inter-conversion algorithms between FOF and mesh representations with a Laplacian constraint and an automaton-based discontinuity matcher, improving both quality and robustness. We validate the strengths of our approach on different datasets and real-captured data, where FOF-X achieves new state-of-the-art results. The code will be released for research purposes.
Authors: Zipeng Qi, Hao Chen, Haotian Zhang, Zhengxia Zou, Zhenwei Shi
Abstract: In this paper, we propose a novel semantic splatting approach based on Gaussian Splatting to achieve efficient and low-latency. Our method projects the RGB attributes and semantic features of point clouds onto the image plane, simultaneously rendering RGB images and semantic segmentation results. Leveraging the explicit structure of point clouds and a one-time rendering strategy, our approach significantly enhances efficiency during optimization and rendering. Additionally, we employ SAM2 to generate pseudo-labels for boundary regions, which often lack sufficient supervision, and introduce two-level aggregation losses at the 2D feature map and 3D spatial levels to improve the view-consistent and spatial continuity.
Authors: Jinqing Zhang, Yanan Zhang, Qingjie Liu, Yunhong Wang
Abstract: Occupancy prediction has garnered increasing attention in recent years for its comprehensive fine-grained environmental representation and strong generalization to open-set objects. However, cumbersome voxel features and 3D convolution operations inevitably introduce large overheads in both memory and computation, obstructing the deployment of occupancy prediction approaches in real-time autonomous driving systems. Although some methods attempt to efficiently predict 3D occupancy from 2D Bird's-Eye-View (BEV) features through the Channel-to-Height mechanism, BEV features are insufficient to store all the height information of the scene, which limits performance. This paper proposes LightOcc, an innovative 3D occupancy prediction framework that leverages Lightweight Spatial Embedding to effectively supplement the height clues for the BEV-based representation while maintaining its deployability. Firstly, Global Spatial Sampling is used to obtain the Single-Channel Occupancy from multi-view depth distribution. Spatial-to-Channel mechanism then takes the arbitrary spatial dimension of Single-Channel Occupancy as the feature dimension and extracts Tri-Perspective Views (TPV) Embeddings by 2D convolution. Finally, TPV Embeddings will interact with each other by Lightweight TPV Interaction module to obtain the Spatial Embedding that is optimal supplementary to BEV features. Sufficient experimental results show that LightOcc significantly increases the prediction accuracy of the baseline and achieves state-of-the-art performance on the Occ3D-nuScenes benchmark.
Authors: Yiren Song, Shengtao Lou, Xiaokang Liu, Hai Ci, Pei Yang, Jiaming Liu, Mike Zheng Shou
Abstract: Diffusion models have revolutionized generative modeling with their exceptional ability to produce high-fidelity images. However, misuse of such potent tools can lead to the creation of fake news or disturbing content targeting individuals, resulting in significant social harm. In this paper, we introduce Anti-Reference, a novel method that protects images from the threats posed by reference-based generation techniques by adding imperceptible adversarial noise to the images. We propose a unified loss function that enables joint attacks on fine-tuning-based customization methods, non-fine-tuning customization methods, and human-centric driving methods. Based on this loss, we train a Adversarial Noise Encoder to predict the noise or directly optimize the noise using the PGD method. Our method shows certain transfer attack capabilities, effectively challenging both gray-box models and some commercial APIs. Extensive experiments validate the performance of Anti-Reference, establishing a new benchmark in image security.
Authors: Tianshuo Peng, Mingsheng Li, Hongbin Zhou, Renqiu Xia, Renrui Zhang, Lei Bai, Song Mao, Bin Wang, Conghui He, Aojun Zhou, Botian Shi, Tao Chen, Bo Zhang, Xiangyu Yue
Abstract: Recent advancements in Large Multi-modal Models (LMMs) underscore the importance of scaling by increasing image-text paired data, achieving impressive performance on general tasks. Despite their effectiveness in broad applications, generalist models are primarily trained on web-scale datasets dominated by natural images, resulting in the sacrifice of specialized capabilities for domain-specific tasks that require extensive domain prior knowledge. Moreover, directly integrating expert models tailored for specific domains is challenging due to the representational gap and imbalanced optimization between the generalist model and experts. To address these challenges, we introduce Chimera, a scalable and low-cost multi-modal pipeline designed to boost the ability of existing LMMs with domain-specific experts. Specifically, we design a progressive training strategy to integrate features from expert models into the input of a generalist LMM. To address the imbalanced optimization caused by the well-aligned general visual encoder, we introduce a novel Generalist-Specialist Collaboration Masking (GSCM) mechanism. This results in a versatile model that excels across the chart, table, math, and document domains, achieving state-of-the-art performance on multi-modal reasoning and visual content extraction tasks, both of which are challenging tasks for assessing existing LMMs.
Authors: Xiao Zhang, Ruoxi Jiang, Rebecca Willett, Michael Maire
Abstract: We introduce nested diffusion models, an efficient and powerful hierarchical generative framework that substantially enhances the generation quality of diffusion models, particularly for images of complex scenes. Our approach employs a series of diffusion models to progressively generate latent variables at different semantic levels. Each model in this series is conditioned on the output of the preceding higher-level models, culminating in image generation. Hierarchical latent variables guide the generation process along predefined semantic pathways, allowing our approach to capture intricate structural details while significantly improving image quality. To construct these latent variables, we leverage a pre-trained visual encoder, which learns strong semantic visual representations, and modulate its capacity via dimensionality reduction and noise injection. Across multiple datasets, our system demonstrates significant enhancements in image quality for both unconditional and class/text conditional generation. Moreover, our unconditional generation system substantially outperforms the baseline conditional system. These advancements incur minimal computational overhead as the more abstract levels of our hierarchy work with lower-dimensional representations.
Authors: Bimarsha Khanal, Paras Poudel, Anish Chapagai, Bijan Regmi, Sitaram Pokhrel, Salik Ram Khanal
Abstract: Plant diseases significantly impact our food supply, causing problems for farmers, economies reliant on agriculture, and global food security. Accurate and timely plant disease diagnosis is crucial for effective treatment and minimizing yield losses. Despite advancements in agricultural technology, a precise and early diagnosis remains a challenge, especially in underdeveloped regions where agriculture is crucial and agricultural experts are scarce. However, adopting Deep Learning applications can assist in accurately identifying diseases without needing plant pathologists. In this study, the effectiveness of various computer vision models for detecting paddy diseases is evaluated and proposed the best deep learning-based disease detection system. Both classification and detection using the Paddy Doctor dataset, which contains over 20,000 annotated images of paddy leaves for disease diagnosis are tested and evaluated. For detection, we utilized the YOLOv8 model-based model were used for paddy disease detection and CNN models and the Vision Transformer were used for disease classification. The average mAP50 of 69% for detection tasks was achieved and the Vision Transformer classification accuracy was 99.38%. It was found that detection models are effective at identifying multiple diseases simultaneously with less computing power, whereas classification models, though computationally expensive, exhibit better performance for classifying single diseases. Additionally, a mobile application was developed to enable farmers to identify paddy diseases instantly. Experiments with the app showed encouraging results in utilizing the trained models for both disease classification and treatment guidance.
Authors: Aymen Sekhri, Seyed Ali Amirshahi, Mohamed-Chaker Larabi
Abstract: Augmented Reality (AR) is a major immersive media technology that enriches our perception of reality by overlaying digital content (the foreground) onto physical environments (the background). It has far-reaching applications, from entertainment and gaming to education, healthcare, and industrial training. Nevertheless, challenges such as visual confusion and classical distortions can result in user discomfort when using the technology. Evaluating AR quality of experience becomes essential to measure user satisfaction and engagement, facilitating the refinement necessary for creating immersive and robust experiences. Though, the scarcity of data and the distinctive characteristics of AR technology render the development of effective quality assessment metrics challenging. This paper presents a deep learning-based objective metric designed specifically for assessing image quality for AR scenarios. The approach entails four key steps, (1) fine-tuning a self-supervised pre-trained vision transformer to extract prominent features from reference images and distilling this knowledge to improve representations of distorted images, (2) quantifying distortions by computing shift representations, (3) employing cross-attention-based decoders to capture perceptual quality features, and (4) integrating regularization techniques and label smoothing to address the overfitting problem. To validate the proposed approach, we conduct extensive experiments on the ARIQA dataset. The results showcase the superior performance of our proposed approach across all model variants, namely TransformAR, TransformAR-KD, and TransformAR-KD+ in comparison to existing state-of-the-art methods.
Authors: Anton Baumann, Rui Li, Marcus Klasson, Santeri Mentu, Shyamgopal Karthik, Zeynep Akata, Arno Solin, Martin Trapp
Abstract: Vision-language models (VLMs), such as CLIP and SigLIP, have found remarkable success in classification, retrieval, and generative tasks. For this, VLMs deterministically map images and text descriptions to a joint latent space in which their similarity is assessed using the cosine similarity. However, a deterministic mapping of inputs fails to capture uncertainties over concepts arising from domain shifts when used in downstream tasks. In this work, we propose post-hoc uncertainty estimation in VLMs that does not require additional training. Our method leverages a Bayesian posterior approximation over the last layers in VLMs and analytically quantifies uncertainties over cosine similarities. We demonstrate its effectiveness for uncertainty quantification and support set selection in active learning. Compared to baselines, we obtain improved and well-calibrated predictive uncertainties, interpretable uncertainty estimates, and sample-efficient active learning. Our results show promise for safety-critical applications of large-scale models.
Authors: Hyeonho Jeong, Chun-Hao Paul Huang, Jong Chul Ye, Niloy Mitra, Duygu Ceylan
Abstract: While recent foundational video generators produce visually rich output, they still struggle with appearance drift, where objects gradually degrade or change inconsistently across frames, breaking visual coherence. We hypothesize that this is because there is no explicit supervision in terms of spatial tracking at the feature level. We propose Track4Gen, a spatially aware video generator that combines video diffusion loss with point tracking across frames, providing enhanced spatial supervision on the diffusion features. Track4Gen merges the video generation and point tracking tasks into a single network by making minimal changes to existing video generation architectures. Using Stable Video Diffusion as a backbone, Track4Gen demonstrates that it is possible to unify video generation and point tracking, which are typically handled as separate tasks. Our extensive evaluations show that Track4Gen effectively reduces appearance drift, resulting in temporally stable and visually coherent video generation. Project page: hyeonho99.github.io/Track4Gen
Authors: Shuning Chang, Pichao Wang, Jiasheng Tang, Yi Yang
Abstract: Diffusion Transformers (DiT) deliver impressive generative performance but face prohibitive computational demands due to both the quadratic complexity of token-based self-attention and the need for extensive sampling steps. While recent research has focused on accelerating sampling, the structural inefficiencies of DiT remain underexplored. We propose FlexDiT, a framework that dynamically adapts token density across both spatial and temporal dimensions to achieve computational efficiency without compromising generation quality. Spatially, FlexDiT employs a three-segment architecture that allocates token density based on feature requirements at each layer: Poolingformer in the bottom layers for efficient global feature extraction, Sparse-Dense Token Modules (SDTM) in the middle layers to balance global context with local detail, and dense tokens in the top layers to refine high-frequency details. Temporally, FlexDiT dynamically modulates token density across denoising stages, progressively increasing token count as finer details emerge in later timesteps. This synergy between FlexDiT's spatially adaptive architecture and its temporal pruning strategy enables a unified framework that balances efficiency and fidelity throughout the generation process. Our experiments demonstrate FlexDiT's effectiveness, achieving a 55% reduction in FLOPs and a 175% improvement in inference speed on DiT-XL with only a 0.09 increase in FID score on 512$\times$512 ImageNet images, a 56% reduction in FLOPs across video generation datasets including FaceForensics, SkyTimelapse, UCF101, and Taichi-HD, and a 69% improvement in inference speed on PixArt-$\alpha$ on text-to-image generation task with a 0.24 FID score decrease. FlexDiT provides a scalable solution for high-quality diffusion-based generation compatible with further sampling optimization techniques.
Authors: Zhenghong Zhou, Jie An, Jiebo Luo
Abstract: Precise camera pose control is crucial for video generation with diffusion models. Existing methods require fine-tuning with additional datasets containing paired videos and camera pose annotations, which are both data-intensive and computationally costly, and can disrupt the pre-trained model distribution. We introduce Latent-Reframe, which enables camera control in a pre-trained video diffusion model without fine-tuning. Unlike existing methods, Latent-Reframe operates during the sampling stage, maintaining efficiency while preserving the original model distribution. Our approach reframes the latent code of video frames to align with the input camera trajectory through time-aware point clouds. Latent code inpainting and harmonization then refine the model latent space, ensuring high-quality video generation. Experimental results demonstrate that Latent-Reframe achieves comparable or superior camera control precision and video quality to training-based methods, without the need for fine-tuning on additional datasets.
Authors: Yuemei Ren, Liang Liao, Stephen John Maybank, Yanning Zhang, Xin Liu
Abstract: This paper addresses the challenge of spectral-spatial feature extraction for hyperspectral image classification by introducing a novel tensor-based framework. The proposed approach incorporates circular convolution into a tensor structure to effectively capture and integrate both spectral and spatial information. Building upon this framework, the traditional Principal Component Analysis (PCA) technique is extended to its tensor-based counterpart, referred to as Tensor Principal Component Analysis (TPCA). The proposed TPCA method leverages the inherent multi-dimensional structure of hyperspectral data, thereby enabling more effective feature representation. Experimental results on benchmark hyperspectral datasets demonstrate that classification models using TPCA features consistently outperform those using traditional PCA and other state-of-the-art techniques. These findings highlight the potential of the tensor-based framework in advancing hyperspectral image analysis.
Authors: Karlo Koledic, Luka Petrovic, Ivan Markovic, Ivan Petrovic
Abstract: Generalizing metric monocular depth estimation presents a significant challenge due to its ill-posed nature, while the entanglement between camera parameters and depth amplifies issues further, hindering multi-dataset training and zero-shot accuracy. This challenge is particularly evident in autonomous vehicles and mobile robotics, where data is collected with fixed camera setups, limiting the geometric diversity. Yet, this context also presents an opportunity: the fixed relationship between the camera and the ground plane imposes additional perspective geometry constraints, enabling depth regression via vertical image positions of objects. However, this cue is highly susceptible to overfitting, thus we propose a novel canonical representation that maintains consistency across varied camera setups, effectively disentangling depth from specific parameters and enhancing generalization across datasets. We also propose a novel architecture that adaptively and probabilistically fuses depths estimated via object size and vertical image position cues. A comprehensive evaluation demonstrates the effectiveness of the proposed approach on five autonomous driving datasets, achieving accurate metric depth estimation for varying resolutions, aspect ratios and camera setups. Notably, we achieve comparable accuracy to existing zero-shot methods, despite training on a single dataset with a single-camera setup.
Authors: Leo Fillioux, Julio Silva-Rodr\'iguez, Ismail Ben Ayed, Paul-Henry Courn\`ede, Maria Vakalopoulou, Stergios Christodoulidis, Jose Dolz
Abstract: Recent advances in self-supervision and constrastive learning have brought the performance of foundation models to unprecedented levels in a variety of tasks. Fueled by this progress, these models are becoming the prevailing approach for a wide array of real-world vision problems, including risk-sensitive and high-stakes applications. However, ensuring safe deployment in these scenarios requires a more comprehensive understanding of their uncertainty modeling capabilities, which has been barely explored. In this work, we delve into the behavior of vision and vision-language foundation models under Conformal Prediction (CP), a statistical framework that provides theoretical guarantees of marginal coverage of the true class. Across extensive experiments including popular vision classification benchmarks, well-known foundation vision models, and three CP methods, our findings reveal that foundation models are well-suited for conformalization procedures, particularly those integrating Vision Transformers. Furthermore, we show that calibrating the confidence predictions of these models leads to efficiency degradation of the conformal set on adaptive CP methods. In contrast, few-shot adaptation to downstream tasks generally enhances conformal scores, where we identify Adapters as a better conformable alternative compared to Prompt Learning strategies. Our empirical study identifies APS as particularly promising in the context of vision foundation models, as it does not violate the marginal coverage property across multiple challenging, yet realistic scenarios.
Authors: Ruoxin Wang, Tianyi Tang, Haiming Du, Yuxuan Cheng, Yu Wang, Lingjie Yang, Xiaohui Duan, Yunfang Yu, Yu Zhou, Donglong Chen
Abstract: Brain tumor segmentation models have aided diagnosis in recent years. However, they face MRI complexity and variability challenges, including irregular shapes and unclear boundaries, leading to noise, misclassification, and incomplete segmentation, thereby limiting accuracy. To address these issues, we adhere to an outstanding Convolutional Neural Networks (CNNs) design paradigm and propose a novel network named A4-Unet. In A4-Unet, Deformable Large Kernel Attention (DLKA) is incorporated in the encoder, allowing for improved capture of multi-scale tumors. Swin Spatial Pyramid Pooling (SSPP) with cross-channel attention is employed in a bottleneck further to study long-distance dependencies within images and channel relationships. To enhance accuracy, a Combined Attention Module (CAM) with Discrete Cosine Transform (DCT) orthogonality for channel weighting and convolutional element-wise multiplication is introduced for spatial weighting in the decoder. Attention gates (AG) are added in the skip connection to highlight the foreground while suppressing irrelevant background information. The proposed network is evaluated on three authoritative MRI brain tumor benchmarks and a proprietary dataset, and it achieves a 94.4% Dice score on the BraTS 2020 dataset, thereby establishing multiple new state-of-the-art benchmarks. The code is available here: https://github.com/WendyWAAAAANG/A4-Unet.
Authors: Ashish Goswami, Satyam Kumar Modi, Santhosh Rishi Deshineni, Harman Singh, Prathosh A. P, Parag Singla
Abstract: Text-to-image (T2I) generation has seen significant progress with diffusion models, enabling generation of photo-realistic images from text prompts. Despite this progress, existing methods still face challenges in following complex text prompts, especially those requiring compositional and multi-step reasoning. Given such complex instructions, SOTA models often make mistakes in faithfully modeling object attributes, and relationships among them. In this work, we present an alternate paradigm for T2I synthesis, decomposing the task of complex multi-step generation into three steps, (a) Generate: we first generate an image using existing diffusion models (b) Plan: we make use of Multi-Modal LLMs (MLLMs) to identify the mistakes in the generated image expressed in terms of individual objects and their properties, and produce a sequence of corrective steps required in the form of an edit-plan. (c) Edit: we make use of an existing text-guided image editing models to sequentially execute our edit-plan over the generated image to get the desired image which is faithful to the original instruction. Our approach derives its strength from the fact that it is modular in nature, is training free, and can be applied over any combination of image generation and editing models. As an added contribution, we also develop a model capable of compositional editing, which further helps improve the overall accuracy of our proposed approach. Our method flexibly trades inference time compute with performance on compositional text prompts. We perform extensive experimental evaluation across 3 benchmarks and 10 T2I models including DALLE-3 and the latest -- SD-3.5-Large. Our approach not only improves the performance of the SOTA models, by upto 3 points, it also reduces the performance gap between weaker and stronger models. $\href{https://dair-iitd.github.io/GraPE/}{https://dair-iitd.github.io/GraPE/}$
URLs: https://dair-iitd.github.io/GraPE/, https://dair-iitd.github.io/GraPE/
Authors: Eric Dolores-Cuenca, Aldo Guzman-Saenz, Sangil Kim, Susana Lopez-Moreno, Jose Mendoza-Cortes
Abstract: The paper ``Tropical Geometry of Deep Neural Networks'' by L. Zhang et al. introduces an equivalence between integer-valued neural networks (IVNN) with activation $\text{ReLU}_{t}$ and tropical rational functions, which come with a map to polytopes. Here, IVNN refers to a network with integer weights but real biases, and $\text{ReLU}_{t}$ is defined as $\text{ReLU}_{t}(x)=\max(x,t)$ for $t\in\mathbb{R}\cup\{-\infty\}$. For every poset with $n$ points, there exists a corresponding order polytope, i.e., a convex polytope in the unit cube $[0,1]^n$ whose coordinates obey the inequalities of the poset. We study neural networks whose associated polytope is an order polytope. We then explain how posets with four points induce neural networks that can be interpreted as $2\times 2$ convolutional filters. These poset filters can be added to any neural network, not only IVNN. Similarly to maxout, poset convolutional filters update the weights of the neural network during backpropagation with more precision than average pooling, max pooling, or mixed pooling, without the need to train extra parameters. We report experiments that support our statements. We also prove that the assignment from a poset to an order polytope (and to certain tropical polynomials) is one to one, and we define the structure of algebra over the operad of posets on tropical polynomials.
Authors: Calvin Glisson, Qiuxiao Chen
Abstract: Autonomous driving has garnered significant attention in recent research, and Bird's-Eye-View (BEV) map segmentation plays a vital role in the field, providing the basis for safe and reliable operation. While data augmentation is a commonly used technique for improving BEV map segmentation networks, existing approaches predominantly focus on manipulating spatial domain representations. In this work, we investigate the potential of frequency domain data augmentation for camera-based BEV map segmentation. We observe that high-frequency information in camera images is particularly crucial for accurate segmentation. Based on this insight, we propose High-frequency Shuffle Data Augmentation (HSDA), a novel data augmentation strategy that enhances a network's ability to interpret high-frequency image content. This approach encourages the network to distinguish relevant high-frequency information from noise, leading to improved segmentation results for small and intricate image regions, as well as sharper edge and detail perception. Evaluated on the nuScenes dataset, our method demonstrates broad applicability across various BEV map segmentation networks, achieving a new state-of-the-art mean Intersection over Union (mIoU) of 61.3% for camera-only systems. This significant improvement underscores the potential of frequency domain data augmentation for advancing the field of autonomous driving perception. Code has been released: https://github.com/Zarhult/HSDA
Authors: Lei Su, Yang Du
Abstract: We focus on tertiary lymphoid structure (TLS) semantic segmentation in whole slide image (WSI). Unlike TLS binary segmentation, TLS semantic segmentation identifies boundaries and maturity, which requires integrating contextual information to discover discriminative features. Due to the extensive scale of WSI (e.g., 100,000 \times 100,000 pixels), the segmentation of TLS is usually carried out through a patch-based strategy. However, this prevents the model from accessing information outside of the patches, limiting the performance. To address this issue, we propose GCUNet, a GNN-based contextual learning network for TLS semantic segmentation. Given an image patch (target) to be segmented, GCUNet first progressively aggregates long-range and fine-grained context outside the target. Then, a Detail and Context Fusion block (DCFusion) is designed to integrate the context and detail of the target to predict the segmentation mask. We build four TLS semantic segmentation datasets, called TCGA-COAD, TCGA-LUSC, TCGA-BLCA and INHOUSE-PAAD, and make the former three datasets (comprising 826 WSIs and 15,276 TLSs) publicly available to promote the TLS semantic segmentation. Experiments on these datasets demonstrate the superiority of GCUNet, achieving at least 7.41% improvement in mF1 compared with SOTA.
Authors: Qiyu Liao, Xin Yuan, Min Xu, Dadong Wang
Abstract: In Fine-Grained Visual Classification (FGVC), distinguishing highly similar subcategories remains a formidable challenge, often necessitating datasets with extensive variability. The acquisition and annotation of such FGVC datasets are notably difficult and costly, demanding specialized knowledge to identify subtle distinctions among closely related categories. Our study introduces a novel approach employing the Sequence Latent Diffusion Model (SLDM) for augmenting FGVC datasets, called Sequence Generative Image Augmentation (SGIA). Our method features a unique Bridging Transfer Learning (BTL) process, designed to minimize the domain gap between real and synthetically augmented data. This approach notably surpasses existing methods in generating more realistic image samples, providing a diverse range of pose transformations that extend beyond the traditional rigid transformations and style changes in generative augmentation. We demonstrate the effectiveness of our augmented dataset with substantial improvements in FGVC tasks on various datasets, models, and training strategies, especially in few-shot learning scenarios. Our method outperforms conventional image augmentation techniques in benchmark tests on three FGVC datasets, showcasing superior realism, variability, and representational quality. Our work sets a new benchmark and outperforms the previous state-of-the-art models in classification accuracy by 0.5% for the CUB-200-2011 dataset and advances the application of generative models in FGVC data augmentation.
Authors: Kangyu Zhu, Peng Xia, Yun Li, Hongtu Zhu, Sheng Wang, Huaxiu Yao
Abstract: The advancement of Large Vision-Language Models (LVLMs) has propelled their application in the medical field. However, Medical LVLMs (Med-LVLMs) encounter factuality challenges due to modality misalignment, where the models prioritize textual knowledge over visual input, leading to hallucinations that contradict information in medical images. Previous attempts to enhance modality alignment in Med-LVLMs through preference optimization have inadequately mitigated clinical relevance in preference data, making these samples easily distinguishable and reducing alignment effectiveness. To address this challenge, we propose MMedPO, a novel multimodal medical preference optimization approach that considers the clinical relevance of preference samples to enhance Med-LVLM alignment. MMedPO curates multimodal preference data by introducing two types of dispreference: (1) plausible hallucinations injected through target Med-LVLMs or GPT-4o to produce medically inaccurate responses, and (2) lesion region neglect achieved through local lesion-noising, disrupting visual understanding of critical areas. We then calculate clinical relevance for each sample based on scores from multiple Med-LLMs and visual tools, and integrate these scores into the preference optimization process as weights, enabling effective alignment. Our experiments demonstrate that MMedPO significantly enhances factual accuracy in Med-LVLMs, achieving substantial improvements over existing preference optimization methods by averaging 14.2% and 51.7% across the Med-VQA and report generation tasks. Our code are available in https://github.com/aiming-lab/MMedPO.
Authors: Zonglin Meng, Yun Zhang, Zhaoliang Zheng, Zhihao Zhao, Jiaqi Ma
Abstract: Cooperative perception has attracted wide attention given its capability to leverage shared information across connected automated vehicles (CAVs) and smart infrastructures to address sensing occlusion and range limitation issues. However, existing research overlooks the fragile multi-sensor correlations in multi-agent settings, as the heterogeneous agent sensor measurements are highly susceptible to environmental factors, leading to weakened inter-agent sensor interactions. The varying operational conditions and other real-world factors inevitably introduce multifactorial noise and consequentially lead to multi-sensor misalignment, making the deployment of multi-agent multi-modality perception particularly challenging in the real world. In this paper, we propose AgentAlign, a real-world heterogeneous agent cross-modality feature alignment framework, to effectively address these multi-modality misalignment issues. Our method introduces a cross-modality feature alignment space (CFAS) and heterogeneous agent feature alignment (HAFA) mechanism to harmonize multi-modality features across various agents dynamically. Additionally, we present a novel V2XSet-noise dataset that simulates realistic sensor imperfections under diverse environmental conditions, facilitating a systematic evaluation of our approach's robustness. Extensive experiments on the V2X-Real and V2XSet-Noise benchmarks demonstrate that our framework achieves state-of-the-art performance, underscoring its potential for real-world applications in cooperative autonomous driving. The controllable V2XSet-Noise dataset and generation pipeline will be released in the future.
Authors: Yuan Wang, Ouxiang Li, Tingting Mu, Yanbin Hao, Kuien Liu, Xiang Wang, Xiangnan He
Abstract: The success of text-to-image generation enabled by diffuion models has imposed an urgent need to erase unwanted concepts, e.g., copyrighted, offensive, and unsafe ones, from the pre-trained models in a precise, timely, and low-cost manner. The twofold demand of concept erasure requires a precise removal of the target concept during generation (i.e., erasure efficacy), while a minimal impact on non-target content generation (i.e., prior preservation). Existing methods are either computationally costly or face challenges in maintaining an effective balance between erasure efficacy and prior preservation. To improve, we propose a precise, fast, and low-cost concept erasure method, called Adaptive Vaule Decomposer (AdaVD), which is training-free. This method is grounded in a classical linear algebraic orthogonal complement operation, implemented in the value space of each cross-attention layer within the UNet of diffusion models. An effective shift factor is designed to adaptively navigate the erasure strength, enhancing prior preservation without sacrificing erasure efficacy. Extensive experimental results show that the proposed AdaVD is effective at both single and multiple concept erasure, showing a 2- to 10-fold improvement in prior preservation as compared to the second best, meanwhile achieving the best or near best erasure efficacy, when comparing with both training-based and training-free state of the arts. AdaVD supports a series of diffusion models and downstream image generation tasks, the code is available on the project page: https://github.com/WYuan1001/AdaVD
Authors: Xinpeng Liu, Junxuan Liang, Chenshuo Zhang, Zixuan Cai, Cewu Lu, Yong-Lu Li
Abstract: Analyses of human motion kinematics have achieved tremendous advances. However, the production mechanism, known as human dynamics, is still undercovered. In this paper, we aim to push data-driven human dynamics understanding forward. We identify a major obstacle to this as the heterogeneity of existing human motion understanding efforts. Specifically, heterogeneity exists in not only the diverse kinematics representations and hierarchical dynamics representations but also in the data from different domains, namely biomechanics and reinforcement learning. With an in-depth analysis of the existing heterogeneity, we propose to emphasize the beneath homogeneity: all of them represent the homogeneous fact of human motion, though from different perspectives. Given this, we propose Homogeneous Dynamics Space (HDyS) as a fundamental space for human dynamics by aggregating heterogeneous data and training a homogeneous latent space with inspiration from the inverse-forward dynamics procedure. Leveraging the heterogeneous representations and datasets, HDyS achieves decent mapping between human kinematics and dynamics. We demonstrate the feasibility of HDyS with extensive experiments and applications. The project page is https://foruck.github.io/HDyS.
Authors: Xueluan Gong, Bowei Tian, Meng Xue, Yuan Wu, Yanjiao Chen, Qian Wang
Abstract: Recent studies have revealed the vulnerability of Deep Neural Network (DNN) models to backdoor attacks. However, existing backdoor attacks arbitrarily set the trigger mask or use a randomly selected trigger, which restricts the effectiveness and robustness of the generated backdoor triggers. In this paper, we propose a novel attention-based mask generation methodology that searches for the optimal trigger shape and location. We also introduce a Quality-of-Experience (QoE) term into the loss function and carefully adjust the transparency value of the trigger in order to make the backdoored samples to be more natural. To further improve the prediction accuracy of the victim model, we propose an alternating retraining algorithm in the backdoor injection process. The victim model is retrained with mixed poisoned datasets in even iterations and with only benign samples in odd iterations. Besides, we launch the backdoor attack under a co-optimized attack framework that alternately optimizes the backdoor trigger and backdoored model to further improve the attack performance. Apart from DNN models, we also extend our proposed attack method against vision transformers. We evaluate our proposed method with extensive experiments on VGG-Flower, CIFAR-10, GTSRB, CIFAR-100, and ImageNette datasets. It is shown that we can increase the attack success rate by as much as 82\% over baselines when the poison ratio is low and achieve a high QoE of the backdoored samples. Our proposed backdoor attack framework also showcases robustness against state-of-the-art backdoor defenses.
Authors: Connor Malone, Somayeh Hussaini, Tobias Fischer, Michael Milford
Abstract: Visual Place Recognition (VPR) enables coarse localization by comparing query images to a reference database of geo-tagged images. Recent breakthroughs in deep learning architectures and training regimes have led to methods with improved robustness to factors like environment appearance change, but with the downside that the required training and/or matching compute scales with the number of distinct environmental conditions encountered. Here, we propose Hyperdimensional One Place Signatures (HOPS) to simultaneously improve the performance, compute and scalability of these state-of-the-art approaches by fusing the descriptors from multiple reference sets captured under different conditions. HOPS scales to any number of environmental conditions by leveraging the Hyperdimensional Computing framework. Extensive evaluations demonstrate that our approach is highly generalizable and consistently improves recall performance across all evaluated VPR methods and datasets by large margins. Arbitrarily fusing reference images without compute penalty enables numerous other useful possibilities, three of which we demonstrate here: descriptor dimensionality reduction with no performance penalty, stacking synthetic images, and coarse localization to an entire traverse or environmental section.
Authors: Yuming Li, Peidong Jia, Daiwei Hong, Yueru Jia, Qi She, Rui Zhao, Ming Lu, Shanghang Zhang
Abstract: Training-free high-resolution (HR) image generation has garnered significant attention due to the high costs of training large diffusion models. Most existing methods begin by reconstructing the overall structure and then proceed to refine the local details. Despite their advancements, they still face issues with repetitive patterns in HR image generation. Besides, HR generation with diffusion models incurs significant computational costs. Thus, parallel generation is essential for interactive applications. To solve the above limitations, we introduce a novel method named ASGDiffusion for parallel HR generation with Asynchronous Structure Guidance (ASG) using pre-trained diffusion models. To solve the pattern repetition problem of HR image generation, ASGDiffusion leverages the low-resolution (LR) noise weighted by the attention mask as the structure guidance for the denoising step to ensure semantic consistency. The proposed structure guidance can significantly alleviate the pattern repetition problem. To enable parallel generation, we further propose a parallelism strategy, which calculates the patch noises and structure guidance asynchronously. By leveraging multi-GPU parallel acceleration, we significantly accelerate generation speed and reduce memory usage per GPU. Extensive experiments demonstrate that our method effectively and efficiently addresses common issues like pattern repetition and achieves state-of-the-art HR generation.
Authors: Huaxin Zhang, Xiaohao Xu, Xiang Wang, Jialong Zuo, Xiaonan Huang, Changxin Gao, Shanjun Zhang, Li Yu, Nong Sang
Abstract: How can we enable models to comprehend video anomalies occurring over varying temporal scales and contexts? Traditional Video Anomaly Understanding (VAU) methods focus on frame-level anomaly prediction, often missing the interpretability of complex and diverse real-world anomalies. Recent multimodal approaches leverage visual and textual data but lack hierarchical annotations that capture both short-term and long-term anomalies. To address this challenge, we introduce HIVAU-70k, a large-scale benchmark for hierarchical video anomaly understanding across any granularity. We develop a semi-automated annotation engine that efficiently scales high-quality annotations by combining manual video segmentation with recursive free-text annotation using large language models (LLMs). This results in over 70,000 multi-granular annotations organized at clip-level, event-level, and video-level segments. For efficient anomaly detection in long videos, we propose the Anomaly-focused Temporal Sampler (ATS). ATS integrates an anomaly scorer with a density-aware sampler to adaptively select frames based on anomaly scores, ensuring that the multimodal LLM concentrates on anomaly-rich regions, which significantly enhances both efficiency and accuracy. Extensive experiments demonstrate that our hierarchical instruction data markedly improves anomaly comprehension. The integrated ATS and visual-language model outperform traditional methods in processing long videos. Our benchmark and model are publicly available at https://github.com/pipixin321/HolmesVAU.
Authors: Fan Liu, Chenwei Dong, Chuanyi Zhang, Hualiang Zhou, Jun Zhou
Abstract: Many researchers collect data from the internet through crowd-sourcing or web crawling to alleviate the data-hungry challenge associated with cross-modal matching. Although such practice does not require expensive annotations, it inevitably introduces mismatched pairs and results in a noisy correspondence problem. Current approaches leverage the memorization effect of deep neural networks to distinguish noise and perform re-weighting. However, briefly lowering the weight of noisy pairs cannot eliminate the negative impact of noisy correspondence in the training process. In this paper, we propose a novel self-drop and dual-weight approach, which achieves elaborate data processing by qua-partitioning the data. Specifically, our approach partitions all data into four types: clean and significant, clean yet insignificant, vague, and noisy. We analyze the effect of noisy and clean data pairs and find that for vision-language pre-training models, a small number of clean samples is more valuable than a majority of noisy ones. Based on this observation, we employ self-drop to discard noisy samples to effectively mitigate the impact of noise. In addition, we adopt a dual-weight strategy to ensure that the model focuses more on significant samples while appropriately leveraging vague samples. Compared to the prior works, our approach is more robust and demonstrates relatively more stable performance on noisy datasets, especially under a high noise ratio. Extensive experiments on three widely used datasets, including Flickr30K, MS-COCO, and Conceptual Captions, validate the effectiveness of our approach. The source code is available at https://github.com/DongChenwei2000/SDD.
Authors: Yuzhu Ji, Chuanxia Zheng, Tat-Jen Cham
Abstract: Human motion transfer aims at animating a static source image with a driving video. While recent advances in one-shot human motion transfer have led to significant improvement in results, it remains challenging for methods with 2D body landmarks, skeleton and semantic mask to accurately capture correspondences between source and driving poses due to the large variation in motion and articulation complexity. In addition, the accuracy and precision of DensePose degrade the image quality for neural-rendering-based methods. To address the limitations and by both considering the importance of appearance and geometry for motion transfer, in this work, we proposed a unified framework that combines multi-scale feature warping and neural texture mapping to recover better 2D appearance and 2.5D geometry, partly by exploiting the information from DensePose, yet adapting to its inherent limited accuracy. Our model takes advantage of multiple modalities by jointly training and fusing them, which allows it to robust neural texture features that cope with geometric errors as well as multi-scale dense motion flow that better preserves appearance. Experimental results with full and half-view body video datasets demonstrate that our model can generalize well and achieve competitive results, and that it is particularly effective in handling challenging cases such as those with substantial self-occlusions.
Authors: Zeng You, Zhiquan Wen, Yaofo Chen, Xin Li, Runhao Zeng, Yaowei Wang, Mingkui Tan
Abstract: Long video understanding has become a critical task in computer vision, driving advancements across numerous applications from surveillance to content retrieval. Existing video understanding methods suffer from two challenges when dealing with long video understanding: intricate long-context relationship modeling and interference from redundancy. To tackle these challenges, we introduce Fine-Detailed Video Story generation (FDVS), which interprets long videos into detailed textual representations. Specifically, to achieve fine-grained modeling of long-temporal content, we propose a Bottom-up Video Interpretation Mechanism that progressively interprets video content from clips to video. To avoid interference from redundant information in videos, we introduce a Semantic Redundancy Reduction mechanism that removes redundancy at both the visual and textual levels. Our method transforms long videos into hierarchical textual representations that contain multi-granularity information of the video. With these representations, FDVS is applicable to various tasks without any fine-tuning. We evaluate the proposed method across eight datasets spanning three tasks. The performance demonstrates the effectiveness and versatility of our method.
Authors: Lingjun Mao, Zineng Tang, Alane Suhr
Abstract: We study the perception of color illusions by vision-language models. Color illusion, where a person's visual system perceives color differently from actual color, is well-studied in human vision. However, it remains underexplored whether vision-language models (VLMs), trained on large-scale human data, exhibit similar perceptual biases when confronted with such color illusions. We propose an automated framework for generating color illusion images, resulting in RCID (Realistic Color Illusion Dataset), a dataset of 19,000 realistic illusion images. Our experiments show that all studied VLMs exhibit perceptual biases similar human vision. Finally, we train a model to distinguish both human perception and actual pixel differences.
Authors: Haijing Liu, Tao Pu, Hefeng Wu, Keze Wang, Liang Lin
Abstract: Benefiting from the generalization capability of CLIP, recent vision language pre-training (VLP) models have demonstrated an impressive ability to capture virtually any visual concept in daily images. However, due to the presence of unseen categories in open-vocabulary settings, existing algorithms struggle to effectively capture strong semantic correlations between categories, resulting in sub-optimal performance on the open-vocabulary multi-label recognition (OV-MLR). Furthermore, the substantial variation in the number of discriminative areas across diverse object categories is misaligned with the fixed-number patch matching used in current methods, introducing noisy visual cues that hinder the accurate capture of target semantics. To tackle these challenges, we propose a novel category-adaptive cross-modal semantic refinement and transfer (C$^2$SRT) framework to explore the semantic correlation both within each category and across different categories, in a category-adaptive manner. The proposed framework consists of two complementary modules, i.e., intra-category semantic refinement (ISR) module and inter-category semantic transfer (IST) module. Specifically, the ISR module leverages the cross-modal knowledge of the VLP model to adaptively find a set of local discriminative regions that best represent the semantics of the target category. The IST module adaptively discovers a set of most correlated categories for a target category by utilizing the commonsense capabilities of LLMs to construct a category-adaptive correlation graph and transfers semantic knowledge from the correlated seen categories to unseen ones. Extensive experiments on OV-MLR benchmarks clearly demonstrate that the proposed C$^2$SRT framework outperforms current state-of-the-art algorithms.
Authors: Ziyuan Qu, Zihao Zou, Vivek Boominathan, Praneeth Chakravarthula, Adithya Pediredla
Abstract: Event cameras, which feature pixels that independently respond to changes in brightness, are becoming increasingly popular in high-speed applications due to their lower latency, reduced bandwidth requirements, and enhanced dynamic range compared to traditional frame-based cameras. Numerous imaging and vision techniques have leveraged event cameras for high-speed scene understanding by capturing high-framerate, high-dynamic range videos, primarily utilizing the temporal advantages inherent to event cameras. Additionally, imaging and vision techniques have utilized the light field-a complementary dimension to temporal information-for enhanced scene understanding. In this work, we propose "Event Fields", a new approach that utilizes innovative optical designs for event cameras to capture light fields at high speed. We develop the underlying mathematical framework for Event Fields and introduce two foundational frameworks to capture them practically: spatial multiplexing to capture temporal derivatives and temporal multiplexing to capture angular derivatives. To realize these, we design two complementary optical setups one using a kaleidoscope for spatial multiplexing and another using a galvanometer for temporal multiplexing. We evaluate the performance of both designs using a custom-built simulator and real hardware prototypes, showcasing their distinct benefits. Our event fields unlock the full advantages of typical light fields-like post-capture refocusing and depth estimation-now supercharged for high-speed and high-dynamic range scenes. This novel light-sensing paradigm opens doors to new applications in photography, robotics, and AR/VR, and presents fresh challenges in rendering and machine learning.
Authors: Yohei Yamashita, Chihiro Nakatani, Norimichi Ukita
Abstract: This paper addresses a new virtual try-on problem of fitting any size of clothes to a reference person in the image domain. While previous image-based virtual try-on methods can produce highly natural try-on images, these methods fit the clothes on the person without considering the relative relationship between the physical sizes of the clothes and the person. Different from these methods, our method achieves size-variable virtual try-on in which the image size of the try-on clothes is changed depending on this relative relationship of the physical sizes. To relieve the difficulty in maintaining the physical size of the closes while synthesizing the high-fidelity image of the whole clothes, our proposed method focuses on the residual between the silhouettes of the clothes in the reference and try-on images. We also develop a size-variable virtual try-on dataset consisting of 1,524 images provided by 26 subjects. Furthermore, we propose an evaluation metric for size-variable virtual-try-on. Quantitative and qualitative experimental results show that our method can achieve size-variable virtual try-on better than general virtual try-on methods.
Authors: Yanqi Cheng, Carola-Bibiane Sch\"onlieb, Angelica I Aviles-Rivero
Abstract: The use of Plug-and-Play (PnP) methods has become a central approach for solving inverse problems, with denoisers serving as regularising priors that guide optimisation towards a clean solution. In this work, we introduce KAN-PnP, an optimisation framework that incorporates Kolmogorov-Arnold Networks (KANs) as denoisers within the Plug-and-Play (PnP) paradigm. KAN-PnP is specifically designed to solve inverse problems with single-instance priors, where only a single noisy observation is available, eliminating the need for large datasets typically required by traditional denoising methods. We show that KANs, based on the Kolmogorov-Arnold representation theorem, serve effectively as priors in such settings, providing a robust approach to denoising. We prove that the KAN denoiser is Lipschitz continuous, ensuring stability and convergence in optimisation algorithms like PnP-ADMM, even in the context of single-shot learning. Additionally, we provide theoretical guarantees for KAN-PnP, demonstrating its convergence under key conditions: the convexity of the data fidelity term, Lipschitz continuity of the denoiser, and boundedness of the regularisation functional. These conditions are crucial for stable and reliable optimisation. Our experimental results show, on super-resolution and joint optimisation, that KAN-PnP outperforms exiting methods, delivering superior performance in single-shot learning with minimal data. The method exhibits strong convergence properties, achieving high accuracy with fewer iterations.
Authors: Kim Sung-Bin, Arda Senocak, Hyunwoo Ha, Tae-Hyun Oh
Abstract: How does audio describe the world around us? In this work, we propose a method for generating images of visual scenes from diverse in-the-wild sounds. This cross-modal generation task is challenging due to the significant information gap between auditory and visual signals. We address this challenge by designing a model that aligns audio-visual modalities by enriching audio features with visual information and translating them into the visual latent space. These features are then fed into the pre-trained image generator to produce images. To enhance image quality, we use sound source localization to select audio-visual pairs with strong cross-modal correlations. Our method achieves substantially better results on the VEGAS and VGGSound datasets compared to previous work and demonstrates control over the generation process through simple manipulations to the input waveform or latent space. Furthermore, we analyze the geometric properties of the learned embedding space and demonstrate that our learning approach effectively aligns audio-visual signals for cross-modal generation. Based on this analysis, we show that our method is agnostic to specific design choices, showing its generalizability by integrating various model architectures and different types of audio-visual data.
Authors: Qinfeng Zhu, Yuan Fang, Lei Fan
Abstract: Crack detection is a critical task in structural health monitoring, aimed at assessing the structural integrity of bridges, buildings, and roads to prevent potential failures. Vision-based crack detection has become the mainstream approach due to its ease of implementation and effectiveness. Fusing infrared (IR) channels with red, green and blue (RGB) channels can enhance feature representation and thus improve crack detection. However, IR and RGB channels often differ in resolution. To align them, higher-resolution RGB images typically need to be downsampled to match the IR image resolution, which leads to the loss of fine details. Moreover, crack detection performance is restricted by the limited receptive fields and high computational complexity of traditional image segmentation networks. Inspired by the recently proposed Mamba neural architecture, this study introduces a two-stage paradigm called MSCrackMamba, which leverages Vision Mamba along with a super-resolution network to address these challenges. Specifically, to align IR and RGB channels, we first apply super-resolution to IR channels to match the resolution of RGB channels for data fusion. Vision Mamba is then adopted as the backbone network, while UperNet is employed as the decoder for crack detection. Our approach is validated on the large-scale Crack Detection dataset Crack900, demonstrating an improvement of 3.55% in mIoU compared to the best-performing baseline methods.
Authors: Jaden Mu
Abstract: Autonomous vehicles (AVs) increasingly use DNN-based object detection models in vision-based perception. Correct detection and classification of obstacles is critical to ensure safe, trustworthy driving decisions. Adversarial patches aim to fool a DNN with intentionally generated patterns concentrated in a localized region of an image. In particular, object vanishing patch attacks can cause object detection models to fail to detect most or all objects in a scene, posing a significant practical threat to AVs. This work proposes ADAV (Adversarial Defense for Autonomous Vehicles), a novel defense methodology against object vanishing patch attacks specifically designed for autonomous vehicles. Unlike existing defense methods which have high latency or are designed for static images, ADAV runs in real-time and leverages contextual information from prior frames in an AV's video feed. ADAV checks if the object detector's output for the target frame is temporally consistent with the output from a previous reference frame to detect the presence of a patch. If the presence of a patch is detected, ADAV uses gradient-based attribution to localize adversarial pixels that break temporal consistency. This two stage procedure allows ADAV to efficiently process clean inputs, and both stages are optimized to be low latency. ADAV is evaluated using real-world driving data from the Berkeley Deep Drive BDD100K dataset, and demonstrates high adversarial and clean performance.
Authors: Marsha Mariya Kappan, Eduardo Benitez Sandoval, Erik Meijering, Francisco Cruz
Abstract: Pose estimation is a critical task in computer vision with a wide range of applications from activity monitoring to human-robot interaction. However,most of the existing methods are computationally expensive or have complex architecture. Here we propose a lightweight attention based pose estimation network that utilizes depthwise separable convolution and Convolutional Block Attention Module on an hourglass backbone. The network significantly reduces the computational complexity (floating point operations) and the model size (number of parameters) containing only about 10% of parameters of original eight stack Hourglass network.Experiments were conducted on COCO and MPII datasets using a two stack hourglass backbone. The results showed that our model performs well in comparison to six other lightweight pose estimation models with an average precision of 72.07. The model achieves this performance with only 2.3M parameters and 3.7G FLOPs.
Authors: Seungtae Nam, Xiangyu Sun, Gyeongjin Kang, Younggeun Lee, Seungjun Oh, Eunbyung Park
Abstract: Generalized feed-forward Gaussian models have achieved significant progress in sparse-view 3D reconstruction by leveraging prior knowledge from large multi-view datasets. However, these models often struggle to represent high-frequency details due to the limited number of Gaussians. While the densification strategy used in per-scene 3D Gaussian splatting (3D-GS) optimization can be adapted to the feed-forward models, it may not be ideally suited for generalized scenarios. In this paper, we propose Generative Densification, an efficient and generalizable method to densify Gaussians generated by feed-forward models. Unlike the 3D-GS densification strategy, which iteratively splits and clones raw Gaussian parameters, our method up-samples feature representations from the feed-forward models and generates their corresponding fine Gaussians in a single forward pass, leveraging the embedded prior knowledge for enhanced generalization. Experimental results on both object-level and scene-level reconstruction tasks demonstrate that our method outperforms state-of-the-art approaches with comparable or smaller model sizes, achieving notable improvements in representing fine details.
Authors: Michael Yeung, Toya Teramoto, Songtao Wu, Tatsuo Fujiwara, Kenji Suzuki, Tamaki Kojima
Abstract: The use of large-scale, web-scraped datasets to train face recognition models has raised significant privacy and bias concerns. Synthetic methods mitigate these concerns and provide scalable and controllable face generation to enable fair and accurate face recognition. However, existing synthetic datasets display limited intraclass and interclass diversity and do not match the face recognition performance obtained using real datasets. Here, we propose VariFace, a two-stage diffusion-based pipeline to create fair and diverse synthetic face datasets to train face recognition models. Specifically, we introduce three methods: Face Recognition Consistency to refine demographic labels, Face Vendi Score Guidance to improve interclass diversity, and Divergence Score Conditioning to balance the identity preservation-intraclass diversity trade-off. When constrained to the same dataset size, VariFace considerably outperforms previous synthetic datasets (0.9200 $\rightarrow$ 0.9405) and achieves comparable performance to face recognition models trained with real data (Real Gap = -0.0065). In an unconstrained setting, VariFace not only consistently achieves better performance compared to previous synthetic methods across dataset sizes but also, for the first time, outperforms the real dataset (CASIA-WebFace) across six evaluation datasets. This sets a new state-of-the-art performance with an average face verification accuracy of 0.9567 (Real Gap = +0.0097) across LFW, CFP-FP, CPLFW, AgeDB, and CALFW datasets and 0.9366 (Real Gap = +0.0380) on the RFW dataset.
Authors: Sungpyo Kim, Jeonghyeok Do, Jaehyup Lee, Munchurl Kim
Abstract: Conventional methods for PAN-sharpening often struggle to restore fine details due to limitations in leveraging high-frequency information. Moreover, diffusion-based approaches lack sufficient conditioning to fully utilize Panchromatic (PAN) images and low-resolution multispectral (LRMS) inputs effectively. To address these challenges, we propose an uncertainty-aware knowledge distillation diffusion framework with details enhancement for PAN-sharpening, called U-Know-DiffPAN. The U-Know-DiffPAN incorporates uncertainty-aware knowledge distillation for effective transfer of feature details from our teacher model to a student one. The teacher model in our U-Know-DiffPAN captures frequency details through freqeuncy selective attention, facilitating accurate reverse process learning. By conditioning the encoder on compact vector representations of PAN and LRMS and the decoder on Wavelet transforms, we enable rich frequency utilization. So, the high-capacity teacher model distills frequency-rich features into a lightweight student model aided by an uncertainty map. From this, the teacher model can guide the student model to focus on difficult image regions for PAN-sharpening via the usage of the uncertainty map. Extensive experiments on diverse datasets demonstrate the robustness and superior performance of our U-Know-DiffPAN over very recent state-of-the-art PAN-sharpening methods.
Authors: Yunheng Li, Yuxuan Li, Quansheng Zeng, Wenhai Wang, Qibin Hou, Ming-Ming Cheng
Abstract: Pre-trained vision-language models (VLMs), such as CLIP, have demonstrated impressive zero-shot recognition capability, but still underperform in dense prediction tasks. Self-distillation recently is emerging as a promising approach for fine-tuning VLMs to better adapt to local regions without requiring extensive annotations. However, previous state-of-the-art approaches often suffer from significant `foreground bias', where models tend to wrongly identify background regions as foreground objects. To alleviate this issue, we propose DenseVLM, a framework designed to learn unbiased region-language alignment from powerful pre-trained VLM representations. By leveraging the pre-trained VLM to retrieve categories for unlabeled regions, DenseVLM effectively decouples the interference between foreground and background region features, ensuring that each region is accurately aligned with its corresponding category. We show that DenseVLM can be seamlessly integrated into open-vocabulary object detection and image segmentation tasks, leading to notable performance improvements. Furthermore, it exhibits promising zero-shot scalability when training on more extensive and diverse datasets.
Authors: Kartik Patwari, David Schneider, Xiaoxiao Sun, Chen-Nee Chuah, Lingjuan Lyu, Vivek Sharma
Abstract: Growing privacy concerns and regulations like GDPR and CCPA necessitate pseudonymization techniques that protect identity in image datasets. However, retaining utility is also essential. Traditional methods like masking and blurring degrade quality and obscure critical context, especially in human-centric images. We introduce Rendering-Refined Stable Diffusion (RefSD), a pipeline that combines 3D-rendering with Stable Diffusion, enabling prompt-based control over human attributes while preserving posture. Unlike standard diffusion models that fail to retain posture or GANs that lack realism and flexible attribute control, RefSD balances posture preservation, realism, and customization. We also propose HumanGenAI, a framework for human perception and utility evaluation. Human perception assessments reveal attribute-specific strengths and weaknesses of RefSD. Our utility experiments show that models trained on RefSD pseudonymized data outperform those trained on real data in detection tasks, with further performance gains when combining RefSD with real data. For classification tasks, we consistently observe performance improvements when using RefSD data with real data, confirming the utility of our pseudonymized data.
Authors: Zheng Chen, Chenming Wu, Zhelun Shen, Chen Zhao, Weicai Ye, Haocheng Feng, Errui Ding, Song-Hai Zhang
Abstract: Wide-baseline panoramic images are frequently used in applications like VR and simulations to minimize capturing labor costs and storage needs. However, synthesizing novel views from these panoramic images in real time remains a significant challenge, especially due to panoramic imagery's high resolution and inherent distortions. Although existing 3D Gaussian splatting (3DGS) methods can produce photo-realistic views under narrow baselines, they often overfit the training views when dealing with wide-baseline panoramic images due to the difficulty in learning precise geometry from sparse 360$^{\circ}$ views. This paper presents \textit{Splatter-360}, a novel end-to-end generalizable 3DGS framework designed to handle wide-baseline panoramic images. Unlike previous approaches, \textit{Splatter-360} performs multi-view matching directly in the spherical domain by constructing a spherical cost volume through a spherical sweep algorithm, enhancing the network's depth perception and geometry estimation. Additionally, we introduce a 3D-aware bi-projection encoder to mitigate the distortions inherent in panoramic images and integrate cross-view attention to improve feature interactions across multiple viewpoints. This enables robust 3D-aware feature representations and real-time rendering capabilities. Experimental results on the HM3D~\cite{hm3d} and Replica~\cite{replica} demonstrate that \textit{Splatter-360} significantly outperforms state-of-the-art NeRF and 3DGS methods (e.g., PanoGRF, MVSplat, DepthSplat, and HiSplat) in both synthesis quality and generalization performance for wide-baseline panoramic images. Code and trained models are available at \url{https://3d-aigc.github.io/Splatter-360/}.
Authors: Shi Qiu, Binzhu Xie, Qixuan Liu, Pheng-Ann Heng
Abstract: 3D Gaussian Splatting (3DGS) has attracted significant attention for its potential to revolutionize 3D representation, rendering, and interaction. Despite the rapid growth of 3DGS research, its direct application to Extended Reality (XR) remains underexplored. Although many studies recognize the potential of 3DGS for XR, few have explicitly focused on or demonstrated its effectiveness within XR environments. In this paper, we aim to synthesize innovations in 3DGS that show specific potential for advancing XR research and development. We conduct a comprehensive review of publicly available 3DGS papers, with a focus on those referencing XR-related concepts. Additionally, we perform an in-depth analysis of innovations explicitly relevant to XR and propose a taxonomy to highlight their significance. Building on these insights, we propose several prospective XR research areas where 3DGS can make promising contributions, yet remain rarely touched. By investigating the intersection of 3DGS and XR, this paper provides a roadmap to push the boundaries of XR using cutting-edge 3DGS techniques.
Authors: Li Yin, Calvin Yeung, Qingrui Hu, Jun Ichikawa, Hirotsugu Azechi, Susumu Takahashi, Keisuke Fujii
Abstract: Multi-object tracking (MOT) is crucial for various multi-agent analyses such as evaluating team sports tactics and player movements and performance. While pedestrian tracking has advanced with Tracking-by-Detection MOT, team sports like basketball pose unique challenges. These challenges include players' unpredictable movements, frequent close interactions, and visual similarities that complicate pose labeling and lead to significant occlusions, frequent ID switches, and high manual annotation costs. To address these challenges, we propose a novel pose-based virtual marker (VM) MOT method for team sports, named Sports-vmTracking. This method builds on the vmTracking approach developed for multi-animal tracking with active learning. First, we constructed a 3x3 basketball pose dataset for VMs and applied active learning to enhance model performance in generating VMs. Then, we overlaid the VMs on video to identify players, extract their poses with unique IDs, and convert these into bounding boxes for comparison with automated MOT methods. Using our 3x3 basketball dataset, we demonstrated that our VM configuration has been highly effective, and reduced the need for manual corrections and labeling during pose model training while maintaining high accuracy. Our approach achieved an average HOTA score of 72.3%, over 10 points higher than other state-of-the-art methods without VM, and resulted in 0 ID switches. Beyond improving performance in handling occlusions and minimizing ID switches, our framework could substantially increase the time and cost efficiency compared to traditional manual annotation.
Authors: Quansong He, Xiaojun Yao, Jun Wu, Zhang Yi, Tao He
Abstract: In recent years, advanced U-like networks have demonstrated remarkable performance in medical image segmentation tasks. However, their drawbacks, including excessive parameters, high computational complexity, and slow inference speed, pose challenges for practical implementation in scenarios with limited computational resources. Existing lightweight U-like networks have alleviated some of these problems, but they often have pre-designed structures and consist of inseparable modules, limiting their application scenarios. In this paper, we propose three plug-and-play decoders by employing different discretization methods of the neural memory Ordinary Differential Equations (nmODEs). These decoders integrate features at various levels of abstraction by processing information from skip connections and performing numerical operations on upward path. Through experiments on the PH2, ISIC2017, and ISIC2018 datasets, we embed these decoders into different U-like networks, demonstrating their effectiveness in significantly reducing the number of parameters and FLOPs while maintaining performance. In summary, the proposed discretized nmODEs decoders are capable of reducing the number of parameters by about 20% ~ 50% and FLOPs by up to 74%, while possessing the potential to adapt to all U-like networks. Our code is available at https://github.com/nayutayuki/Lightweight-nmODE-Decoders-For-U-like-networks.
URLs: https://github.com/nayutayuki/Lightweight-nmODE-Decoders-For-U-like-networks.
Authors: Lianyu Hu, Fanhua Shang, Liang Wan, Wei Feng
Abstract: In this paper, we introduce iLLaVA, a simple method that can be seamlessly deployed upon current Large Vision-Language Models (LVLMs) to greatly increase the throughput with nearly lossless model performance, without a further requirement to train. iLLaVA achieves this by finding and gradually merging the redundant tokens with an accurate and fast algorithm, which can merge hundreds of tokens within only one step. While some previous methods have explored directly pruning or merging tokens in the inference stage to accelerate models, our method excels in both performance and throughput by two key designs. First, while most previous methods only try to save the computations of Large Language Models (LLMs), our method accelerates the forward pass of both image encoders and LLMs in LVLMs, which both occupy a significant part of time during inference. Second, our method recycles the beneficial information from the pruned tokens into existing tokens, which avoids directly dropping context tokens like previous methods to cause performance loss. iLLaVA can nearly 2$\times$ the throughput, and reduce the memory costs by half with only a 0.2\% - 0.5\% performance drop across models of different scales including 7B, 13B and 34B. On tasks across different domains including single-image, multi-images and videos, iLLaVA demonstrates strong generalizability with consistently promising efficiency. We finally offer abundant visualizations to show the merging processes of iLLaVA in each step, which show insights into the distribution of computing resources in LVLMs. Code is available at https://github.com/hulianyuyy/iLLaVA.
Authors: Jiuyi Xu, Meida Chen, Andrew Feng, Yangming Shi, Zifan Yu
Abstract: In the domain of the U.S. Army modeling and simulation, the availability of high quality annotated 3D data is pivotal to creating virtual environments for training and simulations. Traditional methodologies for 3D semantic and instance segmentation, such as KpConv, RandLA, Mask3D, etc., are designed to train on extensive labeled datasets to obtain satisfactory performance in practical tasks. This requirement presents a significant challenge, given the inherent scarcity of manually annotated 3D datasets, particularly for the military use cases. Recognizing this gap, our previous research leverages the One World Terrain data repository manually annotated databases, as showcased at IITSEC 2019 and 2021, to enrich the training dataset for deep learning models. However, collecting and annotating large scale 3D data for specific tasks remains costly and inefficient. To this end, the objective of this research is to design and develop a comprehensive and efficient framework for 3D segmentation tasks to assist in 3D data annotation. This framework integrates Grounding DINO and Segment anything Model, augmented by an enhancement in 2D image rendering via 3D mesh. Furthermore, the authors have also developed a user friendly interface that facilitates the 3D annotation process, offering intuitive visualization of rendered images and the 3D point cloud.
Authors: Dongxu Wei, Zhiqi Li, Peidong Liu
Abstract: Prior works employing pixel-based Gaussian representation have demonstrated efficacy in feed-forward sparse-view reconstruction. However, such representation necessitates cross-view overlap for accurate depth estimation, and is challenged by object occlusions and frustum truncations. As a result, these methods require scene-centric data acquisition to maintain cross-view overlap and complete scene visibility to circumvent occlusions and truncations, which limits their applicability to scene-centric reconstruction. In contrast, in autonomous driving scenarios, a more practical paradigm is ego-centric reconstruction, which is characterized by minimal cross-view overlap and frequent occlusions and truncations. The limitations of pixel-based representation thus hinder the utility of prior works in this task. In light of this, this paper conducts an in-depth analysis of different representations, and introduces Omni-Gaussian representation with tailored network design to complement their strengths and mitigate their drawbacks. Experiments show that our method significantly surpasses state-of-the-art methods, pixelSplat and MVSplat, in ego-centric reconstruction, and achieves comparable performance to prior works in scene-centric reconstruction. Furthermore, we extend our method with diffusion models, pioneering feed-forward multi-modal generation of 3D driving scenes.
Authors: Xiang Fang, Arvind Easwaran, Blaise Genest, Ponnuthurai Nagaratnam Suganthan
Abstract: Previous OOD detection systems only focus on the semantic gap between ID and OOD samples. Besides the semantic gap, we are faced with two additional gaps: the domain gap between source and target domains, and the class-imbalance gap between different classes. In fact, similar objects from different domains should belong to the same class. In this paper, we introduce a realistic yet challenging setting: class-imbalanced cross-domain OOD detection (CCOD), which contains a well-labeled (but usually small) source set for training and conducts OOD detection on an unlabeled (but usually larger) target set for testing. We do not assume that the target domain contains only OOD classes or that it is class-balanced: the distribution among classes of the target dataset need not be the same as the source dataset. To tackle this challenging setting with an OOD detection system, we propose a novel uncertainty-aware adaptive semantic alignment (UASA) network based on a prototype-based alignment strategy. Specifically, we first build label-driven prototypes in the source domain and utilize these prototypes for target classification to close the domain gap. Rather than utilizing fixed thresholds for OOD detection, we generate adaptive sample-wise thresholds to handle the semantic gap. Finally, we conduct uncertainty-aware clustering to group semantically similar target samples to relieve the class-imbalance gap. Extensive experiments on three challenging benchmarks demonstrate that our proposed UASA outperforms state-of-the-art methods by a large margin.
Authors: Meng Zhang, Jun Li
Abstract: Achieving efficient, high-fidelity, high-resolution garment simulation is challenging due to its computational demands. Conversely, low-resolution garment simulation is more accessible and ideal for low-budget devices like smartphones. In this paper, we introduce a lightweight, learning-based method for garment dynamic super-resolution, designed to efficiently enhance high-resolution, high-frequency details in low-resolution garment simulations. Starting with low-resolution garment simulation and underlying body motion, we utilize a mesh-graph-net to compute super-resolution features based on coarse garment dynamics and garment-body interactions. These features are then used by a hyper-net to construct an implicit function of detailed wrinkle residuals for each coarse mesh triangle. Considering the influence of coarse garment shapes on detailed wrinkle performance, we correct the coarse garment shape and predict detailed wrinkle residuals using these implicit functions. Finally, we generate detailed high-resolution garment geometry by applying the detailed wrinkle residuals to the corrected coarse garment. Our method enables roll-out prediction by iteratively using its predictions as input for subsequent frames, producing fine-grained wrinkle details to enhance the low-resolution simulation. Despite training on a small dataset, our network robustly generalizes to different body shapes, motions, and garment types not present in the training data. We demonstrate significant improvements over state-of-the-art alternatives, particularly in enhancing the quality of high-frequency, fine-grained wrinkle details.
Authors: Patrick Ramos, Nicolas Gonthier, Selina Khan, Yuta Nakashima, Noa Garcia
Abstract: Object detection in art is a valuable tool for the digital humanities, as it allows for faster identification of objects in artistic and historical images compared to humans. However, annotating such images poses significant challenges due to the need for specialized domain expertise. We present NADA (no annotations for detection in art), a pipeline that leverages diffusion models' art-related knowledge for object detection in paintings without the need for full bounding box supervision. Our method, which supports both weakly-supervised and zero-shot scenarios and does not require any fine-tuning of its pretrained components, consists of a class proposer based on large vision-language models and a class-conditioned detector based on Stable Diffusion. NADA is evaluated on two artwork datasets, ArtDL 2.0 and IconArt, outperforming prior work in weakly-supervised detection, while being the first work for zero-shot object detection in art. Code is available at https://github.com/patrick-john-ramos/nada
Authors: Bingchen Gong, Diego Gomez, Abdullah Hamdi, Abdelrahman Eldesokey, Ahmed Abdelreheem, Peter Wonka, Maks Ovsjanikov
Abstract: We propose a novel zero-shot approach for keypoint detection on 3D shapes. Point-level reasoning on visual data is challenging as it requires precise localization capability, posing problems even for powerful models like DINO or CLIP. Traditional methods for 3D keypoint detection rely heavily on annotated 3D datasets and extensive supervised training, limiting their scalability and applicability to new categories or domains. In contrast, our method utilizes the rich knowledge embedded within Multi-Modal Large Language Models (MLLMs). Specifically, we demonstrate, for the first time, that pixel-level annotations used to train recent MLLMs can be exploited for both extracting and naming salient keypoints on 3D models without any ground truth labels or supervision. Experimental evaluations demonstrate that our approach achieves competitive performance on standard benchmarks compared to supervised methods, despite not requiring any 3D keypoint annotations during training. Our results highlight the potential of integrating language models for localized 3D shape understanding. This work opens new avenues for cross-modal learning and underscores the effectiveness of MLLMs in contributing to 3D computer vision challenges.
Authors: Qifan Yu, Zhebei Shen, Zhongqi Yue, Yang Wu, Wenqiao Zhang, Yunfei Li, Juncheng Li, Siliang Tang, Yueting Zhuang
Abstract: Instruction tuning fine-tunes pre-trained Multi-modal Large Language Models (MLLMs) to handle real-world tasks. However, the rapid expansion of visual instruction datasets introduces data redundancy, leading to excessive computational costs. We propose a collaborative framework, DataTailor, which leverages three key principles--informativeness, uniqueness, and representativeness--for effective data selection. We argue that a valuable sample should be informative of the task, non-redundant, and represent the sample distribution (i.e., not an outlier). We further propose practical ways to score against each principle, which automatically adapts to a given dataset without tedious hyperparameter tuning. Comprehensive experiments on various benchmarks demonstrate that DataTailor achieves 100.8% of the performance of full-data fine-tuning with only 15% of the data, significantly reducing computational costs while maintaining superior results. This exemplifies the "Less is More" philosophy in MLLM development.
Authors: Yunpeng Liu, Boxiao Liu, Yi Zhang, Xingzhong Hou, Guanglu Song, Yu Liu, Haihang You
Abstract: Significant advances have been made in the sampling efficiency of diffusion models and flow matching models, driven by Consistency Distillation (CD), which trains a student model to mimic the output of a teacher model at a later timestep. However, we found that the learning complexity of the student model varies significantly across different timesteps, leading to suboptimal performance in CD.To address this issue, we propose the Curriculum Consistency Model (CCM), which stabilizes and balances the learning complexity across timesteps. Specifically, we regard the distillation process at each timestep as a curriculum and introduce a metric based on Peak Signal-to-Noise Ratio (PSNR) to quantify the learning complexity of this curriculum, then ensure that the curriculum maintains consistent learning complexity across different timesteps by having the teacher model iterate more steps when the noise intensity is low. Our method achieves competitive single-step sampling Fr\'echet Inception Distance (FID) scores of 1.64 on CIFAR-10 and 2.18 on ImageNet 64x64.Moreover, we have extended our method to large-scale text-to-image models and confirmed that it generalizes well to both diffusion models (Stable Diffusion XL) and flow matching models (Stable Diffusion 3). The generated samples demonstrate improved image-text alignment and semantic structure, since CCM enlarges the distillation step at large timesteps and reduces the accumulated error.
Authors: Jinbo Yan, Rui Peng, Luyang Tang, Ronggang Wang
Abstract: Reconstructing dynamic scenes from video sequences is a highly promising task in the multimedia domain. While previous methods have made progress, they often struggle with slow rendering and managing temporal complexities such as significant motion and object appearance/disappearance. In this paper, we propose SaRO-GS as a novel dynamic scene representation capable of achieving real-time rendering while effectively handling temporal complexities in dynamic scenes. To address the issue of slow rendering speed, we adopt a Gaussian primitive-based representation and optimize the Gaussians in 4D space, which facilitates real-time rendering with the assistance of 3D Gaussian Splatting. Additionally, to handle temporally complex dynamic scenes, we introduce a Scale-aware Residual Field. This field considers the size information of each Gaussian primitive while encoding its residual feature and aligns with the self-splitting behavior of Gaussian primitives. Furthermore, we propose an Adaptive Optimization Schedule, which assigns different optimization strategies to Gaussian primitives based on their distinct temporal properties, thereby expediting the reconstruction of dynamic regions. Through evaluations on monocular and multi-view datasets, our method has demonstrated state-of-the-art performance. Please see our project page at https://yjb6.github.io/SaRO-GS.github.io.
Authors: Zi-Wei Sun, Ze-Xi hua, Heng-Chao Li, Yan Li
Abstract: In order to avoid the impact of hard samples on the training process of the Flying Bird Object Detection model (FBOD model, in our previous work, we designed the FBOD model according to the characteristics of flying bird objects in surveillance video), the Self-Paced Learning strategy with Easy Sample Prior Based on Confidence (SPL-ESP-BC), a new model training strategy, is proposed. Firstly, the loss-based Minimizer Function in Self-Paced Learning (SPL) is improved, and the confidence-based Minimizer Function is proposed, which makes it more suitable for one-class object detection tasks. Secondly, to give the model the ability to judge easy and hard samples at the early stage of training by using the SPL strategy, an SPL strategy with Easy Sample Prior (ESP) is proposed. The FBOD model is trained using the standard training strategy with easy samples first, then the SPL strategy with all samples is used to train it. Combining the strategy of the ESP and the Minimizer Function based on confidence, the SPL-ESP-BC model training strategy is proposed. Using this strategy to train the FBOD model can make it to learn the characteristics of the flying bird object in the surveillance video better, from easy to hard. The experimental results show that compared with the standard training strategy that does not distinguish between easy and hard samples, the AP50 of the FBOD model trained by the SPL-ESP-BC is increased by 2.1%, and compared with other loss-based SPL strategies, the FBOD model trained with SPL-ESP-BC strategy has the best comprehensive detection performance.
Authors: Mingjie Xu, Mengyang Wu, Yuzhi Zhao, Jason Chun Lok Li, Weifeng Ou
Abstract: Scene Graph Generation (SGG) converts visual scenes into structured graph representations, providing deeper scene understanding for complex vision tasks. However, existing SGG models often overlook essential spatial relationships and struggle with generalization in open-vocabulary contexts. To address these limitations, we propose LLaVA-SpaceSGG, a multimodal large language model (MLLM) designed for open-vocabulary SGG with enhanced spatial relation modeling. To train it, we collect the SGG instruction-tuning dataset, named SpaceSGG. This dataset is constructed by combining publicly available datasets and synthesizing data using open-source models within our data construction pipeline. It combines object locations, object relations, and depth information, resulting in three data formats: spatial SGG description, question-answering, and conversation. To enhance the transfer of MLLMs' inherent capabilities to the SGG task, we introduce a two-stage training paradigm. Experiments show that LLaVA-SpaceSGG outperforms other open-vocabulary SGG methods, boosting recall by 8.6% and mean recall by 28.4% compared to the baseline. Our codebase, dataset, and trained models are publicly accessible on GitHub at the following URL: https://github.com/Endlinc/LLaVA-SpaceSGG.
Authors: Florian Strohm, Mihai B\^ace, Andreas Bulling
Abstract: We present HAIFAI - a novel collaborative human-AI system to tackle the challenging task of reconstructing a visual representation of a face that exists only in a person's mind. Users iteratively rank images presented by the AI system based on their resemblance to a mental image. These rankings, in turn, allow the system to extract relevant image features, fuse them into a unified feature vector, and use a generative model to reconstruct the mental image. We also propose an extension called HAIFAI-X that allows users to manually refine and further improve the reconstruction using an easy-to-use slider interface. To avoid the need for tedious human data collection for model training, we introduce a computational user model of human ranking behaviour. For this, we collected a small face ranking dataset through an online crowd-sourcing study containing data from 275 participants. We evaluate HAIFAI and HAIFAI-X in a 12-participant user study and show that HAIFAI outperforms the previous state of the art regarding reconstruction quality, usability, perceived workload, and reconstruction speed. HAIFAI-X achieves even better reconstruction quality at the cost of reduced usability, perceived workload, and increased reconstruction time. We further validate the reconstructions in a subsequent face ranking study with 18 participants and show that HAIFAI-X achieves a new state-of-the-art identification rate of 60.6%. These findings represent a significant advancement towards developing new collaborative intelligent systems capable of reliably and effortlessly reconstructing a user's mental image.
Authors: Mingliang Zhai, Cheng Li, Zengyuan Guo, Ningrui Yang, Xiameng Qin, Yuwei Wu, Sanyuan Zhao, Junyu Han, Ji Tao, Yunde Jia
Abstract: The Multi-modal Large Language Models (MLLMs) with extensive world knowledge have revitalized autonomous driving, particularly in reasoning tasks within perceivable regions. However, when faced with perception-limited areas (dynamic or static occlusion regions), MLLMs struggle to effectively integrate perception ability with world knowledge for reasoning. These perception-limited regions can conceal crucial safety information, especially for vulnerable road users. In this paper, we propose a framework, which aims to improve autonomous driving performance under perceptionlimited conditions by enhancing the integration of perception capabilities and world knowledge. Specifically, we propose a plug-and-play instruction-guided interaction module that bridges modality gaps and significantly reduces the input sequence length, allowing it to adapt effectively to multi-view video inputs. Furthermore, to better integrate world knowledge with driving-related tasks, we have collected and refined a large-scale multi-modal dataset that includes 2 million natural language QA pairs, 1.7 million grounding task data. To evaluate the model's utilization of world knowledge, we introduce an object-level risk assessment dataset comprising 200K QA pairs, where the questions necessitate multi-step reasoning leveraging world knowledge for resolution. Extensive experiments validate the effectiveness of our proposed method.
Authors: Shuangfei Zhai, Ruixiang Zhang, Preetum Nakkiran, David Berthelot, Jiatao Gu, Huangjie Zheng, Tianrong Chen, Miguel Angel Bautista, Navdeep Jaitly, Josh Susskind
Abstract: Normalizing Flows (NFs) are likelihood-based models for continuous inputs. They have demonstrated promising results on both density estimation and generative modeling tasks, but have received relatively little attention in recent years. In this work, we demonstrate that NFs are more powerful than previously believed. We present \textit{TarFlow}: a simple and scalable architecture that enables highly performant NF models. TarFlow can be thought of as a Transformer-based variant of Masked Autoregressive Flows (MAFs): it consists of a stack of autoregressive Transformer blocks on image patches, alternating the autoregression direction between layers. TarFlow is straightforward to train end-to-end, and capable of directly modeling and generating pixels. We also propose three key techniques to improve sample quality: Gaussian noise augmentation during training, a post training denoising procedure, and an effective guidance method for both class-conditional and unconditional settings. Putting these together, TarFlow sets new state-of-the-art results on likelihood estimation for images, beating the previous best methods by a large margin, and generates samples with quality and diversity comparable to diffusion models, for the first time with a stand-alone NF model. We make our code available at \href{https://github.com/apple/ml-tarflow}{https://github.com/apple/ml-tarflow}.
URLs: https://github.com/apple/ml-tarflow, https://github.com/apple/ml-tarflow
Authors: Ilya A. Petrov, Riccardo Marin, Julian Chibane, Gerard Pons-Moll
Abstract: Modeling 3D human-object interaction (HOI) is a problem of great interest for computer vision and a key enabler for virtual and mixed-reality applications. Existing methods work in a one-way direction: some recover plausible human interactions conditioned on a 3D object; others recover the object pose conditioned on a human pose. Instead, we provide the first unified model - TriDi which works in any direction. Concretely, we generate Human, Object, and Interaction modalities simultaneously with a new three-way diffusion process, allowing to model seven distributions with one network. We implement TriDi as a transformer attending to the various modalities' tokens, thereby discovering conditional relations between them. The user can control the interaction either as a text description of HOI or a contact map. We embed these two representations into a shared latent space, combining the practicality of text descriptions with the expressiveness of contact maps. Using a single network, TriDi unifies all the special cases of prior work and extends to new ones, modeling a family of seven distributions. Remarkably, despite using a single model, TriDi generated samples surpass one-way specialized baselines on GRAB and BEHAVE in terms of both qualitative and quantitative metrics, and demonstrating better diversity. We show the applicability of TriDi to scene population, generating objects for human-contact datasets, and generalization to unseen object geometry. The project page is available at: https://virtualhumans.mpi-inf.mpg.de/tridi.
Authors: Zhen Wan, Yue Ma, Chenyang Qi, Zhiheng Liu, Tao Gui
Abstract: In this paper, we present UniPaint, a unified generative space-time video inpainting framework that enables spatial-temporal inpainting and interpolation. Different from existing methods that treat video inpainting and video interpolation as two distinct tasks, we leverage a unified inpainting framework to tackle them and observe that these two tasks can mutually enhance synthesis performance. Specifically, we first introduce a plug-and-play space-time video inpainting adapter, which can be employed in various personalized models. The key insight is to propose a Mixture of Experts (MoE) attention to cover various tasks. Then, we design a spatial-temporal masking strategy during the training stage to mutually enhance each other and improve performance. UniPaint produces high-quality and aesthetically pleasing results, achieving the best quantitative results across various tasks and scale setups. The code and checkpoints will be available soon.
Authors: Daeun Seo, Hoeseok Yang, Sihyeong Park, Hyungshin Kim
Abstract: Multi-scale image resolution is a de facto standard approach in modern object detectors, such as DETR. This technique allows for the acquisition of various scale information from multiple image resolutions. However, manual hyperparameter selection of the resolution can restrict its flexibility, which is informed by prior knowledge, necessitating human intervention. This work introduces a novel strategy for learnable resolution, called Elastic-DETR, enabling elastic utilization of multiple image resolutions. Our network provides an adaptive scale factor based on the content of the image with a compact scale prediction module (< 2 GFLOPs). The key aspect of our method lies in how to determine the resolution without prior knowledge. We present two loss functions derived from identified key components for resolution optimization: scale loss, which increases adaptiveness according to the image, and distribution loss, which determines the overall degree of scaling based on network performance. By leveraging the resolution's flexibility, we can demonstrate various models that exhibit varying trade-offs between accuracy and computational complexity. We empirically show that our scheme can unleash the potential of a wide spectrum of image resolutions without constraining flexibility. Our models on MS COCO establish a maximum accuracy gain of 3.5%p or 26% decrease in computation than MS-trained DN-DETR.
Authors: Zeru Shi, Zengxi Zhang, Zhiying Jiang, Ruizhe An, Jinyuan Liu
Abstract: Images captured in harsh environments often exhibit blurred details, reduced contrast, and color distortion, which hinder feature detection and matching, thereby affecting the accuracy and robustness of homography estimation. While visual enhancement can improve contrast and clarity, it may introduce visual-tolerant artifacts that obscure the structural integrity of images. Considering the resilience of semantic information against environmental interference, we propose a semantic-driven feature enhancement network for robust homography estimation, dubbed SeFENet. Concretely, we first introduce an innovative hierarchical scale-aware module to expand the receptive field by aggregating multi-scale information, thereby effectively extracting image features under diverse harsh conditions. Subsequently, we propose a semantic-guided constraint module combined with a high-level perceptual framework to achieve degradation-tolerant with semantic feature. A meta-learning-based training strategy is introduced to mitigate the disparity between semantic and structural features. By internal-external alternating optimization, the proposed network achieves implicit semantic-wise feature enhancement, thereby improving the robustness of homography estimation in adverse environments by strengthening the local feature comprehension and context information extraction. Experimental results under both normal and harsh conditions demonstrate that SeFENet significantly outperforms SOTA methods, reducing point match error by at least 41\% on the large-scale datasets.
Authors: Jonathan Ganz, Jonas Ammeling, Emely Rosbach, Ludwig Lausser, Christof A. Bertram, Katharina Breininger, Marc Aubreville
Abstract: Foundation models (FMs), i.e., models trained on a vast amount of typically unlabeled data, have become popular and available recently for the domain of histopathology. The key idea is to extract semantically rich vectors from any input patch, allowing for the use of simple subsequent classification networks potentially reducing the required amounts of labeled data, and increasing domain robustness. In this work, we investigate to which degree this also holds for mitotic figure classification. Utilizing two popular public mitotic figure datasets, we compared linear probing of five publicly available FMs against models trained on ImageNet and a simple ResNet50 end-to-end-trained baseline. We found that the end-to-end-trained baseline outperformed all FM-based classifiers, regardless of the amount of data provided. Additionally, we did not observe the FM-based classifiers to be more robust against domain shifts, rendering both of the above assumptions incorrect.
Authors: Ori Matityahu, Raanan Fattal
Abstract: In this paper we identify the source of a singularity in the training loss of key denoising models, that causes the denoiser's predictions to collapse towards the mean of the source or target distributions. This degeneracy creates false basins of attraction, distorting the denoising trajectories and ultimately increasing the number of steps required to sample these models. We circumvent this artifact by leveraging the deterministic ODE-based samplers, offered by certain denoising diffusion and score-matching models, which establish a well-defined change-of-variables between the source and target distributions. Given this correspondence, we propose a new probability flow model, the Lines Matching Model (LMM), which matches globally straight lines interpolating the two distributions. We demonstrate that the flow fields produced by the LMM exhibit notable temporal consistency, resulting in trajectories with excellent straightness scores. Beyond its sampling efficiency, the LMM formulation allows us to enhance the fidelity of the generated samples by integrating domain-specific reconstruction and adversarial losses, and by optimizing its training for the sampling procedure used. Overall, the LMM achieves state-of-the-art FID scores with minimal NFEs on established benchmark datasets: 1.57/1.39 (NFE=1/2) on CIFAR-10, 1.47/1.17 on ImageNet 64x64, and 2.68/1.54 on AFHQ 64x64. Finally, we provide a theoretical analysis showing that the use of optimal transport to relate the two distributions suffers from a curse of dimensionality, where the pairing set size (mini-batch) must scale exponentially with the signal dimension.
Authors: Yu Zhong, Rui Zhang, Zihao Zhang, Shuo Wang, Chuan Fang, Xishan Zhang, Jiaming Guo, Shaohui Peng, Di Huang, Yanyang Yan, Xing Hu, Ping Tan, Qi Guo
Abstract: Vision-and-Language Navigation (VLN) is a challenging task that requires an agent to navigate through photorealistic environments following natural-language instructions. One main obstacle existing in VLN is data scarcity, leading to poor generalization performance over unseen environments. Tough data argumentation is a promising way for scaling up the dataset, how to generate VLN data both diverse and world-consistent remains problematic. To cope with this issue, we propose the world-consistent data generation (WCGEN), an efficacious data-augmentation framework satisfying both diversity and world-consistency, targeting at enhancing the generalizations of agents to novel environments. Roughly, our framework consists of two stages, the trajectory stage which leverages a point-cloud based technique to ensure spatial coherency among viewpoints, and the viewpoint stage which adopts a novel angle synthesis method to guarantee spatial and wraparound consistency within the entire observation. By accurately predicting viewpoint changes with 3D knowledge, our approach maintains the world-consistency during the generation procedure. Experiments on a wide range of datasets verify the effectiveness of our method, demonstrating that our data augmentation strategy enables agents to achieve new state-of-the-art results on all navigation tasks, and is capable of enhancing the VLN agents' generalization ability to unseen environments.
Authors: Jinglong Yang, Yichen Wu, Jun Cen, Wenjian Huang, Hong Wang, Jianguo Zhang
Abstract: Although the current different types of SAM adaptation methods have achieved promising performance for various downstream tasks, such as prompt-based ones and adapter-based ones, most of them belong to the one-step adaptation paradigm. In real-world scenarios, we are generally confronted with the dynamic scenario where the data comes in a streaming manner. Driven by the practical need, in this paper, we first propose a novel Continual SAM adaptation (CoSAM) benchmark with 8 different task domains and carefully analyze the limitations of the existing SAM one-step adaptation methods in the continual segmentation scenario. Then we propose a novel simple-yet-effective Mixture of Domain Adapters (MoDA) algorithm which utilizes the Global Feature Tokens (GFT) and Global Assistant Tokens (GAT) modules to help the SAM encoder extract well-separated features for different task domains, and then provide the accurate task-specific information for continual learning. Extensive experiments demonstrate that our proposed MoDA obviously surpasses the existing classic continual learning methods, as well as prompt-based and adapter-based approaches for continual segmentation. Moreover, after sequential learning on the CoSAM benchmark with diverse data distributions, our MoDA maintains highly competitive results in the natural image domain, approaching the zero-shot performance of the original SAM, demonstrating its superior capability in knowledge preservation. Notably, the proposed MoDA can be seamlessly integrated into various one-step adaptation methods of SAM, which can consistently bring obvious performance gains. Code is available at \url{https://github.com/yangjl1215/CoSAM}
Authors: Renlong Wu, Zhilu Zhang, Mingyang Chen, Xiaopeng Fan, Zifei Yan, Wangmeng Zuo
Abstract: Recent 4D reconstruction methods have yielded impressive results but rely on sharp videos as supervision. However, motion blur often occurs in videos due to camera shake and object movement, while existing methods render blurry results when using such videos for reconstructing 4D models. Although a few NeRF-based approaches attempted to address the problem, they struggled to produce high-quality results, due to the inaccuracy in estimating continuous dynamic representations within the exposure time. Encouraged by recent works in 3D motion trajectory modeling using 3D Gaussian Splatting (3DGS), we suggest taking 3DGS as the scene representation manner, and propose the first 4D Gaussian Splatting framework to reconstruct a high-quality 4D model from blurry monocular video, named Deblur4DGS. Specifically, we transform continuous dynamic representations estimation within an exposure time into the exposure time estimation. Moreover, we introduce exposure regularization to avoid trivial solutions, as well as multi-frame and multi-resolution consistency ones to alleviate artifacts. Furthermore, to better represent objects with large motion, we suggest blur-aware variable canonical Gaussians. Beyond novel-view synthesis, Deblur4DGS can be applied to improve blurry video from multiple perspectives, including deblurring, frame interpolation, and video stabilization. Extensive experiments on the above four tasks show that Deblur4DGS outperforms state-of-the-art 4D reconstruction methods. The codes are available at https://github.com/ZcsrenlongZ/Deblur4DGS.
Authors: Alexander Gielisse, Nergis T\"omen, Jan van Gemert
Abstract: Most recent works on optical flow use convex upsampling as the last step to obtain high-resolution flow. In this work, we show and discuss several issues and limitations of this currently widely adopted convex upsampling approach. We propose a series of changes, in an attempt to resolve current issues. First, we propose to decouple the weights for the final convex upsampler, making it easier to find the correct convex combination. For the same reason, we also provide extra contextual features to the convex upsampler. Then, we increase the convex mask size by using an attention-based alternative convex upsampler; Transformers for Convex Upsampling. This upsampler is based on the observation that convex upsampling can be reformulated as attention, and we propose to use local attention masks as a drop-in replacement for convex masks to increase the mask size. We provide empirical evidence that a larger mask size increases the likelihood of the existence of the convex combination. Lastly, we propose an alternative training scheme to remove bilinear interpolation artifacts from the model output. Our proposed ideas could theoretically be applied to almost every current state-of-the-art optical flow architecture. On the FlyingChairs + FlyingThings3D training setting we reduce the Sintel Clean training end-point-error of RAFT from 1.42 to 1.26, GMA from 1.31 to 1.18, and that of FlowFormer from 0.94 to 0.90, by solely adapting the convex upsampler.
Authors: Francis Xiatian Zhang, Jingjing Deng, Robert Lieck, Hubert P. H. Shum
Abstract: Surgical workflow anticipation is the task of predicting the timing of relevant surgical events from live video data, which is critical in Robotic-Assisted Surgery (RAS). Accurate predictions require the use of spatial information to model surgical interactions. However, current methods focus solely on surgical instruments, assume static interactions between instruments, and only anticipate surgical events within a fixed time horizon. To address these challenges, we propose an adaptive graph learning framework for surgical workflow anticipation based on a novel spatial representation, featuring three key innovations. First, we introduce a new representation of spatial information based on bounding boxes of surgical instruments and targets, including their detection confidence levels. These are trained on additional annotations we provide for two benchmark datasets. Second, we design an adaptive graph learning method to capture dynamic interactions. Third, we develop a multi-horizon objective that balances learning objectives for different time horizons, allowing for unconstrained predictions. Evaluations on two benchmarks reveal superior performance in short-to-mid-term anticipation, with an error reduction of approximately 3% for surgical phase anticipation and 9% for remaining surgical duration anticipation. These performance improvements demonstrate the effectiveness of our method and highlight its potential for enhancing preparation and coordination within the RAS team. This can improve surgical safety and the efficiency of operating room usage.
Authors: Wei Suo, Ji Ma, Mengyang Sun, Lin Yuanbo Wu, Peng Wang, Yanning Zhang
Abstract: Although Large Vision-Language Models (LVLMs) have achieved impressive results, their high computational cost poses a significant barrier to wider application. To enhance inference efficiency, most existing approaches depend on parameter-dependent or token-dependent strategies to reduce computational demands. However, these methods typically require complex training processes and struggle to consistently select the most relevant tokens. In this paper, we systematically analyze the above challenges and provide a series of valuable insights for inference acceleration. Based on these findings, we propose a novel framework, the Pruning All-Rounder (PAR). Different from previous works, PAR develops a meta-router to adaptively organize pruning flows across both tokens and layers. With a self-supervised learning manner, our method achieves a superior balance between performance and efficiency. Notably, PAR is highly flexible, offering multiple pruning versions to address a range of pruning scenarios. The code for this work will be made publicly available.
Authors: Weijie Tu, Weijian Deng, Dylan Campbell, Yu Yao, Jiyang Zheng, Tom Gedeon, Tongliang Liu
Abstract: As large multimodal models (LMMs) are increasingly deployed across diverse applications, the need for adaptable, real-world model ranking has become paramount. Traditional evaluation methods are largely dataset-centric, relying on fixed, labeled datasets and supervised metrics, which are resource-intensive and may lack generalizability to novel scenarios, highlighting the importance of unsupervised ranking. In this work, we explore unsupervised model ranking for LMMs by leveraging their uncertainty signals, such as softmax probabilities. We evaluate state-of-the-art LMMs (e.g., LLaVA) across visual question answering benchmarks, analyzing how uncertainty-based metrics can reflect model performance. Our findings show that uncertainty scores derived from softmax distributions provide a robust, consistent basis for ranking models across varied tasks. This finding enables the ranking of LMMs on real-world, unlabeled data for visual question answering, providing a practical approach for selecting models across diverse domains without requiring manual annotation.
Authors: Xuesong Zhang, Yunbo Xu, Jia Li, Zhenzhen Hu, Richnag Hong
Abstract: Navigating unseen environments based on natural language instructions remains difficult for egocentric agents in Vision-and-Language Navigation (VLN). While recent advancements have yielded promising outcomes, they primarily rely on RGB images for environmental representation, often overlooking the underlying semantic knowledge and spatial cues. Intuitively, humans inherently ground textual semantics within the spatial layout during indoor navigation. Inspired by this, we propose a versatile Semantic Understanding and Spatial Awareness (SUSA) architecture to facilitate navigation. SUSA includes a Textual Semantic Understanding (TSU) module, which narrows the modality gap between instructions and environments by generating and associating the descriptions of environmental landmarks in the agent's immediate surroundings. Additionally, a Depth-based Spatial Perception (DSP) module incrementally constructs a depth exploration map, enabling a more nuanced comprehension of environmental layouts. Experimental results demonstrate that SUSA hybrid semantic-spatial representations effectively enhance navigation performance, setting new state-of-the-art performance across three VLN benchmarks (REVERIE, R2R, and SOON). The source code will be publicly available.
Authors: Fei Wu, Pablo Marquez-Neila, Hedyeh Rafi-Tarii, Raphael Sznitman
Abstract: Multi-class semantic segmentation remains a cornerstone challenge in computer vision. Yet, dataset creation remains excessively demanding in time and effort, especially for specialized domains. Active Learning (AL) mitigates this challenge by selecting data points for annotation strategically. However, existing patch-based AL methods often overlook boundary pixels critical information, essential for accurate segmentation. We present OREAL, a novel patch-based AL method designed for multi-class semantic segmentation. OREAL enhances boundary detection by employing maximum aggregation of pixel-wise uncertainty scores. Additionally, we introduce one-vs-rest entropy, a novel uncertainty score function that computes class-wise uncertainties while achieving implicit class balancing during dataset creation. Comprehensive experiments across diverse datasets and model architectures validate our hypothesis.
Authors: Yixiong Fang, Ziran Yang, Zhaorun Chen, Zhuokai Zhao, Jiawei Zhou
Abstract: Large vision-language models (LVLMs) demonstrate remarkable capabilities in multimodal tasks but are prone to misinterpreting visual inputs, often resulting in hallucinations and unreliable outputs. To address these challenges, we propose Dropout Decoding, a novel inference-time approach that quantifies the uncertainty of visual tokens and selectively masks uncertain tokens to improve decoding. Our method measures the uncertainty of each visual token by projecting it onto the text space and decomposing it into aleatoric and epistemic components. Specifically, we focus on epistemic uncertainty, which captures perception-related errors more effectively. Inspired by dropout regularization, we introduce uncertainty-guided token dropout, which applies the dropout principle to input visual tokens instead of model parameters, and during inference rather than training. By aggregating predictions from an ensemble of masked decoding contexts, Dropout Decoding robustly mitigates errors arising from visual token misinterpretations. Evaluations on benchmarks including CHAIR, THRONE, and MMBench demonstrate that Dropout Decoding significantly reduces object hallucinations (OH) and enhances both reliability and quality of LVLM outputs across diverse visual contexts.
Authors: Yihong Xu, Yuan Yin, Tuan-Hung Vu, Alexandre Boulch, \'Eloi Zablocki, Matthieu Cord
Abstract: Motion forecasting (MF) for autonomous driving aims at anticipating trajectories of surrounding agents in complex urban scenarios. In this work, we investigate a mixed strategy in MF training that first pre-train motion forecasters on pseudo-labeled data, then fine-tune them on annotated data. To obtain pseudo-labeled trajectories, we propose a simple pipeline that leverages off-the-shelf single-frame 3D object detectors and non-learning trackers. The whole pre-training strategy including pseudo-labeling is coined as PPT. Our extensive experiments demonstrate that: (1) combining PPT with supervised fine-tuning on annotated data achieves superior performance on diverse testbeds, especially under annotation-efficient regimes, (2) scaling up to multiple datasets improves the previous state-of-the-art and (3) PPT helps enhance cross-dataset generalization. Our findings showcase PPT as a promising pre-training solution for robust motion forecasting in diverse autonomous driving contexts.
Authors: Xiaoqian Zhou, Zhen Huang, Heqin Zhu, Qingsong Yao, S. Kevin Zhou
Abstract: Accurate anatomical landmark detection in medical images is crucial for clinical applications. Existing methods often struggle to balance global context with computational efficiency, particularly with high-resolution images. This paper introduces the Hybrid Attention Network(HAN), a novel hybrid architecture integrating CNNs and Transformers. Its core is the BiFormer module, utilizing Bi-Level Routing Attention (BRA) for efficient attention to relevant image regions. This, combined with Convolutional Attention Blocks (CAB) enhanced by CBAM, enables precise local feature refinement guided by the global context. A Feature Fusion Correction Module (FFCM) integrates multi-scale features, mitigating resolution loss. Deep supervision with MSE loss on multi-resolution heatmaps optimizes the model. Experiments on five diverse datasets demonstrate state-of-the-art performance, surpassing existing methods in accuracy, robustness, and efficiency. The HAN provides a promising solution for accurate and efficient anatomical landmark detection in complex medical images. Our codes and data will be released soon at: \url{https://github.com/MIRACLE-Center/}.
Authors: Shidan He, Lei Liu, Shen Zhao
Abstract: Anomaly synthesis is a crucial approach to augment abnormal data for advancing anomaly inspection. Based on the knowledge from the large-scale pre-training, existing text-to-image anomaly synthesis methods predominantly focus on textual information or coarse-aligned visual features to guide the entire generation process. However, these methods often lack sufficient descriptors to capture the complicated characteristics of realistic anomalies (e.g., the fine-grained visual pattern of anomalies), limiting the realism and generalization of the generation process. To this end, we propose a novel anomaly synthesis framework called AnomalyControl to learn cross-modal semantic features as guidance signals, which could encode the generalized anomaly cues from text-image reference prompts and improve the realism of synthesized abnormal samples. Specifically, AnomalyControl adopts a flexible and non-matching prompt pair (i.e., a text-image reference prompt and a targeted text prompt), where a Cross-modal Semantic Modeling (CSM) module is designed to extract cross-modal semantic features from the textual and visual descriptors. Then, an Anomaly-Semantic Enhanced Attention (ASEA) mechanism is formulated to allow CSM to focus on the specific visual patterns of the anomaly, thus enhancing the realism and contextual relevance of the generated anomaly features. Treating cross-modal semantic features as the prior, a Semantic Guided Adapter (SGA) is designed to encode effective guidance signals for the adequate and controllable synthesis process. Extensive experiments indicate that AnomalyControl can achieve state-of-the-art results in anomaly synthesis compared with existing methods while exhibiting superior performance for downstream tasks.
Authors: Pascal Clausen, Li Ma, Mingming He, Ahmet Levent Tasel, Oliver Pilarski, Paul Debevec
Abstract: We present a technique for fitting high dynamic range illumination (HDRI) sequences using anisotropic spherical Gaussians (ASGs) while preserving temporal consistency in the compressed HDRI maps. Our approach begins with an optimization network that iteratively minimizes a composite loss function, which includes both reconstruction and diffuse losses. This allows us to represent all-frequency signals with a small number of ASGs, optimizing their directions, sharpness, and intensity simultaneously for an individual HDRI. To extend this optimization into the temporal domain, we introduce a temporal consistency loss, ensuring a consistent approximation across the entire HDRI sequence.
Authors: Jan Rathjens, Shirin Reyhanian, David Kappel, Laurenz Wiskott
Abstract: Understanding the mechanisms underlying deep neural networks in computer vision remains a fundamental challenge. While many prior approaches have focused on visualizing intermediate representations within deep neural networks, particularly convolutional neural networks, these techniques have yet to be thoroughly explored in transformer-based vision models. In this study, we apply the approach of training inverse models to reconstruct input images from intermediate layers within a Detection Transformer, showing that this approach is efficient and feasible for transformer-based vision models. Through qualitative and quantitative evaluations of reconstructed images across model stages, we demonstrate critical properties of Detection Transformers, including contextual shape preservation, inter-layer correlation, and robustness to color perturbations, illustrating how these characteristics emerge within the model's architecture. Our findings contribute to a deeper understanding of transformer-based vision models. The code for reproducing our experiments will be made available at github.com/wiskott-lab/inverse-detection-transformer.
Authors: Melo Castillo Angie Nataly, Martin Serrano Sergio, Salinas Carlota, Sotelo Miguel Angel
Abstract: Pedestrian detection is a critical task in autonomous driving, aimed at enhancing safety and reducing risks on the road. Over recent years, significant advancements have been made in improving detection performance. However, these achievements still fall short of human perception, particularly in cases involving occluded pedestrians, especially entirely invisible ones. In this work, we present the Occlusion-Rich Road Scenes with Pedestrians (OccluRoads) dataset, which features a diverse collection of road scenes with partially and fully occluded pedestrians in both real and virtual environments. All scenes are meticulously labeled and enriched with contextual information that encapsulates human perception in such scenarios. Using this dataset, we developed a pipeline to predict the presence of occluded pedestrians, leveraging Knowledge Graph (KG), Knowledge Graph Embedding (KGE), and a Bayesian inference process. Our approach achieves a F1 score of 0.91, representing an improvement of up to 42% compared to traditional machine learning models.
Authors: Adil Karjauv, Noor Fathima, Ioannis Lelekas, Fatih Porikli, Amir Ghodrati, Amirhossein Habibian
Abstract: Recent progress in diffusion-based video editing has shown remarkable potential for practical applications. However, these methods remain prohibitively expensive and challenging to deploy on mobile devices. In this study, we introduce a series of optimizations that render mobile video editing feasible. Building upon the existing image editing model, we first optimize its architecture and incorporate a lightweight autoencoder. Subsequently, we extend classifier-free guidance distillation to multiple modalities, resulting in a threefold on-device speedup. Finally, we reduce the number of sampling steps to one by introducing a novel adversarial distillation scheme which preserves the controllability of the editing process. Collectively, these optimizations enable video editing at 12 frames per second on mobile devices, while maintaining high quality. Our results are available at https://qualcomm-ai-research.github.io/mobile-video-editing/
URLs: https://qualcomm-ai-research.github.io/mobile-video-editing/
Authors: Dongchen Han, Yifan Pu, Zhuofan Xia, Yizeng Han, Xuran Pan, Xiu Li, Jiwen Lu, Shiji Song, Gao Huang
Abstract: Widely adopted in modern Vision Transformer designs, Softmax attention can effectively capture long-range visual information; however, it incurs excessive computational cost when dealing with high-resolution inputs. In contrast, linear attention naturally enjoys linear complexity and has great potential to scale up to higher-resolution images. Nonetheless, the unsatisfactory performance of linear attention greatly limits its practical application in various scenarios. In this paper, we take a step forward to close the gap between the linear and Softmax attention with novel theoretical analyses, which demystify the core factors behind the performance deviations. Specifically, we present two key perspectives to understand and alleviate the limitations of linear attention: the injective property and the local modeling ability. Firstly, we prove that linear attention is not injective, which is prone to assign identical attention weights to different query vectors, thus adding to severe semantic confusion since different queries correspond to the same outputs. Secondly, we confirm that effective local modeling is essential for the success of Softmax attention, in which linear attention falls short. The aforementioned two fundamental differences significantly contribute to the disparities between these two attention paradigms, which is demonstrated by our substantial empirical validation in the paper. In addition, more experiment results indicate that linear attention, as long as endowed with these two properties, can outperform Softmax attention across various tasks while maintaining lower computation complexity. Code is available at https://github.com/LeapLabTHU/InLine.
Authors: Ziya Erko\c{c}, Can G\"umeli, Chaoyang Wang, Matthias Nie{\ss}ner, Angela Dai, Peter Wonka, Hsin-Ying Lee, Peiye Zhuang
Abstract: We propose a training-free approach to 3D editing that enables the editing of a single shape within a few minutes. The edited 3D mesh aligns well with the prompts, and remains identical for regions that are not intended to be altered. To this end, we first project the 3D object onto 4-view images and perform synchronized multi-view image editing along with user-guided text prompts and user-provided rough masks. However, the targeted regions to be edited are ambiguous due to projection from 3D to 2D. To ensure precise editing only in intended regions, we develop a 3D segmentation pipeline that detects edited areas in 3D space, followed by a merging algorithm to seamlessly integrate edited 3D regions with the original input. Extensive experiments demonstrate the superiority of our method over previous approaches, enabling fast, high-quality editing while preserving unintended regions.
Authors: Chun-Peng Chang, Alain Pagani, Didier Stricker
Abstract: Multimodal Large Language Models (MLLMs) have made significant progress in tasks such as image captioning and question answering. However, while these models can generate realistic captions, they often struggle with providing precise instructions, particularly when it comes to localizing and disambiguating objects in complex 3D environments. This capability is critical as MLLMs become more integrated with collaborative robotic systems. In scenarios where a target object is surrounded by similar objects (distractors), robots must deliver clear, spatially-aware instructions to guide humans effectively. We refer to this challenge as contextual object localization and disambiguation, which imposes stricter constraints than conventional 3D dense captioning, especially regarding ensuring target exclusivity. In response, we propose simple yet effective techniques to enhance the model's ability to localize and disambiguate target objects. Our approach not only achieves state-of-the-art performance on conventional metrics that evaluate sentence similarity, but also demonstrates improved 3D spatial understanding through 3D visual grounding model.
Authors: Weitao Wang, Haoran Xu, Yuxiao Yang, Zhifang Liu, Jun Meng, Haoqian Wang
Abstract: Recent years have witnessed remarkable progress in 3D content generation. However, corresponding evaluation methods struggle to keep pace. Automatic approaches have proven challenging to align with human preferences, and the mixed comparison of text- and image-driven methods often leads to unfair evaluations. In this paper, we present a comprehensive framework to better align and evaluate multi-view diffusion models with human preferences. To begin with, we first collect and filter a standardized image prompt set from DALL$\cdot$E and Objaverse, which we then use to generate multi-view assets with several multi-view diffusion models. Through a systematic ranking pipeline on these assets, we obtain a human annotation dataset with 16k expert pairwise comparisons and train a reward model, coined MVReward, to effectively encode human preferences. With MVReward, image-driven 3D methods can be evaluated against each other in a more fair and transparent manner. Building on this, we further propose Multi-View Preference Learning (MVP), a plug-and-play multi-view diffusion tuning strategy. Extensive experiments demonstrate that MVReward can serve as a reliable metric and MVP consistently enhances the alignment of multi-view diffusion models with human preferences.
Authors: Ioannis Sarridis, Christos Koutlis, Symeon Papadopoulos, Christos Diou
Abstract: Mitigating biases in computer vision models is an essential step towards the trustworthiness of artificial intelligence models. Existing bias mitigation methods focus on a small set of predefined biases, limiting their applicability in visual datasets where multiple, possibly unknown biases exist. To address this limitation, we introduce MAVias, an open-set bias mitigation approach leveraging foundation models to discover spurious associations between visual attributes and target classes. MAVias first captures a wide variety of visual features in natural language via a foundation image tagging model, and then leverages a large language model to select those visual features defining the target class, resulting in a set of language-coded potential visual biases. We then translate this set of potential biases into vision-language embeddings and introduce an in-processing bias mitigation approach to prevent the model from encoding information related to them. Our experiments on diverse datasets, including CelebA, Waterbirds, ImageNet, and UrbanCars, show that MAVias effectively detects and mitigates a wide range of biases in visual recognition tasks outperforming current state-of-the-art.
Authors: Johanna Vielhaben, Dilyara Bareeva, Jim Berend, Wojciech Samek, Nils Strodthoff
Abstract: Vision transformers (ViTs) can be trained using various learning paradigms, from fully supervised to self-supervised. Diverse training protocols often result in significantly different feature spaces, which are usually compared through alignment analysis. However, current alignment measures quantify this relationship in terms of a single scalar value, obscuring the distinctions between common and unique features in pairs of representations that share the same scalar alignment. We address this limitation by combining alignment analysis with concept discovery, which enables a breakdown of alignment into single concepts encoded in feature space. This fine-grained comparison reveals both universal and unique concepts across different representations, as well as the internal structure of concepts within each of them. Our methodological contributions address two key prerequisites for concept-based alignment: 1) For a description of the representation in terms of concepts that faithfully capture the geometry of the feature space, we define concepts as the most general structure they can possibly form - arbitrary manifolds, allowing hidden features to be described by their proximity to these manifolds. 2) To measure distances between concept proximity scores of two representations, we use a generalized Rand index and partition it for alignment between pairs of concepts. We confirm the superiority of our novel concept definition for alignment analysis over existing linear baselines in a sanity check. The concept-based alignment analysis of representations from four different ViTs reveals that increased supervision correlates with a reduction in the semantic structure of learned representations.
Authors: Zitong Huang, Ze Chen, Yuanze Li, Bowen Dong, Erjin Zhou, Yong Liu, Rick Siow Mong Goh, Chun-Mei Feng, Wangmeng Zuo
Abstract: Few-Shot Class-Incremental Learning has shown remarkable efficacy in efficient learning new concepts with limited annotations. Nevertheless, the heuristic few-shot annotations may not always cover the most informative samples, which largely restricts the capability of incremental learner. We aim to start from a pool of large-scale unlabeled data and then annotate the most informative samples for incremental learning. Based on this premise, this paper introduces the Active Class-Incremental Learning (ACIL). The objective of ACIL is to select the most informative samples from the unlabeled pool to effectively train an incremental learner, aiming to maximize the performance of the resulting model. Note that vanilla active learning algorithms suffer from class-imbalanced distribution among annotated samples, which restricts the ability of incremental learning. To achieve both class balance and informativeness in chosen samples, we propose Class-Balanced Selection (CBS) strategy. Specifically, we first cluster the features of all unlabeled images into multiple groups. Then for each cluster, we employ greedy selection strategy to ensure that the Gaussian distribution of the sampled features closely matches the Gaussian distribution of all unlabeled features within the cluster. Our CBS can be plugged and played into those CIL methods which are based on pretrained models with prompts tunning technique. Extensive experiments under ACIL protocol across five diverse datasets demonstrate that CBS outperforms both random selection and other SOTA active learning approaches. Code is publicly available at https://github.com/1170300714/CBS.
Authors: Alejandro Marco Montejano, Angela Sanchez Perez, Javier Barrachina, David Ortiz-Perez, Manuel Benavent-Lledo, Jose Garcia-Rodriguez
Abstract: The rapid evolution of digital image manipulation techniques poses significant challenges for content verification, with models such as stable diffusion and mid-journey producing highly realistic, yet synthetic, images that can deceive human perception. This research develops and evaluates convolutional neural networks (CNNs) specifically tailored for the detection of these manipulated images. The study implements a comparative analysis of three progressively complex CNN architectures, assessing their ability to classify and localize manipulations across various facial image modifications. Regularization and optimization techniques were systematically incorporated to improve feature extraction and performance. The results indicate that the proposed models achieve an accuracy of up to 76\% in distinguishing manipulated images from genuine ones, surpassing traditional approaches. This research not only highlights the potential of CNNs in enhancing the robustness of digital media verification tools, but also provides insights into effective architectural adaptations and training strategies for low-computation environments. Future work will build on these findings by extending the architectures to handle more diverse manipulation techniques and integrating multi-modal data for improved detection capabilities.
Authors: Alessandro Serra, Francesco Ortu, Emanuele Panizon, Lucrezia Valeriani, Lorenzo Basile, Alessio Ansuini, Diego Doimo, Alberto Cazzaniga
Abstract: Recent advances in multimodal training have significantly improved the integration of image understanding and generation within a unified model. This study investigates how vision-language models (VLMs) handle image-understanding tasks, specifically focusing on how visual information is processed and transferred to the textual domain. We compare VLMs that generate both images and text with those that output only text, highlighting key differences in information flow. We find that in models with multimodal outputs, image and text embeddings are more separated within the residual stream. Additionally, models vary in how information is exchanged from visual to textual tokens. VLMs that only output text exhibit a distributed communication pattern, where information is exchanged through multiple image tokens. In contrast, models trained for image and text generation rely on a single token that acts as a narrow gate for the visual information. We demonstrate that ablating this single token significantly deteriorates performance on image understanding tasks. Furthermore, modifying this token enables effective steering of the image semantics, showing that targeted, local interventions can reliably control the model's global behavior.
Authors: Xiao Wang, Yu Jin, Wentao Wu, Wei Zhang, Lin Zhu, Bo Jiang, Yonghong Tian
Abstract: Object detection in event streams has emerged as a cutting-edge research area, demonstrating superior performance in low-light conditions, scenarios with motion blur, and rapid movements. Current detectors leverage spiking neural networks, Transformers, or convolutional neural networks as their core architectures, each with its own set of limitations including restricted performance, high computational overhead, or limited local receptive fields. This paper introduces a novel MoE (Mixture of Experts) heat conduction-based object detection algorithm that strikingly balances accuracy and computational efficiency. Initially, we employ a stem network for event data embedding, followed by processing through our innovative MoE-HCO blocks. Each block integrates various expert modules to mimic heat conduction within event streams. Subsequently, an IoU-based query selection module is utilized for efficient token extraction, which is then channeled into a detection head for the final object detection process. Furthermore, we are pleased to introduce EvDET200K, a novel benchmark dataset for event-based object detection. Captured with a high-definition Prophesee EVK4-HD event camera, this dataset encompasses 10 distinct categories, 200,000 bounding boxes, and 10,054 samples, each spanning 2 to 5 seconds. We also provide comprehensive results from over 15 state-of-the-art detectors, offering a solid foundation for future research and comparison. The source code of this paper will be released on: https://github.com/Event-AHU/OpenEvDET
Authors: Shuaiting Li, Juncan Deng, Zeyu Wang, Hong Gu, Kedong Xu, Haibin Shen, Kejie Huang
Abstract: Text-to-image generation of Stable Diffusion models has achieved notable success due to its remarkable generation ability. However, the repetitive denoising process is computationally intensive during inference, which renders Diffusion models less suitable for real-world applications that require low latency and scalability. Recent studies have employed post-training quantization (PTQ) and quantization-aware training (QAT) methods to compress Diffusion models. Nevertheless, prior research has often neglected to examine the consistency between results generated by quantized models and those from floating-point models. This consistency is crucial in fields such as content creation, design, and edge deployment, as it can significantly enhance both efficiency and system stability for practitioners. To ensure that quantized models generate high-quality and consistent images, we propose an efficient quantization framework for Stable Diffusion models. Our approach features a Serial-to-Parallel calibration pipeline that addresses the consistency of both the calibration and inference processes, as well as ensuring training stability. Based on this pipeline, we further introduce a mix-precision quantization strategy, multi-timestep activation quantization, and time information precalculation techniques to ensure high-fidelity generation in comparison to floating-point models. Through extensive experiments with Stable Diffusion v1-4, v2-1, and XL 1.0, we have demonstrated that our method outperforms the current state-of-the-art techniques when tested on prompts from the COCO validation dataset and the Stable-Diffusion-Prompts dataset. Under W4A8 quantization settings, our approach enhances both distribution similarity and visual similarity by 45%-60%.
Authors: Shun Zhang, Xuechao Zou, Kai Li, Congyan Lang, Shiying Wang, Pin Tao, Tengfei Cao
Abstract: Fine-grained remote sensing image segmentation is essential for accurately identifying detailed objects in remote sensing images. Recently, vision transformer models (VTM) pretrained on large-scale datasets have shown strong zero-shot generalization, indicating that they have learned the general knowledge of object understanding. We introduce a novel end-to-end learning paradigm combining knowledge guidance with domain refinement to enhance performance. We present two key components: the Feature Alignment Module (FAM) and the Feature Modulation Module (FMM). FAM aligns features from a CNN-based backbone with those from the pretrained VTM's encoder using channel transformation and spatial interpolation, and transfers knowledge via KL divergence and L2 normalization constraint. FMM further adapts the knowledge to the specific domain to address domain shift. We also introduce a fine-grained grass segmentation dataset and demonstrate, through experiments on two datasets, that our method achieves a significant improvement of 2.57 mIoU on the grass dataset and 3.73 mIoU on the cloud dataset. The results highlight the potential of combining knowledge transfer and domain adaptation to overcome domain-related challenges and data limitations. The project page is available at https://xavierjiezou.github.io/KTDA/.
Authors: Chunwei Wang, Guansong Lu, Junwei Yang, Runhui Huang, Jianhua Han, Lu Hou, Wei Zhang, Hang Xu
Abstract: In this paper, we introduce ILLUME, a unified multimodal large language model (MLLM) that seamlessly integrates multimodal understanding and generation capabilities within a single large language model through a unified next-token prediction formulation. To address the large dataset size typically required for image-text alignment, we propose to enhance data efficiency through the design of a vision tokenizer that incorporates semantic information and a progressive multi-stage training procedure. This approach reduces the dataset size to just 15M for pretraining -- over four times fewer than what is typically needed -- while achieving competitive or even superior performance with existing unified MLLMs, such as Janus. Additionally, to promote synergistic enhancement between understanding and generation capabilities, which is under-explored in previous works, we introduce a novel self-enhancing multimodal alignment scheme. This scheme supervises the MLLM to self-assess the consistency between text descriptions and self-generated images, facilitating the model to interpret images more accurately and avoid unrealistic and incorrect predictions caused by misalignment in image generation. Based on extensive experiments, our proposed ILLUME stands out and competes with state-of-the-art unified MLLMs and specialized models across various benchmarks for multimodal understanding, generation, and editing.
Authors: Jiangning Zhang, Teng Hu, Haoyang He, Zhucun Xue, Yabiao Wang, Chengjie Wang, Yong Liu, Xiangtai Li, Dacheng Tao
Abstract: This work focuses on developing parameter-efficient and lightweight models for dense predictions while trading off parameters, FLOPs, and performance. Our goal is to set up the new frontier of the 5M magnitude lightweight model on various downstream tasks. Inverted Residual Block (IRB) serves as the infrastructure for lightweight CNNs, but no counterparts have been recognized by attention-based design. Our work rethinks the lightweight infrastructure of efficient IRB and practical components in Transformer from a unified perspective, extending CNN-based IRB to attention-based models and abstracting a one-residual Meta Mobile Block (MMBlock) for lightweight model design. Following neat but effective design criterion, we deduce a modern Improved Inverted Residual Mobile Block (i2RMB) and improve a hierarchical Efficient MOdel (EMOv2) with no elaborate complex structures. Considering the imperceptible latency for mobile users when downloading models under 4G/5G bandwidth and ensuring model performance, we investigate the performance upper limit of lightweight models with a magnitude of 5M. Extensive experiments on various vision recognition, dense prediction, and image generation tasks demonstrate the superiority of our EMOv2 over state-of-the-art methods, e.g., EMOv2-1M/2M/5M achieve 72.3, 75.8, and 79.4 Top-1 that surpass equal-order CNN-/Attention-based models significantly. At the same time, EMOv2-5M equipped RetinaNet achieves 41.5 mAP for object detection tasks that surpasses the previous EMO-5M by +2.6. When employing the more robust training recipe, our EMOv2-5M eventually achieves 82.9 Top-1 accuracy, which elevates the performance of 5M magnitude models to a new level. Code is available at https://github.com/zhangzjn/EMOv2.
Authors: Yuxuan Xue, Xianghui Xie, Riccardo Marin, Gerard Pons-Moll
Abstract: Creating realistic 3D objects and clothed avatars from a single RGB image is an attractive yet challenging problem. Due to its ill-posed nature, recent works leverage powerful prior from 2D diffusion models pretrained on large datasets. Although 2D diffusion models demonstrate strong generalization capability, they cannot guarantee the generated multi-view images are 3D consistent. In this paper, we propose Gen-3Diffusion: Realistic Image-to-3D Generation via 2D & 3D Diffusion Synergy. We leverage a pre-trained 2D diffusion model and a 3D diffusion model via our elegantly designed process that synchronizes two diffusion models at both training and sampling time. The synergy between the 2D and 3D diffusion models brings two major advantages: 1) 2D helps 3D in generalization: the pretrained 2D model has strong generalization ability to unseen images, providing strong shape priors for the 3D diffusion model; 2) 3D helps 2D in multi-view consistency: the 3D diffusion model enhances the 3D consistency of 2D multi-view sampling process, resulting in more accurate multi-view generation. We validate our idea through extensive experiments in image-based objects and clothed avatar generation tasks. Results show that our method generates realistic 3D objects and avatars with high-fidelity geometry and texture. Extensive ablations also validate our design choices and demonstrate the strong generalization ability to diverse clothing and compositional shapes. Our code and pretrained models will be publicly released on https://yuxuan-xue.com/gen-3diffusion.
Authors: Baorui Ma, Huachen Gao, Haoge Deng, Zhengxiong Luo, Tiejun Huang, Lulu Tang, Xinlong Wang
Abstract: Recent 3D generation models typically rely on limited-scale 3D `gold-labels' or 2D diffusion priors for 3D content creation. However, their performance is upper-bounded by constrained 3D priors due to the lack of scalable learning paradigms. In this work, we present See3D, a visual-conditional multi-view diffusion model trained on large-scale Internet videos for open-world 3D creation. The model aims to Get 3D knowledge by solely Seeing the visual contents from the vast and rapidly growing video data -- You See it, You Got it. To achieve this, we first scale up the training data using a proposed data curation pipeline that automatically filters out multi-view inconsistencies and insufficient observations from source videos. This results in a high-quality, richly diverse, large-scale dataset of multi-view images, termed WebVi3D, containing 320M frames from 16M video clips. Nevertheless, learning generic 3D priors from videos without explicit 3D geometry or camera pose annotations is nontrivial, and annotating poses for web-scale videos is prohibitively expensive. To eliminate the need for pose conditions, we introduce an innovative visual-condition - a purely 2D-inductive visual signal generated by adding time-dependent noise to the masked video data. Finally, we introduce a novel visual-conditional 3D generation framework by integrating See3D into a warping-based pipeline for high-fidelity 3D generation. Our numerical and visual comparisons on single and sparse reconstruction benchmarks show that See3D, trained on cost-effective and scalable video data, achieves notable zero-shot and open-world generation capabilities, markedly outperforming models trained on costly and constrained 3D datasets. Please refer to our project page at: https://vision.baai.ac.cn/see3d
Authors: Dongyue Lu, Lingdong Kong, Gim Hee Lee, Camille Simon Chane, Wei Tsang Ooi
Abstract: Event cameras offer unparalleled advantages for real-time perception in dynamic environments, thanks to their microsecond-level temporal resolution and asynchronous operation. Existing event-based object detection methods, however, are limited by fixed-frequency paradigms and fail to fully exploit the high-temporal resolution and adaptability of event cameras. To address these limitations, we propose FlexEvent, a novel event camera object detection framework that enables detection at arbitrary frequencies. Our approach consists of two key components: FlexFuser, an adaptive event-frame fusion module that integrates high-frequency event data with rich semantic information from RGB frames, and FAL, a frequency-adaptive learning mechanism that generates frequency-adjusted labels to enhance model generalization across varying operational frequencies. This combination allows our method to detect objects with high accuracy in both fast-moving and static scenarios, while adapting to dynamic environments. Extensive experiments on large-scale event camera datasets demonstrate that our approach surpasses state-of-the-art methods, achieving significant improvements in both standard and high-frequency settings. Notably, our method maintains robust performance when scaling from 20 Hz to 90 Hz and delivers accurate detection up to 180 Hz, proving its effectiveness in extreme conditions. Our framework sets a new benchmark for event-based object detection and paves the way for more adaptable, real-time vision systems.
Authors: Aqib Nazir Mir, Iqra Nissar, Mumtaz Ahmed, Sarfaraz Masood, Danish Raza Rizvi
Abstract: Deep learning holds tremendous potential in healthcare for uncovering hidden patterns within extensive clinical datasets, aiding in the diagnosis of various diseases. Parkinson's disease (PD) is a neurodegenerative condition characterized by the deterioration of brain function. In the initial stages of PD, automatic diagnosis poses a challenge due to the similarity in behavior between individuals with PD and those who are healthy. Our objective is to propose an effective model that can aid in the early detection of Parkinson's disease. We employed the VGRF gait signal dataset sourced from Physionet for distinguishing between healthy individuals and those diagnosed with Parkinson's disease. This paper introduces a novel deep learning architecture based on the LSTM network for automatically detecting freezing of gait episodes in Parkinson's disease patients. In contrast to conventional machine learning algorithms, this method eliminates manual feature engineering and proficiently captures prolonged temporal dependencies in gait patterns, thereby improving the diagnosis of Parkinson's disease. The LSTM network resolves the issue of vanishing gradients by employing memory blocks in place of self-connected hidden units, allowing for optimal information assimilation. To prevent overfitting, dropout and L2 regularization techniques have been employed. Additionally, the stochastic gradient-based optimizer Adam is used for the optimization process. The results indicate that our proposed approach surpasses current state-of-the-art models in FOG episode detection, achieving an accuracy of 97.71%, sensitivity of 99%, precision of 98%, and specificity of 96%. This demonstrates its potential as a superior classification method for Parkinson's disease detection.
Authors: Sahil Sethi, Sai Reddy, Mansi Sakarvadia, Jordan Serotte, Darlington Nwaudo, Nicholas Maassen, Lewis Shi
Abstract: Bankart lesions, or anterior-inferior glenoid labral tears, are diagnostically challenging on standard MRIs due to their subtle imaging features-often necessitating invasive MRI arthrograms (MRAs). This study develops deep learning (DL) models to detect Bankart lesions on both standard MRIs and MRAs, aiming to improve diagnostic accuracy and reduce reliance on MRAs. We curated a dataset of 586 shoulder MRIs (335 standard, 251 MRAs) from 558 patients who underwent arthroscopy. Ground truth labels were derived from intraoperative findings, the gold standard for Bankart lesion diagnosis. Separate DL models for MRAs and standard MRIs were trained using the Swin Transformer architecture, pre-trained on a public knee MRI dataset. Predictions from sagittal, axial, and coronal views were ensembled to optimize performance. The models were evaluated on a 20% hold-out test set (117 MRIs: 46 MRAs, 71 standard MRIs). Bankart lesions were identified in 31.9% of MRAs and 8.6% of standard MRIs. The models achieved AUCs of 0.87 (86% accuracy, 83% sensitivity, 86% specificity) and 0.90 (85% accuracy, 82% sensitivity, 86% specificity) on standard MRIs and MRAs, respectively. These results match or surpass radiologist performance on our dataset and reported literature metrics. Notably, our model's performance on non-invasive standard MRIs matched or surpassed the radiologists interpreting MRAs. This study demonstrates the feasibility of using DL to address the diagnostic challenges posed by subtle pathologies like Bankart lesions. Our models demonstrate potential to improve diagnostic confidence, reduce reliance on invasive imaging, and enhance accessibility to care.
Authors: Hongze Mi, Jinyuan Li, Xuying Zhang, Haoran Cheng, Jiahao Wang, Di Sun, Gang Pan
Abstract: Multimodal Entity Linking (MEL) is extensively utilized in the domains of information retrieval. However, existing MEL methods typically utilize mention words as mentions for retrieval. This results in a significant dependence of MEL on mention words, thereby constraining its capacity to effectively leverage information from both images and text. In situations where mention words are absent, MEL methods struggle to leverage image-text pairs for entity linking. To solve these issues, we introduce a Visual Prompts guided Multimodal Entity Linking (VP-MEL) task. VP-MEL directly marks specific regions within the image. These markers are referred to as visual prompts in VP-MEL. Without mention words, VP-MEL aims to utilize marked image-text pairs to align visual prompts with specific entities in the knowledge bases. A new dataset for the VP-MEL task, VPWiki, is proposed in this paper. Moreover, we propose a framework named FBMEL, which enhances the significance of visual prompts and fully leverages the information in image-text pairs. Experimental results on the VPWiki dataset demonstrate that FBMEL outperforms baseline methods across multiple benchmarks for the VP-MEL task.
Authors: Caiyun Xie, Dengpan Ye, Yunming Zhang, Long Tang, Yunna Lv, Jiacheng Deng, Jiawei Song
Abstract: The security of AI-generated content (AIGC) detection based on GANs and diffusion models is closely related to the credibility of multimedia content. Malicious adversarial attacks can evade these developing AIGC detection. However, most existing adversarial attacks focus only on GAN-generated facial images detection, struggle to be effective on multi-class natural images and diffusion-based detectors, and exhibit poor invisibility. To fill this gap, we first conduct an in-depth analysis of the vulnerability of AIGC detectors and discover the feature that detectors vary in vulnerability to different post-processing. Then, considering the uncertainty of detectors in real-world scenarios, and based on the discovery, we propose a Realistic-like Robust Black-box Adversarial attack (R$^2$BA) with post-processing fusion optimization. Unlike typical perturbations, R$^2$BA uses real-world post-processing, i.e., Gaussian blur, JPEG compression, Gaussian noise and light spot to generate adversarial examples. Specifically, we use a stochastic particle swarm algorithm with inertia decay to optimize post-processing fusion intensity and explore the detector's decision boundary. Guided by the detector's fake probability, R$^2$BA enhances/weakens the detector-vulnerable/detector-robust post-processing intensity to strike a balance between adversariality and invisibility. Extensive experiments on popular/commercial AIGC detectors and datasets demonstrate that R$^2$BA exhibits impressive anti-detection performance, excellent invisibility, and strong robustness in GAN-based and diffusion-based cases. Compared to state-of-the-art white-box and black-box attacks, R$^2$BA shows significant improvements of 15% and 21% in anti-detection performance under the original and robust scenario respectively, offering valuable insights for the security of AIGC detection in real-world applications.
Authors: Simone Azeglio, Olivier Marre, Peter Neri, Ulisse Ferrari
Abstract: We propose a novel approach to image classification inspired by complex nonlinear biological visual processing, whereby classical convolutional neural networks (CNNs) are equipped with learnable higher-order convolutions. Our model incorporates a Volterra-like expansion of the convolution operator, capturing multiplicative interactions akin to those observed in early and advanced stages of biological visual processing. We evaluated this approach on synthetic datasets by measuring sensitivity to testing higher-order correlations and performance in standard benchmarks (MNIST, FashionMNIST, CIFAR10, CIFAR100 and Imagenette). Our architecture outperforms traditional CNN baselines, and achieves optimal performance with expansions up to 3rd/4th order, aligning remarkably well with the distribution of pixel intensities in natural images. Through systematic perturbation analysis, we validate this alignment by isolating the contributions of specific image statistics to model performance, demonstrating how different orders of convolution process distinct aspects of visual information. Furthermore, Representational Similarity Analysis reveals distinct geometries across network layers, indicating qualitatively different modes of visual information processing. Our work bridges neuroscience and deep learning, offering a path towards more effective, biologically inspired computer vision models. It provides insights into visual information processing and lays the groundwork for neural networks that better capture complex visual patterns, particularly in resource-constrained scenarios.
Authors: Andrei-Robert Alexandrescu, Razvan-Gabriel Petec, Alexandru Manole, Laura-Silvia Diosan
Abstract: Deep Learning became an ubiquitous paradigm due to its extraordinary effectiveness and applicability in numerous domains. However, the approach suffers from the high demand of data required to achieve the potential of this type of model. An ever-increasing sub-field of Artificial Intelligence, Image Synthesis, aims to address this limitation through the design of intelligent models capable of creating original and realistic images, endeavour which could drastically reduce the need for real data. The Stable Diffusion generation paradigm recently propelled state-of-the-art approaches to exceed all previous benchmarks. In this work, we propose the ContRail framework based on the novel Stable Diffusion model ControlNet, which we empower through a multi-modal conditioning method. We experiment with the task of synthetic railway image generation, where we improve the performance in rail-specific tasks, such as rail semantic segmentation by enriching the dataset with realistic synthetic images.
Authors: Harish Thangaraj, Diya Katariya, Eshaan Joshi, Sangeetha N
Abstract: Pediatric brain tumors, particularly gliomas, represent a significant cause of cancer related mortality in children with complex infiltrative growth patterns that complicate treatment. Early, accurate segmentation of these tumors in neuroimaging data is crucial for effective diagnosis and intervention planning. This study presents a novel 3D UNet architecture with a spatial attention mechanism tailored for automated segmentation of pediatric gliomas. Using the BraTS pediatric glioma dataset with multiparametric MRI data, the proposed model captures multi-scale features and selectively attends to tumor relevant regions, enhancing segmentation precision and reducing interference from surrounding tissue. The model's performance is quantitatively evaluated using the Dice similarity coefficient and HD95, demonstrating improved delineation of complex glioma structured. This approach offers a promising advancement in automating pediatric glioma segmentation, with the potential to improve clinical decision making and outcomes.
Authors: Howard Zhang, Yuval Alaluf, Sizhuo Ma, Achuta Kadambi, Jian Wang, Kfir Aberman
Abstract: Face image restoration aims to enhance degraded facial images while addressing challenges such as diverse degradation types, real-time processing demands, and, most crucially, the preservation of identity-specific features. Existing methods often struggle with slow processing times and suboptimal restoration, especially under severe degradation, failing to accurately reconstruct finer-level identity details. To address these issues, we introduce InstantRestore, a novel framework that leverages a single-step image diffusion model and an attention-sharing mechanism for fast and personalized face restoration. Additionally, InstantRestore incorporates a novel landmark attention loss, aligning key facial landmarks to refine the attention maps, enhancing identity preservation. At inference time, given a degraded input and a small (~4) set of reference images, InstantRestore performs a single forward pass through the network to achieve near real-time performance. Unlike prior approaches that rely on full diffusion processes or per-identity model tuning, InstantRestore offers a scalable solution suitable for large-scale applications. Extensive experiments demonstrate that InstantRestore outperforms existing methods in quality and speed, making it an appealing choice for identity-preserving face restoration.
Authors: Wei-Hsiang Yu, Yen-Yu Lin, Ming-Hsuan Yang, Yi-Hsuan Tsai
Abstract: Recent advances in vision-language models (VLMs) have made significant progress in downstream tasks that require quantitative concepts such as facial age estimation and image quality assessment, enabling VLMs to explore applications like image ranking and retrieval. However, existing studies typically focus on the reasoning based on a single image and heavily depend on text prompting, limiting their ability to learn comprehensive understanding from multiple images. To address this, we propose an effective yet efficient approach that reframes the CLIP model into a learning-to-rank task and introduces a lightweight adapter to augment CLIP for text-guided image ranking. Specifically, our approach incorporates learnable prompts to adapt to new instructions for ranking purposes and an auxiliary branch with ranking-aware attention, leveraging text-conditioned visual differences for additional supervision in image ranking. Our ranking-aware adapter consistently outperforms fine-tuned CLIPs on various tasks and achieves competitive results compared to state-of-the-art models designed for specific tasks like facial age estimation and image quality assessment. Overall, our approach primarily focuses on ranking images with a single instruction, which provides a natural and generalized way of learning from visual differences across images, bypassing the need for extensive text prompts tailored to individual tasks. Code is available: https://github.com/uynaes/RankingAwareCLIP.
Authors: Antoine Gu\'edon, Tomoki Ichikawa, Kohei Yamashita, Ko Nishino
Abstract: We present a novel appearance model that simultaneously realizes explicit high-quality 3D surface mesh recovery and photorealistic novel view synthesis from sparse view samples. Our key idea is to model the underlying scene geometry Mesh as an Atlas of Charts which we render with 2D Gaussian surfels (MAtCha Gaussians). MAtCha distills high-frequency scene surface details from an off-the-shelf monocular depth estimator and refines it through Gaussian surfel rendering. The Gaussian surfels are attached to the charts on the fly, satisfying photorealism of neural volumetric rendering and crisp geometry of a mesh model, i.e., two seemingly contradicting goals in a single model. At the core of MAtCha lies a novel neural deformation model and a structure loss that preserve the fine surface details distilled from learned monocular depths while addressing their fundamental scale ambiguities. Results of extensive experimental validation demonstrate MAtCha's state-of-the-art quality of surface reconstruction and photorealism on-par with top contenders but with dramatic reduction in the number of input views and computational time. We believe MAtCha will serve as a foundational tool for any visual application in vision, graphics, and robotics that require explicit geometry in addition to photorealism. Our project page is the following: https://anttwo.github.io/matcha/
Authors: Viktor Rudnev, Gereon Fox, Mohamed Elgharib, Christian Theobalt, Vladislav Golyanik
Abstract: Volumetric reconstruction of dynamic scenes is an important problem in computer vision. It is especially challenging in poor lighting and with fast motion. It is partly due to the limitations of RGB cameras: To capture fast motion without much blur, the framerate must be increased, which in turn requires more lighting. In contrast, event cameras, which record changes in pixel brightness asynchronously, are much less dependent on lighting, making them more suitable for recording fast motion. We hence propose the first method to spatiotemporally reconstruct a scene from sparse multi-view event streams and sparse RGB frames. We train a sequence of cross-faded time-conditioned NeRF models, one per short recording segment. The individual segments are supervised with a set of event- and RGB-based losses and sparse-view regularisation. We assemble a real-world multi-view camera rig with six static event cameras around the object and record a benchmark multi-view event stream dataset of challenging motions. Our work outperforms RGB-based baselines, producing state-of-the-art results, and opens up the topic of multi-view event-based reconstruction as a new path for fast scene capture beyond RGB cameras. The code and the data will be released soon at https://4dqv.mpi-inf.mpg.de/DynEventNeRF/
Authors: XuDong Wang, Xingyi Zhou, Alireza Fathi, Trevor Darrell, Cordelia Schmid
Abstract: We present Visual Lexicon, a novel visual language that encodes rich image information into the text space of vocabulary tokens while retaining intricate visual details that are often challenging to convey in natural language. Unlike traditional methods that prioritize either high-level semantics (e.g., CLIP) or pixel-level reconstruction (e.g., VAE), ViLex simultaneously captures rich semantic content and fine visual details, enabling high-quality image generation and comprehensive visual scene understanding. Through a self-supervised learning pipeline, ViLex generates tokens optimized for reconstructing input images using a frozen text-to-image (T2I) diffusion model, preserving the detailed information necessary for high-fidelity semantic-level reconstruction. As an image embedding in the language space, ViLex tokens leverage the compositionality of natural languages, allowing them to be used independently as "text tokens" or combined with natural language tokens to prompt pretrained T2I models with both visual and textual inputs, mirroring how we interact with vision-language models (VLMs). Experiments demonstrate that ViLex achieves higher fidelity in image reconstruction compared to text embeddings--even with a single ViLex token. Moreover, ViLex successfully performs various DreamBooth tasks in a zero-shot, unsupervised manner without fine-tuning T2I models. Additionally, ViLex serves as a powerful vision encoder, consistently improving vision-language model performance across 15 benchmarks relative to a strong SigLIP baseline.
Authors: Yi-Lun Lee, Yi-Hsuan Tsai, Wei-Chen Chiu
Abstract: While large vision-language models (LVLMs) have shown impressive capabilities in generating plausible responses correlated with input visual contents, they still suffer from hallucinations, where the generated text inaccurately reflects visual contents. To address this, recent approaches apply contrastive decoding to calibrate the model's response via contrasting output distributions with original and visually distorted samples, demonstrating promising hallucination mitigation in a training-free manner. However, the potential of changing information in visual inputs is not well-explored, so a deeper investigation into the behaviors of visual contrastive decoding is of great interest. In this paper, we first explore various methods for contrastive decoding to change visual contents, including image downsampling and editing. Downsampling images reduces the detailed textual information while editing yields new contents in images, providing new aspects as visual contrastive samples. To further study benefits by using different contrastive samples, we analyze probability-level metrics, including entropy and distribution distance. Interestingly, the effect of these samples in mitigating hallucinations varies a lot across LVLMs and benchmarks. Based on our analysis, we propose a simple yet effective method to combine contrastive samples, offering a practical solution for applying contrastive decoding across various scenarios. Extensive experiments are conducted to validate the proposed fusion method among different benchmarks.
Authors: Xin Fei, Wenzhao Zheng, Yueqi Duan, Wei Zhan, Masayoshi Tomizuka, Kurt Keutzer, Jiwen Lu
Abstract: Realtime 4D reconstruction for dynamic scenes remains a crucial challenge for autonomous driving perception. Most existing methods rely on depth estimation through self-supervision or multi-modality sensor fusion. In this paper, we propose Driv3R, a DUSt3R-based framework that directly regresses per-frame point maps from multi-view image sequences. To achieve streaming dense reconstruction, we maintain a memory pool to reason both spatial relationships across sensors and dynamic temporal contexts to enhance multi-view 3D consistency and temporal integration. Furthermore, we employ a 4D flow predictor to identify moving objects within the scene to direct our network focus more on reconstructing these dynamic regions. Finally, we align all per-frame pointmaps consistently to the world coordinate system in an optimization-free manner. We conduct extensive experiments on the large-scale nuScenes dataset to evaluate the effectiveness of our method. Driv3R outperforms previous frameworks in 4D dynamic scene reconstruction, achieving 15x faster inference speed compared to methods requiring global alignment. Code: https://github.com/Barrybarry-Smith/Driv3R.
Authors: Yanbo Xu, Jayanth Srinivasa, Gaowen Liu, Shubham Tulsiani
Abstract: Score distillation of 2D diffusion models has proven to be a powerful mechanism to guide 3D optimization, for example enabling text-based 3D generation or single-view reconstruction. A common limitation of existing score distillation formulations, however, is that the outputs of the (mode-seeking) optimization are limited in diversity despite the underlying diffusion model being capable of generating diverse samples. In this work, inspired by the sampling process in denoising diffusion, we propose a score formulation that guides the optimization to follow generation paths defined by random initial seeds, thus ensuring diversity. We then present an approximation to adopt this formulation for scenarios where the optimization may not precisely follow the generation paths (e.g. a 3D representation whose renderings evolve in a co-dependent manner). We showcase the applications of our `Diverse Score Distillation' (DSD) formulation across tasks such as 2D optimization, text-based 3D inference, and single-view reconstruction. We also empirically validate DSD against prior score distillation formulations and show that it significantly improves sample diversity while preserving fidelity.
Authors: Nicolas Dufour, David Picard, Vicky Kalogeiton, Loic Landrieu
Abstract: Global visual geolocation predicts where an image was captured on Earth. Since images vary in how precisely they can be localized, this task inherently involves a significant degree of ambiguity. However, existing approaches are deterministic and overlook this aspect. In this paper, we aim to close the gap between traditional geolocalization and modern generative methods. We propose the first generative geolocation approach based on diffusion and Riemannian flow matching, where the denoising process operates directly on the Earth's surface. Our model achieves state-of-the-art performance on three visual geolocation benchmarks: OpenStreetView-5M, YFCC-100M, and iNat21. In addition, we introduce the task of probabilistic visual geolocation, where the model predicts a probability distribution over all possible locations instead of a single point. We introduce new metrics and baselines for this task, demonstrating the advantages of our diffusion-based approach. Codes and models will be made available.
Authors: Ruihan Gao, Kangle Deng, Gengshan Yang, Wenzhen Yuan, Jun-Yan Zhu
Abstract: 3D generation methods have shown visually compelling results powered by diffusion image priors. However, they often fail to produce realistic geometric details, resulting in overly smooth surfaces or geometric details inaccurately baked in albedo maps. To address this, we introduce a new method that incorporates touch as an additional modality to improve the geometric details of generated 3D assets. We design a lightweight 3D texture field to synthesize visual and tactile textures, guided by 2D diffusion model priors on both visual and tactile domains. We condition the visual texture generation on high-resolution tactile normals and guide the patch-based tactile texture refinement with a customized TextureDreambooth. We further present a multi-part generation pipeline that enables us to synthesize different textures across various regions. To our knowledge, we are the first to leverage high-resolution tactile sensing to enhance geometric details for 3D generation tasks. We evaluate our method in both text-to-3D and image-to-3D settings. Our experiments demonstrate that our method provides customized and realistic fine geometric textures while maintaining accurate alignment between two modalities of vision and touch.
Authors: M. Hamza Mughal, Rishabh Dabral, Merel C. J. Scholman, Vera Demberg, Christian Theobalt
Abstract: Non-verbal communication often comprises of semantically rich gestures that help convey the meaning of an utterance. Producing such semantic co-speech gestures has been a major challenge for the existing neural systems that can generate rhythmic beat gestures, but struggle to produce semantically meaningful gestures. Therefore, we present RAG-Gesture, a diffusion-based gesture generation approach that leverages Retrieval Augmented Generation (RAG) to produce natural-looking and semantically rich gestures. Our neuro-explicit gesture generation approach is designed to produce semantic gestures grounded in interpretable linguistic knowledge. We achieve this by using explicit domain knowledge to retrieve exemplar motions from a database of co-speech gestures. Once retrieved, we then inject these semantic exemplar gestures into our diffusion-based gesture generation pipeline using DDIM inversion and retrieval guidance at the inference time without any need of training. Further, we propose a control paradigm for guidance, that allows the users to modulate the amount of influence each retrieval insertion has over the generated sequence. Our comparative evaluations demonstrate the validity of our approach against recent gesture generation approaches. The reader is urged to explore the results on our project page.
Authors: Vincent Tao Hu, Bj\"orn Ommer
Abstract: In generative models, two paradigms have gained attraction in various applications: next-set prediction-based Masked Generative Models and next-noise prediction-based Non-Autoregressive Models, e.g., Diffusion Models. In this work, we propose using discrete-state models to connect them and explore their scalability in the vision domain. First, we conduct a step-by-step analysis in a unified design space across two types of models including timestep-independence, noise schedule, temperature, guidance strength, etc in a scalable manner. Second, we re-cast typical discriminative tasks, e.g., image segmentation, as an unmasking process from [MASK]tokens on a discrete-state model. This enables us to perform various sampling processes, including flexible conditional sampling by only training once to model the joint distribution. All aforementioned explorations lead to our framework named Discrete Interpolants, which enables us to achieve state-of-the-art or competitive performance compared to previous discrete-state based methods in various benchmarks, like ImageNet256, MS COCO, and video dataset FaceForensics. In summary, by leveraging [MASK] in discrete-state models, we can bridge Masked Generative and Non-autoregressive Diffusion models, as well as generative and discriminative tasks.
Authors: Li Zhou, Changsheng Fang, Bahareh Morovati, Yongtong Liu, Shuo Han, Yongshun Xu, Hengyong Yu
Abstract: This paper introduces $\rho$-NeRF, a self-supervised approach that sets a new standard in novel view synthesis (NVS) and computed tomography (CT) reconstruction by modeling a continuous volumetric radiance field enriched with physics-based attenuation priors. The $\rho$-NeRF represents a three-dimensional (3D) volume through a fully-connected neural network that takes a single continuous four-dimensional (4D) coordinate, spatial location $(x, y, z)$ and an initialized attenuation value ($\rho$), and outputs the attenuation coefficient at that position. By querying these 4D coordinates along X-ray paths, the classic forward projection technique is applied to integrate attenuation data across the 3D space. By matching and refining pre-initialized attenuation values derived from traditional reconstruction algorithms like Feldkamp-Davis-Kress algorithm (FDK) or conjugate gradient least squares (CGLS), the enriched schema delivers superior fidelity in both projection synthesis and image recognition.
Authors: Guilherme G. D. Fernandes, Vitor S. P. P. Oliveira, Jo\~ao P. I. Astolfo
Abstract: The mapping of ocean floor layers is a current challenge for the oil industry. Existing solution methods involve mapping through seismic methods and wave inversion, which are complex and computationally expensive. The introduction of artificial neural networks, specifically UNet, to predict velocity models based on seismic shots reflected from the ocean floor shows promise for optimising this process. In this study, two neural network architectures are validated for velocity model inversion and compared in terms of stability metrics such as loss function and similarity coefficient, as well as the differences between predicted and actual models. Indeed, neural networks prove promising as a solution to this challenge, achieving S{\o}rensen-Dice coefficient values above 70%.
Authors: Beguet Florian, Lanquetin Sandrine, Raffin Romain
Abstract: Mesh segmentation represents a crucial task in computer graphics and geometric analysis, with diverse applications spanning texture mapping, animation, and beyond. This paper introduces an innovative Reeb graph-based mesh segmentation method that seamlessly integrates geometric and topological features to achieve flexible and robust segmentation results. The proposed approach encompasses three primary phases. First, an enhanced topological skeleton construction efficiently captures the Reeb graph structure while preserving degenerate critical points. Second, a topological simplification process employing critical point cancellation reduces graph complexity while maintaining essential shape features and correspondences. Finally, a region growing algorithm leverages both Reeb graph adjacency and mesh vertex connectivity to generate contiguous, semantically meaningful segments. The presented method exhibits computational efficiency, achieving a complexity of $O(n \log n$) for a mesh containing n vertices. Its versatility and effectiveness are validated through application to both local geometry-based segmentation using the Shape Index and part-based decomposition utilizing the Shape Diameter Function. This flexible framework establishes a solid foundation for advanced analysis and applications across various domains, offering new possibilities for mesh processing and understanding.
Authors: Marien Renaud, Arthur Leclaire, Nicolas Papadakis
Abstract: One key ingredient of image restoration is to define a realistic prior on clean images to complete the missing information in the observation. State-of-the-art restoration methods rely on a neural network to encode this prior. Moreover, typical image distributions are invariant to some set of transformations, such as rotations or flips. However, most deep architectures are not designed to represent an invariant image distribution. Recent works have proposed to overcome this difficulty by including equivariance properties within a Plug-and-Play paradigm. In this work, we propose a unified framework named Equivariant Regularization by Denoising (ERED) based on equivariant denoisers and stochastic optimization. We analyze the convergence of this algorithm and discuss its practical benefit.
Authors: Ung Hwang, Chang-Hun Lee, Kijung Yoon
Abstract: Osteoporosis is a widespread and chronic metabolic bone disease that often remains undiagnosed and untreated due to limited access to bone mineral density (BMD) tests like Dual-energy X-ray absorptiometry (DXA). In response to this challenge, current advancements are pivoting towards detecting osteoporosis by examining alternative indicators from peripheral bone areas, with the goal of increasing screening rates without added expenses or time. In this paper, we present a method to predict osteoporosis using hand and wrist X-ray images, which are both widely accessible and affordable, though their link to DXA-based data is not thoroughly explored. We employ a sophisticated image segmentation model that utilizes a mixture of probabilistic U-Net decoders, specifically designed to capture predictive uncertainty in the segmentation of the ulna, radius, and metacarpal bones. This model is formulated as an optimal transport (OT) problem, enabling it to handle the inherent uncertainties in image segmentation more effectively. Further, we adopt a self-supervised learning (SSL) approach to extract meaningful representations without the need for explicit labels, and move on to classify osteoporosis in a supervised manner. Our method is evaluated on a dataset with 192 individuals, cross-referencing their verified osteoporosis conditions against the standard DXA test. With a notable classification score, this integration of uncertainty-aware segmentation and self-supervised learning represents a pioneering effort in leveraging vision-based techniques for the early detection of osteoporosis from peripheral skeletal sites.
Authors: R. Prashanth
Abstract: Early and accurate detection of Parkinson's disease (PD) is a crucial diagnostic challenge carrying immense clinical significance, for effective treatment regimens and patient management. For instance, a group of subjects termed SWEDD who are clinically diagnosed as PD, but show normal Single Photon Emission Computed Tomography (SPECT) scans, change their diagnosis as non-PD after few years of follow up, and in the meantime, they are treated with PD medications which do more harm than good. In this work, machine learning models are developed using features from SPECT images to detect early PD and SWEDD subjects from normal. These models were observed to perform with high accuracy. It is inferred from the study that these diagnostic models carry potential to help PD clinicians in the diagnostic process
Authors: Priya Sundaresan, Hengyuan Hu, Quan Vuong, Jeannette Bohg, Dorsa Sadigh
Abstract: While imitation learning (IL) offers a promising framework for teaching robots various behaviors, learning complex tasks remains challenging. Existing IL policies struggle to generalize effectively across visual and spatial variations even for simple tasks. In this work, we introduce SPHINX: Salient Point-based Hybrid ImitatioN and eXecution, a flexible IL policy that leverages multimodal observations (point clouds and wrist images), along with a hybrid action space of low-frequency, sparse waypoints and high-frequency, dense end effector movements. Given 3D point cloud observations, SPHINX learns to infer task-relevant points within a point cloud, or salient points, which support spatial generalization by focusing on semantically meaningful features. These salient points serve as anchor points to predict waypoints for long-range movement, such as reaching target poses in free-space. Once near a salient point, SPHINX learns to switch to predicting dense end-effector movements given close-up wrist images for precise phases of a task. By exploiting the strengths of different input modalities and action representations for different manipulation phases, SPHINX tackles complex tasks in a sample-efficient, generalizable manner. Our method achieves 86.7% success across 4 real-world and 2 simulated tasks, outperforming the next best state-of-the-art IL baseline by 41.1% on average across 440 real world trials. SPHINX additionally generalizes to novel viewpoints, visual distractors, spatial arrangements, and execution speeds with a 1.7x speedup over the most competitive baseline. Our website (http://sphinx-manip.github.io) provides open-sourced code for data collection, training, and evaluation, along with supplementary videos.
Authors: Abdulrahman Kerim, Leandro Soriano Marcolino, Erickson R. Nascimento, Richard Jiang
Abstract: Supervised machine learning methods require large-scale training datasets to perform well in practice. Synthetic data has been showing great progress recently and has been used as a complement to real data. However, there is yet a great urge to assess the usability of synthetically generated data. To this end, we propose a novel UCB-based training procedure combined with a dynamic usability metric. Our proposed metric integrates low-level and high-level information from synthetic images and their corresponding real and synthetic datasets, surpassing existing traditional metrics. By utilizing a UCB-based dynamic approach ensures continual enhancement of model learning. Unlike other approaches, our method effectively adapts to changes in the machine learning model's state and considers the evolving utility of training samples during the training process. We show that our metric is an effective way to rank synthetic images based on their usability. Furthermore, we propose a new attribute-aware bandit pipeline for generating synthetic data by integrating a Large Language Model with Stable Diffusion. Quantitative results show that our approach can boost the performance of a wide range of supervised classifiers. Notably, we observed an improvement of up to 10% in classification accuracy compared to traditional approaches, demonstrating the effectiveness of our approach. Our source code, datasets, and additional materials are publically available at https://github.com/A-Kerim/Synthetic-Data-Usability-2024.
URLs: https://github.com/A-Kerim/Synthetic-Data-Usability-2024.
Authors: Muhammad Umair Danish
Abstract: With the recent advancements in the field of information industry, critical data in the form of digital images is best understood by the human brain. Therefore, digital images play a significant part and backbone role in many areas such as image processing, vision computing, robotics, and bio-medical. Such use of digital images is practically implementable in various real-time scenarios like biological sciences, medicine, gaming technology, computer information and communication technology, data and statistical science, radiological sciences and medical imaging technology, and medical lab technology. However, when any digital image is sent electronically or captured via camera, it is likely to get corrupted or degraded by the available of degradation factors. To eradicate this problem, several image denoising algorithms have been proposed in the literature focusing on robust, low-cost and fast techniques to improve output performance. Consequently, in this research project, an earnest effort has been made to study various image denoising algorithms. A specific focus is given to the start-of-the-art techniques namely: NL-means, K-SVD, and BM3D. The standard images, natural images, texture images, synthetic images, and images from other datasets have been tested via these algorithms, and a detailed set of convincing results have been provided for efficient comparison.
Authors: Jiong Lin, Lechen Zhang, Kwansoo Lee, Jialong Ning, Judah Goldfeder, Hod Lipson
Abstract: Robot description models are essential for simulation and control, yet their creation often requires significant manual effort. To streamline this modeling process, we introduce AutoURDF, an unsupervised approach for constructing description files for unseen robots from point cloud frames. Our method leverages a cluster-based point cloud registration model that tracks the 6-DoF transformations of point clusters. Through analyzing cluster movements, we hierarchically address the following challenges: (1) moving part segmentation, (2) body topology inference, and (3) joint parameter estimation. The complete pipeline produces robot description files that are fully compatible with existing simulators. We validate our method across a variety of robots, using both synthetic and real-world scan data. Results indicate that our approach outperforms previous methods in registration and body topology estimation accuracy, offering a scalable solution for automated robot modeling.
Authors: Runhao Zeng, Dingjie Zhou, Qiwei Liang, Junlin Liu, Hui Li, Changxin Huang, Jianqiang Li, Xiping Hu, Fuchun Sun
Abstract: Learning behavior in legged robots presents a significant challenge due to its inherent instability and complex constraints. Recent research has proposed the use of a large language model (LLM) to generate reward functions in reinforcement learning, thereby replacing the need for manually designed rewards by experts. However, this approach, which relies on textual descriptions to define learning objectives, fails to achieve controllable and precise behavior learning with clear directionality. In this paper, we introduce a new video2reward method, which directly generates reward functions from videos depicting the behaviors to be mimicked and learned. Specifically, we first process videos containing the target behaviors, converting the motion information of individuals in the videos into keypoint trajectories represented as coordinates through a video2text transforming module. These trajectories are then fed into an LLM to generate the reward function, which in turn is used to train the policy. To enhance the quality of the reward function, we develop a video-assisted iterative reward refinement scheme that visually assesses the learned behaviors and provides textual feedback to the LLM. This feedback guides the LLM to continually refine the reward function, ultimately facilitating more efficient behavior learning. Experimental results on tasks involving bipedal and quadrupedal robot motion control demonstrate that our method surpasses the performance of state-of-the-art LLM-based reward generation methods by over 37.6% in terms of human normalized score. More importantly, by switching video inputs, we find our method can rapidly learn diverse motion behaviors such as walking and running.
Authors: Kazutoshi Akita, Norimichi Ukita
Abstract: Super-resolution (SR) with arbitrary scale factor and cost-and-quality controllability at test time is essential for various applications. While several arbitrary-scale SR methods have been proposed, these methods require us to modify the model structure and retrain it to control the computational cost and SR quality. To address this limitation, we propose a novel SR method using a Recurrent Neural Network (RNN) with the Fourier representation. In our method, the RNN sequentially estimates Fourier components, each consisting of frequency and amplitude, and aggregates these components to reconstruct an SR image. Since the RNN can adjust the number of recurrences at test time, we can control the computational cost and SR quality in a single model: fewer recurrences (i.e., fewer Fourier components) lead to lower cost but lower quality, while more recurrences (i.e., more Fourier components) lead to better quality but more cost. Experimental results prove that more Fourier components improve the PSNR score. Furthermore, even with fewer Fourier components, our method achieves a lower PSNR drop than other state-of-the-art arbitrary-scale SR methods.
Authors: Cailian Ruan, Chengyue Huang, Yahe Yang
Abstract: This study introduces an evaluation framework for multimodal models in medical imaging diagnostics. We developed a pipeline incorporating data preprocessing, model inference, and preference-based evaluation, expanding an initial set of 500 clinical cases to 3,000 through controlled augmentation. Our method combined medical images with clinical observations to generate assessments, using Claude 3.5 Sonnet for independent evaluation against physician-authored diagnoses. The results indicated varying performance across models, with Llama 3.2-90B outperforming human diagnoses in 85.27% of cases. In contrast, specialized vision models like BLIP2 and Llava showed preferences in 41.36% and 46.77% of cases, respectively. This framework highlights the potential of large multimodal models to outperform human diagnostics in certain tasks.
Authors: Feng Li, Jiusong Luo, Wanjun Xia
Abstract: Speech emotion recognition (SER) remains a challenging yet crucial task due to the inherent complexity and diversity of human emotions. To address this problem, researchers attempt to fuse information from other modalities via multimodal learning. However, existing multimodal fusion techniques often overlook the intricacies of cross-modal interactions, resulting in suboptimal feature representations. In this paper, we propose WavFusion, a multimodal speech emotion recognition framework that addresses critical research problems in effective multimodal fusion, heterogeneity among modalities, and discriminative representation learning. By leveraging a gated cross-modal attention mechanism and multimodal homogeneous feature discrepancy learning, WavFusion demonstrates improved performance over existing state-of-the-art methods on benchmark datasets. Our work highlights the importance of capturing nuanced cross-modal interactions and learning discriminative representations for accurate multimodal SER. Experimental results on two benchmark datasets (IEMOCAP and MELD) demonstrate that WavFusion succeeds over the state-of-the-art strategies on emotion recognition.
Authors: Hao-Chun Yang, Sicheng Dai, Saige Rutherford, Christian Gaser Andre F Marquand, Christian F Beckmann, Thomas Wolfers
Abstract: Unsupervised anomaly detection in brain imaging is challenging. In this paper, we propose a self-supervised masked mesh learning for unsupervised anomaly detection in 3D cortical surfaces. Our framework leverages the intrinsic geometry of the cortical surface to learn a self-supervised representation that captures the underlying structure of the brain. We introduce a masked mesh convolutional neural network (MMN) that learns to predict masked regions of the cortical surface. By training the MMN on a large dataset of healthy subjects, we learn a representation that captures the normal variation in the cortical surface. We then use this representation to detect anomalies in unseen individuals by calculating anomaly scores based on the reconstruction error of the MMN. We evaluate our framework by training on population-scale dataset UKB and HCP-Aging and testing on two datasets of Alzheimer's disease patients ADNI and OASIS3. Our results show that our framework can detect anomalies in cortical thickness, cortical volume, and cortical sulcus features, which are known to be sensitive biomarkers for Alzheimer's disease. Our proposed framework provides a promising approach for unsupervised anomaly detection based on normative variation of cortical features.
Authors: Mozhgan Naderi, Maryam Rastgarpour, Amir Reza Takhsha
Abstract: Alzheimer's Disease (AD) is a progressive neurological disorder that can result in significant cognitive impairment and dementia. Accurate and timely diagnosis is essential for effective treatment and management of this disease. In this study, we proposed two low-parameter Convolutional Neural Networks (CNNs), IR-BRAINNET and Modified-DEMNET, designed to detect the early stages of AD accurately. We also introduced an ensemble model that averages their outputs to reduce variance across the CNNs and enhance AD detection. Both CNNs are trained, and all models are evaluated using a Magnetic Resonance Imaging (MRI) dataset from the Kaggle database. The dataset includes images of four stages of dementia, with an uneven class distribution. To mitigate challenges stemming from the inherent imbalance in the dataset, we employed the Synthetic Minority Over-sampling Technique (SMOTE) to generate additional instances for minority classes. In the NO-SMOTE scenario, despite the imbalanced distribution, the ensemble model achieved 98.28% accuracy, outperforming IR-BRAINNET (97.26%) and Modified-DEMNET (95.54%), with Wilcoxon p-values of 2.9e-3 and 5.20e-6, respectively, indicating significant improvement in correct predictions through the use of the average function. In the SMOTE scenario, the ensemble model achieved 99.92% accuracy (1.64% improvement over NO-SMOTE), IR-BRAINNET reached 99.80% (2.54% improvement), and Modified-DEMNET attained 99.72% (4.18% improvement). Based on the experimental findings, averaging the models' outputs enhanced AD diagnosis in both scenarios, while the diversity in the dataset introduced by SMOTE-generated instances significantly improved performance. Furthermore, the compact models we proposed outperformed those from previous studies, even in the presence of an imbalanced distribution.
Authors: Deepshikha Bhati, Fnu Neha, Md Amiruzzaman, Angela Guercio, Deepak Kumar Shukla, Ben Ward
Abstract: Interpreting complex neural networks is crucial for understanding their decision-making processes, particularly in applications where transparency and accountability are essential. This proposed method addresses this need by focusing on layer-wise Relevance Propagation (LRP), a technique used in explainable artificial intelligence (XAI) to attribute neural network outputs to input features through backpropagated relevance scores. Existing LRP methods often struggle with precision in evaluating individual neuron contributions. To overcome this limitation, we present a novel approach that improves the parsing of selected neurons during LRP backward propagation, using the Visual Geometry Group 16 (VGG16) architecture as a case study. Our method creates neural network graphs to highlight critical paths and visualizes these paths with heatmaps, optimizing neuron selection through accuracy metrics like Mean Squared Error (MSE) and Symmetric Mean Absolute Percentage Error (SMAPE). Additionally, we utilize a deconvolutional visualization technique to reconstruct feature maps, offering a comprehensive view of the network's inner workings. Extensive experiments demonstrate that our approach enhances interpretability and supports the development of more transparent artificial intelligence (AI) systems for computer vision applications. This advancement has the potential to improve the trustworthiness of AI models in real-world machine vision applications, thereby increasing their reliability and effectiveness.
Authors: Reed Chen, Courtney Trutna Paley, Wren Wightman, Lisa Hobson-Webb, Yohei Harada, Felix Jin, Ouwen Huang, Mark Palmeri, Kathryn Nightingale
Abstract: Research ultrasound scanners such as the Verasonics Vantage often lack the advanced image processing algorithms used by clinical systems. Image quality is even lower in plane wave imaging - often used for shear wave elasticity imaging (SWEI) - which sacrifices spatial resolution for temporal resolution. As a result, delay-and-summed images acquired from SWEI have limited interpretability. In this project, a two-stage machine learning model was trained to enhance single plane wave images of muscle acquired with a Verasonics Vantage system. The first stage of the model consists of a U-Net trained to emulate plane wave compounding, histogram matching, and unsharp masking using paired images. The second stage consists of a CycleGAN trained to emulate clinical muscle B-modes using unpaired images. This two-stage model was implemented on the Verasonics Vantage research ultrasound scanner, and its ability to provide high-speed image formation at a frame rate of 28.5 +/- 0.6 FPS from a single plane wave transmit was demonstrated. A reader study with two physicians demonstrated that these processed images had significantly greater structural fidelity and less speckle than the original plane wave images.
Authors: Junha Lee, Sojung An, Sujeong You, Namik Cho
Abstract: Numerical weather prediction (NWP) models are fundamental in meteorology for simulating and forecasting the behavior of various atmospheric variables. The accuracy of precipitation forecasts and the acquisition of sufficient lead time are crucial for preventing hazardous weather events. However, the performance of NWP models is limited by the nonlinear and unpredictable patterns of extreme weather phenomena driven by temporal dynamics. In this regard, we propose a \textbf{S}elf-\textbf{S}upervised \textbf{L}earning with \textbf{P}robabilistic \textbf{D}ensity \textbf{L}abeling (SSLPDL) for estimating rainfall probability by post-processing NWP forecasts. Our post-processing method uses self-supervised learning (SSL) with masked modeling for reconstructing atmospheric physics variables, enabling the model to learn the dependency between variables. The pre-trained encoder is then utilized in transfer learning to a precipitation segmentation task. Furthermore, we introduce a straightforward labeling approach based on probability density to address the class imbalance in extreme weather phenomena like heavy rain events. Experimental results show that SSLPDL surpasses other precipitation forecasting models in regional precipitation post-processing and demonstrates competitive performance in extending forecast lead times. Our code is available at https://github.com/joonha425/SSLPDL
Authors: Qing Wu, Hongjiang Wei, Jingyi Yu, Yuyao Zhang
Abstract: Ring artifacts are prevalent in 3D cone-beam computed tomography (CBCT) due to non-ideal responses of X-ray detectors, severely degrading imaging quality and reliability. Current state-of-the-art (SOTA) ring artifact reduction (RAR) algorithms rely on extensive paired CT samples for supervised learning. While effective, these methods do not fully capture the physical characteristics of ring artifacts, leading to pronounced performance drops when applied to out-of-domain data. Moreover, their applications to 3D CBCT are limited by high memory demands. In this work, we introduce \textbf{Riner}, an unsupervised method formulating 3D CBCT RAR as a multi-parameter inverse problem. Our core innovation is parameterizing the X-ray detector responses as solvable variables within a differential physical model. By jointly optimizing a neural field to represent artifact-free CT images and estimating response parameters directly from raw measurements, Riner eliminates the need for external training data. Moreover, it accommodates diverse CT geometries, enhancing practical usability. Empirical results on both simulated and real-world datasets show that Riner surpasses existing SOTA RAR methods in performance.
Authors: Sertac Kilickaya, Cansu Celebioglu, Levent Eren, Murat Askar
Abstract: Condition monitoring of induction machines is crucial to prevent costly interruptions and equipment failure. Mechanical faults such as misalignment and rotor issues are among the most common problems encountered in industrial environments. To effectively monitor and detect these faults, a variety of sensors, including accelerometers, current sensors, temperature sensors, and microphones, are employed in the field. As a non-contact alternative, thermal imaging offers a powerful monitoring solution by capturing temperature variations in machines with thermal cameras. In this study, we propose using 2-dimensional Self-Organized Operational Neural Networks (Self-ONNs) to diagnose misalignment and broken rotor faults from thermal images of squirrel-cage induction motors. We evaluate our approach by benchmarking its performance against widely used Convolutional Neural Networks (CNNs), including ResNet, EfficientNet, PP-LCNet, SEMNASNet, and MixNet, using a Workswell InfraRed Camera (WIC). Our results demonstrate that Self-ONNs, with their non-linear neurons and self-organizing capability, achieve diagnostic performance comparable to more complex CNN models while utilizing a shallower architecture with just three operational layers. Its streamlined architecture ensures high performance and is well-suited for deployment on edge devices, enabling its use also in more complex multi-function and/or multi-device monitoring systems.
Authors: Juan Yeo, Jinkwan Jang, Kyubyung Chae, Seongkyu Mun, Taesup Kim
Abstract: Recent studies show that pretrained vision models can boost performance in audio downstream tasks. To enhance the performance further, an additional pretraining stage with large scale audio data is typically required to infuse audio specific knowledge into the vision model. However, such approaches require extensive audio data and a carefully designed objective function. In this work, we propose bypassing the pretraining stage by directly fine-tuning the vision model with our Look Aside Adapter (LoAA) designed for efficient audio understanding. Audio spectrum data is represented across two heterogeneous dimensions time and frequency and we refine adapters to facilitate interactions between tokens across these dimensions. Our experiments demonstrate that our adapters allow vision models to reach or surpass the performance of pretrained audio models in various audio and speech tasks, offering a resource efficient and effective solution for leveraging vision models in audio applications.
Authors: Mehwish Mehmood, Shahzaib Iqbal, Tariq Mahmood Khan, Ivor Spence, Muhammad Fahim
Abstract: The analysis of retinal images for the diagnosis of various diseases is one of the emerging areas of research. Recently, the research direction has been inclined towards investigating several changes in retinal blood vessels in subjects with many neurological disorders, including dementia. This research focuses on detecting diseases early by improving the performance of models for segmentation of retinal vessels with fewer parameters, which reduces computational costs and supports faster processing. This paper presents a novel lightweight encoder-decoder model that segments retinal vessels to improve the efficiency of disease detection. It incorporates multi-scale convolutional blocks in the encoder to accurately identify vessels of various sizes and thicknesses. The bottleneck of the model integrates the Focal Modulation Attention and Spatial Feature Refinement Blocks to refine and enhance essential features for efficient segmentation. The decoder upsamples features and integrates them with the corresponding feature in the encoder using skip connections and the spatial feature refinement block at every upsampling stage to enhance feature representation at various scales. The estimated computation complexity of our proposed model is around 29.60 GFLOP with 0.71 million parameters and 2.74 MB of memory size, and it is evaluated using public datasets, that is, DRIVE, CHASE\_DB, and STARE. It outperforms existing models with dice scores of 86.44\%, 84.22\%, and 87.88\%, respectively.
Authors: Meilong Xu, Saumya Gupta, Xiaoling Hu, Chen Li, Shahira Abousamra, Dimitris Samaras, Prateek Prasanna, Chao Chen
Abstract: Accurately modeling multi-class cell topology is crucial in digital pathology, as it provides critical insights into tissue structure and pathology. The synthetic generation of cell topology enables realistic simulations of complex tissue environments, enhances downstream tasks by augmenting training data, aligns more closely with pathologists' domain knowledge, and offers new opportunities for controlling and generalizing the tumor microenvironment. In this paper, we propose a novel approach that integrates topological constraints into a diffusion model to improve the generation of realistic, contextually accurate cell topologies. Our method refines the simulation of cell distributions and interactions, increasing the precision and interpretability of results in downstream tasks such as cell detection and classification. To assess the topological fidelity of generated layouts, we introduce a new metric, Topological Frechet Distance (TopoFD), which overcomes the limitations of traditional metrics like FID in evaluating topological structure. Experimental results demonstrate the effectiveness of our approach in generating multi-class cell layouts that capture intricate topological relationships.
Authors: Joohyuk Park, Yongjeong Oh, Yongjune Kim, Yo-Seb Jeon
Abstract: Semantic communications provide significant performance gains over traditional communications by transmitting task-relevant semantic features through wireless channels. However, most existing studies rely on end-to-end (E2E) training of neural-type encoders and decoders to ensure effective transmission of these semantic features. To enable semantic communications without relying on E2E training, this paper presents a vision transformer (ViT)-based semantic communication system with importance-aware quantization (IAQ) for wireless image transmission. The core idea of the presented system is to leverage the attention scores of a pretrained ViT model to quantify the importance levels of image patches. Based on this idea, our IAQ framework assigns different quantization bits to image patches based on their importance levels. This is achieved by formulating a weighted quantization error minimization problem, where the weight is set to be an increasing function of the attention score. Then, an optimal incremental allocation method and a low-complexity water-filling method are devised to solve the formulated problem. Our framework is further extended for realistic digital communication systems by modifying the bit allocation problem and the corresponding allocation methods based on an equivalent binary symmetric channel (BSC) model. Simulations on single-view and multi-view image classification tasks show that our IAQ framework outperforms conventional image compression methods in both error-free and realistic communication scenarios.
Authors: Seyed Mohsen Hosseini, Mahdieh Soleymani Baghshah
Abstract: A novel method for tackling the problem of imbalanced data in medical image segmentation is proposed in this work. In balanced cross entropy (CE) loss, which is a type of weighted CE loss, the weight assigned to each class is the in-verse of the class frequency. These balancing weights are expected to equalize the effect of each class on the overall loss and prevent the model from being biased towards the majority class. But, as it has been shown in previous studies, this method degrades the performance by a large margin. Therefore, balanced CE is not a popular loss in medical segmentation tasks, and usually a region-based loss, like the Dice loss, is used to address the class imbalance problem. In the pro-posed method, the weighting of cross entropy loss for each class is based on a dilated area of each class mask, and balancing weights are assigned to each class together with its surrounding pixels. The goal of this study is to show that the performance of balanced CE loss can be greatly improved my modifying its weighting strategy. Experiments on different datasets show that the proposed dilated balanced CE (DBCE) loss outperforms the balanced CE loss by a large margin and produces superior results compared to CE loss, and its performance is similar to the performance of the combination of Dice and CE loss. This means that a weighted cross entropy loss with the right weighing strategy can be as effective as a region-based loss in handling the problem of class imbalance in medical segmentation tasks.
Authors: Vincent Gherold, Ioannis Mandralis, Eric Sihite, Adarsh Salagame, Alireza Ramezani, Morteza Gharib
Abstract: Autonomous robots operating in real environments are often faced with decisions on how best to navigate their surroundings. In this work, we address a particular instance of this problem: how can a robot autonomously decide on the energetically optimal path to follow given a high-level objective and information about the surroundings? To tackle this problem we developed a self-supervised learning method that allows the robot to estimate the cost of transport of its surroundings using only vision inputs. We apply our method to the multi-modal mobility morphobot (M4), a robot that can drive, fly, segway, and crawl through its environment. By deploying our system in the real world, we show that our method accurately assigns different cost of transports to various types of environments e.g. grass vs smooth road. We also highlight the low computational cost of our method, which is deployed on an Nvidia Jetson Orin Nano robotic compute unit. We believe that this work will allow multi-modal robotic platforms to unlock their full potential for navigation and exploration tasks.
Authors: Mengyu Sun, Dimeng Xia, Shusen Zhao, Weibin Zhang, Yaobin He
Abstract: In order to improve image quality of projection in industrial applications, generally, a standard method is to increase the current or exposure time, which might cause overexposure of detector units in areas of thin objects or backgrounds. Increasing the projection sampling is a better method to address the issue, but it also leads to significant noise in the reconstructed image. This paper proposed a projection domain denoising algorithm based on the features of the projection domain for this case. This algorithm utilized the similarity of projections of neighboring veiws to reduce image noise quickly and effectively. The availability of the algorithm proposed in this work has been conducted by numerical simulation and practical data experiments.
Authors: L\'ea Demeule, Mahtab Sandhu, Glen Berseth
Abstract: The majority of signal data captured in the real world uses numerous sensors with different resolutions. In practice, however, most deep learning architectures are fixed-resolution; they consider a single resolution at training time and inference time. This is convenient to implement but fails to fully take advantage of the diverse signal data that exists. In contrast, other deep learning architectures are adaptive-resolution; they directly allow various resolutions to be processed at training time and inference time. This benefits robustness and computational efficiency but introduces difficult design constraints that hinder mainstream use. In this work, we address the shortcomings of both fixed-resolution and adaptive-resolution methods by introducing Adaptive Resolution Residual Networks (ARRNs), which inherit the advantages of adaptive-resolution methods and the ease of use of fixed-resolution methods. We construct ARRNs from Laplacian residuals, which serve as generic adaptive-resolution adapters for fixed-resolution layers, and which allow casting high-resolution ARRNs into low-resolution ARRNs at inference time by simply omitting high-resolution Laplacian residuals, thus reducing computational cost on low-resolution signals without compromising performance. We complement this novel component with Laplacian dropout, which regularizes for robustness to a distribution of lower resolutions, and which also regularizes for errors that may be induced by approximate smoothing kernels in Laplacian residuals. We provide a solid grounding for the advantageous properties of ARRNs through a theoretical analysis based on neural operators, and empirically show that ARRNs embrace the challenge posed by diverse resolutions with greater flexibility, robustness, and computational efficiency.
Authors: Fei Yu, Zhe Xiang, Nan Che, Zhuoran Zhang, Yuandi Li, Junxiao Xue, Zhiguo Wan
Abstract: Multimodal semantic communication, which integrates various data modalities such as text, images, and audio, significantly enhances communication efficiency and reliability. It has broad application prospects in fields such as artificial intelligence, autonomous driving, and smart homes. However, current research primarily relies on analog channels and assumes constant channel states (perfect CSI), which is inadequate for addressing dynamic physical channels and noise in real-world scenarios. Existing methods often focus on single modality tasks and fail to handle multimodal stream data, such as video and audio, and their corresponding tasks. Furthermore, current semantic encoding and decoding modules mainly transmit single modality features, neglecting the need for multimodal semantic enhancement and recognition tasks. To address these challenges, this paper proposes a pilot-guided framework for multimodal semantic communication specifically tailored for audio-visual event localization tasks. This framework utilizes digital pilot codes and channel modules to guide the state of analog channels in real-wold scenarios and designs Euler-based multimodal semantic encoding and decoding that consider time-frequency characteristics based on dynamic channel state. This approach effectively handles multimodal stream source data, especially for audio-visual event localization tasks. Extensive numerical experiments demonstrate the robustness of the proposed framework in channel changes and its support for various communication scenarios. The experimental results show that the framework outperforms existing benchmark methods in terms of Signal-to-Noise Ratio (SNR), highlighting its advantage in semantic communication quality.
Authors: Bochuan Cao, Jinyuan Jia, Chuxuan Hu, Wenbo Guo, Zhen Xiang, Jinghui Chen, Bo Li, Dawn Song
Abstract: Backdoor attacks aim to inject a backdoor into a classifier such that it predicts any input with an attacker-chosen backdoor trigger as an attacker-chosen target class. Existing backdoor attacks require either retraining the classifier with some clean data or modifying the model's architecture. As a result, they are 1) not applicable when clean data is unavailable, 2) less efficient when the model is large, and 3) less stealthy due to architecture changes. In this work, we propose DFBA, a novel retraining-free and data-free backdoor attack without changing the model architecture. Technically, our proposed method modifies a few parameters of a classifier to inject a backdoor. Through theoretical analysis, we verify that our injected backdoor is provably undetectable and unremovable by various state-of-the-art defenses under mild assumptions. Our evaluation on multiple datasets further demonstrates that our injected backdoor: 1) incurs negligible classification loss, 2) achieves 100% attack success rates, and 3) bypasses six existing state-of-the-art defenses. Moreover, our comparison with a state-of-the-art non-data-free backdoor attack shows our attack is more stealthy and effective against various defenses while achieving less classification accuracy loss.
Authors: Jiazhao Zhang, Kunyu Wang, Shaoan Wang, Minghan Li, Haoran Liu, Songlin Wei, Zhongyuan Wang, Zhizheng Zhang, He Wang
Abstract: A practical navigation agent must be capable of handling a wide range of interaction demands, such as following instructions, searching objects, answering questions, tracking people, and more. Existing models for embodied navigation fall short of serving as practical generalists in the real world, as they are often constrained by specific task configurations or pre-defined maps with discretized waypoints. In this work, we present Uni-NaVid, the first video-based vision-language-action (VLA) model designed to unify diverse embodied navigation tasks and enable seamless navigation for mixed long-horizon tasks in unseen real-world environments. Uni-NaVid achieves this by harmonizing the input and output data configurations for all commonly used embodied navigation tasks and thereby integrating all tasks in one model. For training Uni-NaVid, we collect 3.6 million navigation data samples in total from four essential navigation sub-tasks and foster synergy in learning across them. Extensive experiments on comprehensive navigation benchmarks clearly demonstrate the advantages of unification modeling in Uni-NaVid and show it achieves state-of-the-art performance. Additionally, real-world experiments confirm the model's effectiveness and efficiency, shedding light on its strong generalizability.
Authors: Junqiao Wang, Zhongliang Yu, Dong Zhou, Jiaqi Shi, Runran Deng
Abstract: The capability of UAVs for efficient autonomous navigation and obstacle avoidance in complex and unknown environments is critical for applications in agricultural irrigation, disaster relief and logistics. In this paper, we propose the DPRL (Distributed Privileged Reinforcement Learning) navigation algorithm, an end-to-end policy designed to address the challenge of high-speed autonomous UAV navigation under partially observable environmental conditions. Our approach combines deep reinforcement learning with privileged learning to overcome the impact of observation data corruption caused by partial observability. We leverage an asymmetric Actor-Critic architecture to provide the agent with privileged information during training, which enhances the model's perceptual capabilities. Additionally, we present a multi-agent exploration strategy across diverse environments to accelerate experience collection, which in turn expedites model convergence. We conducted extensive simulations across various scenarios, benchmarking our DPRL algorithm against the state-of-the-art navigation algorithms. The results consistently demonstrate the superior performance of our algorithm in terms of flight efficiency, robustness and overall success rate.
Authors: Yijie Dang, Weijun Ma, Xiaohu Luo
Abstract: Since the outbreak of the COVID-19 pandemic in 2019, medical imaging has emerged as a primary modality for diagnosing COVID-19 pneumonia. In clinical settings, the segmentation of lung infections from computed tomography images enables rapid and accurate quantification and diagnosis of COVID-19. Segmentation of COVID-19 infections in the lungs poses a formidable challenge, primarily due to the indistinct boundaries and limited contrast presented by ground glass opacity manifestations. Moreover, the confounding similarity between infiltrates, lung tissues, and lung walls further complicates this segmentation task. To address these challenges, this paper introduces a novel deep network architecture, called CAD-Unet, for segmenting COVID-19 lung infections. In this architecture, capsule networks are incorporated into the existing Unet framework. Capsule networks represent a novel network architecture that differs from traditional convolutional neural networks. They utilize vectors for information transfer among capsules, facilitating the extraction of intricate lesion spatial information. Additionally, we design a capsule encoder path and establish a coupling path between the unet encoder and the capsule encoder. This design maximizes the complementary advantages of both network structures while achieving efficient information fusion. \noindent Finally, extensive experiments are conducted on four publicly available datasets, encompassing binary segmentation tasks and multi-class segmentation tasks. The experimental results demonstrate the superior segmentation performance of the proposed model. The code has been released at: https://github.com/AmanoTooko-jie/CAD-Unet.
Authors: Jesse Hagenaars, Yilun Wu, Federico Paredes-Vall\'es, Stein Stroobants, Guido de Croon
Abstract: Event cameras provide low-latency perception for only milliwatts of power. This makes them highly suitable for resource-restricted, agile robots such as small flying drones. Self-supervised learning based on contrast maximization holds great potential for event-based robot vision, as it foregoes the need to high-frequency ground truth and allows for online learning in the robot's operational environment. However, online, onboard learning raises the major challenge of achieving sufficient computational efficiency for real-time learning, while maintaining competitive visual perception performance. In this work, we improve the time and memory efficiency of the contrast maximization learning pipeline. Benchmarking experiments show that the proposed pipeline achieves competitive results with the state of the art on the task of depth estimation from events. Furthermore, we demonstrate the usability of the learned depth for obstacle avoidance through real-world flight experiments. Finally, we compare the performance of different combinations of pre-training and fine-tuning of the depth estimation networks, showing that on-board domain adaptation is feasible given a few minutes of flight.
Authors: Ilke Adalioglu, Serkan Kiranyaz, Mete Ahishali, Aysen Degerli, Tahir Hamid, Rahmat Ghaffar, Ridha Hamila, Moncef Gabbouj
Abstract: Echocardiography is the most widely used imaging to monitor cardiac functions, serving as the first line in early detection of myocardial ischemia and infarction. However, echocardiography often suffers from several artifacts including sensor noise, lack of contrast, severe saturation, and missing myocardial segments which severely limit its usage in clinical diagnosis. In recent years, several machine learning methods have been proposed to improve echocardiography views. Yet, these methods usually address only a specific problem (e.g. denoising) and thus cannot provide a robust and reliable restoration in general. On the other hand, cardiac MRI provides a clean view of the heart without suffering such severe issues. However, due to its significantly higher cost, it is often only afforded by a few major hospitals, hence hindering its use and accessibility. In this pilot study, we propose a novel approach to transform echocardiography into the cardiac MRI view. For this purpose, Echo2MRI dataset, consisting of echocardiography and real cardiac MRI image pairs, is composed and will be shared publicly. A dedicated Cycle-consistent Generative Adversarial Network (Cycle-GAN) is trained to learn the transformation from echocardiography frames to cardiac MRI views. An extensive set of qualitative evaluations shows that the proposed transformer can synthesize high-quality artifact-free synthetic cardiac MRI views from a given sequence of echocardiography frames. Medical evaluations performed by a group of cardiologists further demonstrate that synthetic MRI views are indistinguishable from their original counterparts and are preferred over their initial sequence of echocardiography frames for diagnosis in 78.9% of the cases.
Authors: Aakash Madhav Rao, Debayan Gupta
Abstract: The development of generative models in the past decade has allowed for hyperrealistic data synthesis. While potentially beneficial, this synthetic data generation process has been relatively underexplored in cancer histopathology. One algorithm for synthesising a realistic image is diffusion; it iteratively converts an image to noise and learns the recovery process from this noise [Wang and Vastola, 2023]. While effective, it is highly computationally expensive for high-resolution images, rendering it infeasible for histopathology. The development of Variational Autoencoders (VAEs) has allowed us to learn the representation of complex high-resolution images in a latent space. A vital by-product of this is the ability to compress high-resolution images to space and recover them lossless. The marriage of diffusion and VAEs allows us to carry out diffusion in the latent space of an autoencoder, enabling us to leverage the realistic generative capabilities of diffusion while maintaining reasonable computational requirements. Rombach et al. [2021b] and Yellapragada et al. [2023] build foundational models for this task, paving the way to generate realistic histopathology images. In this paper, we discuss the pitfalls of current methods, namely [Yellapragada et al., 2023] and resolve critical errors while proposing improvements along the way. Our methods achieve an FID score of 21.11, beating its SOTA counterparts in [Yellapragada et al., 2023] by 1.2 FID, while presenting a train-time GPU memory usage reduction of 7%.
Authors: Kuan Xu, Zeyu Jiang, Haozhi Cao, Shenghai Yuan, Chen Wang, Lihua Xie
Abstract: Scene Coordinate Regression (SCR) is a visual localization technique that utilizes deep neural networks (DNN) to directly regress 2D-3D correspondences for camera pose estimation. However, current SCR methods often face challenges in handling repetitive textures and meaningless areas due to their reliance on implicit triangulation. In this paper, we propose an efficient scene coordinate encoding and relocalization method. Compared with the existing SCR methods, we design a unified architecture for both scene encoding and salient keypoint detection, enabling our system to focus on encoding informative regions, thereby significantly enhancing efficiency. Additionally, we introduce a mechanism that leverages sequential information during both map encoding and relocalization, which strengthens implicit triangulation, particularly in repetitive texture environments. Comprehensive experiments conducted across indoor and outdoor datasets demonstrate that the proposed system outperforms other state-of-the-art (SOTA) SCR methods. Our single-frame relocalization mode improves the recall rate of our baseline by 6.4% and increases the running speed from 56Hz to 90Hz. Furthermore, our sequence-based mode increases the recall rate by 11% while maintaining the original efficiency.
Authors: Hongkang Song, Zihui Zhang, Yanpeng Zhou, Jie Hu, Zishuo Wang, Hou Him Chan, Chon Lok Lei, Chen Xu, Yu Xin, Bo Yang
Abstract: Spinal cord tumors significantly contribute to neurological morbidity and mortality. Precise morphometric quantification, encompassing the size, location, and type of such tumors, holds promise for optimizing treatment planning strategies. Although recent methods have demonstrated excellent performance in medical image segmentation, they primarily focus on discerning shapes with relatively large morphology such as brain tumors, ignoring the challenging problem of identifying spinal cord tumors which tend to have tiny sizes, diverse locations, and shapes. To tackle this hard problem of multiclass spinal cord tumor segmentation, we propose a new method, called BATseg, to learn a tumor surface distance field by applying our new multiclass boundary-aware loss function. To verify the effectiveness of our approach, we also introduce the first and large-scale spinal cord tumor dataset. It comprises gadolinium-enhanced T1-weighted 3D MRI scans from 653 patients and contains the four most common spinal cord tumor types: astrocytomas, ependymomas, hemangioblastomas, and spinal meningiomas. Extensive experiments on our dataset and another public kidney tumor segmentation dataset show that our proposed method achieves superior performance for multiclass tumor segmentation.
Authors: Jiayan Chen, Kai Li, Zhanjin Wang, Zhan Wang, Jianqiang Huang
Abstract: Hepatic echinococcosis (HE) is a prevalent disease in economically underdeveloped pastoral areas, where adequate medical resources are usually lacking. Existing methods often ignore multi-scale feature fusion or focus only on feature fusion between adjacent levels, which may lead to insufficient feature fusion. To address these issues, we propose HES-UNet, an efficient and accurate model for HE lesion segmentation. This model combines convolutional layers and attention modules to capture local and global features. During downsampling, the multi-directional downsampling block (MDB) is employed to integrate high-frequency and low-frequency features, effectively extracting image details. The multi-scale aggregation block (MAB) aggregates multi-scale feature information. In contrast, the multi-scale upsampling Block (MUB) learns highly abstract features and supplies this information to the skip connection module to fuse multi-scale features. Due to the distinct regional characteristics of HE, there is currently no publicly available high-quality dataset for training our model. We collected CT slice data from 268 patients at a certain hospital to train and evaluate the model. The experimental results show that HES-UNet achieves state-of-the-art performance on our dataset, achieving an overall Dice Similarity Coefficient (DSC) of 89.21%, which is 1.09% higher than that of TransUNet. The project page is available at https://chenjiayan-qhu.github.io/HES-UNet-page.
Authors: Zilin Wang, Shengqi Chen, Jianrong Dai, Shirui Qin, Ying Cao, Ruiao Zhao, Guohua Wu, Yuan Tang, Jiayun Chen
Abstract: Objective:To develop a no-reference image quality assessment method using automated distortion recognition to boost MRI-guided radiotherapy precision.Methods:We analyzed 106,000 MR images from 10 patients with liver metastasis,captured with the Elekta Unity MR-LINAC.Our No-Reference Quality Assessment Model includes:1)image preprocessing to enhance visibility of key diagnostic features;2)feature extraction and directional analysis using MSCN coefficients across four directions to capture textural attributes and gradients,vital for identifying image features and potential distortions;3)integrative Quality Index(QI)calculation,which integrates features via AGGD parameter estimation and K-means clustering.The QI,based on a weighted MAD computation of directional scores,provides a comprehensive image quality measure,robust against outliers.LOO-CV assessed model generalizability and performance.Tumor tracking algorithm performance was compared with and without preprocessing to verify tracking accuracy enhancements.Results:Preprocessing significantly improved image quality,with the QI showing substantial positive changes and surpassing other metrics.After normalization,the QI's average value was 79.6 times higher than CNR,indicating improved image definition and contrast.It also showed higher sensitivity in detail recognition with average values 6.5 times and 1.7 times higher than Tenengrad gradient and entropy.The tumor tracking algorithm confirmed significant tracking accuracy improvements with preprocessed images,validating preprocessing effectiveness.Conclusions:This study introduces a novel no-reference image quality evaluation method based on automated distortion recognition,offering a new quality control tool for MRIgRT tumor tracking.It enhances clinical application accuracy and facilitates medical image quality assessment standardization, with significant clinical and research value.
Authors: Sooyong Jang, Kuk Jin Jang, Hyonyoung Choi, Yong-Seop Han, Seongjin Lee, Jin-hyun Kim, Insup Lee
Abstract: Timely detection and treatment are essential for maintaining eye health. Visual acuity (VA), which measures the clarity of vision at a distance, is a crucial metric for managing eye health. Machine learning (ML) techniques have been introduced to assist in VA measurement, potentially alleviating clinicians' workloads. However, the inherent uncertainties in ML models make relying solely on them for VA prediction less than ideal. The VA prediction task involves multiple sources of uncertainty, requiring more robust approaches. A promising method is to build prediction sets or intervals rather than point estimates, offering coverage guarantees through techniques like conformal prediction and Probably Approximately Correct (PAC) prediction sets. Despite the potential, to date, these approaches have not been applied to the VA prediction task.To address this, we propose a method for deriving prediction intervals for estimating visual acuity from fundus images with a PAC guarantee. Our experimental results demonstrate that the PAC guarantees are upheld, with performance comparable to or better than that of two prior works that do not provide such guarantees.
Authors: Shanshan Wang, Shoujun Yu, Jian Cheng, Sen Jia, Changjun Tie, Jiayu Zhu, Haohao Peng, Yijing Dong, Jianzhong He, Fan Zhang, Yaowen Xing, Xiuqin Jia, Qi Yang, Qiyuan Tian, Hua Guo, Guobin Li, Hairong Zheng
Abstract: Diffusion magnetic resonance imaging (dMRI) provides critical insights into the microstructural and connectional organization of the human brain. However, the availability of high-field, open-access datasets that include raw k-space data for advanced research remains limited. To address this gap, we introduce Diff5T, a first comprehensive 5.0 Tesla diffusion MRI dataset focusing on the human brain. This dataset includes raw k-space data and reconstructed diffusion images, acquired using a variety of imaging protocols. Diff5T is designed to support the development and benchmarking of innovative methods in artifact correction, image reconstruction, image preprocessing, diffusion modelling and tractography. The dataset features a wide range of diffusion parameters, including multiple b-values and gradient directions, allowing extensive research applications in studying human brain microstructure and connectivity. With its emphasis on open accessibility and detailed benchmarks, Diff5T serves as a valuable resource for advancing human brain mapping research using diffusion MRI, fostering reproducibility, and enabling collaboration across the neuroscience and medical imaging communities.
Authors: Ciro Benito Raggio, Mathias Krohmer Zabaleta, Nils Skupien, Oliver Blanck, Francesco Cicone, Giuseppe Lucio Cascini, Paolo Zaffino, Lucia Migliorelli, Maria Francesca Spadea
Abstract: The generation of Synthetic Computed Tomography (sCT) images has become a pivotal methodology in modern clinical practice, particularly in the context of Radiotherapy (RT) treatment planning. The use of sCT enables the calculation of doses, pushing towards Magnetic Resonance Imaging (MRI) guided radiotherapy treatments. Moreover, with the introduction of MRI-Positron Emission Tomography (PET) hybrid scanners, the derivation of sCT from MRI can improve the attenuation correction of PET images. Deep learning methods for MRI-to-sCT have shown promising results, but their reliance on single-centre training dataset limits generalisation capabilities to diverse clinical settings. Moreover, creating centralised multicentre datasets may pose privacy concerns. To solve the issues, this study introduces FedSynthCT-Brain, a framework based on the Federated Learning (FL) paradigm for MRI-to-sCT in brain imaging. We reproduced a federation across four European and American centres using a U-Net-based model. The approach was implemented using data from centres belonging the federation and it was tested on an unseen dataset from a centre outside the federation. In the case of the unseen centre, the federated model achieved a median Mean Absolute Error (MAE) of 102.0 HU across 23 patients, with an interquartile range of 96.7-110.5 HU. The median (interquartile range) for the Structural Similarity Index (SSIM) and the Peak Signal to Noise Ratio (PNSR) were 0.89 (0.86-0.89) and 26.58 (25.52-27.42), respectively. The analysis of the results showed acceptable performances of the federated approach, thus highlighting the potential of FL to enhance MRI-to-sCT to improve generalisability and advancing safe and equitable clinical applications while fostering collaboration and preserving data privacy.
Authors: Yi-Kai Zhang, Xu-Xiang Zhong, Shiyin Lu, Qing-Guo Chen, De-Chuan Zhan, Han-Jia Ye
Abstract: The rapid advancements in Large Language Models (LLMs) have significantly expanded their applications, ranging from multilingual support to domain-specific tasks and multimodal integration. In this paper, we present OmniEvalKit, a novel benchmarking toolbox designed to evaluate LLMs and their omni-extensions across multilingual, multidomain, and multimodal capabilities. Unlike existing benchmarks that often focus on a single aspect, OmniEvalKit provides a modular, lightweight, and automated evaluation system. It is structured with a modular architecture comprising a Static Builder and Dynamic Data Flow, promoting the seamless integration of new models and datasets. OmniEvalKit supports over 100 LLMs and 50 evaluation datasets, covering comprehensive evaluations across thousands of model-dataset combinations. OmniEvalKit is dedicated to creating an ultra-lightweight and fast-deployable evaluation framework, making downstream applications more convenient and versatile for the AI community.
Authors: Sebastian Dziadzio, Vishaal Udandarao, Karsten Roth, Ameya Prabhu, Zeynep Akata, Samuel Albanie, Matthias Bethge
Abstract: Model merging combines multiple expert models - finetuned from a base foundation model on diverse tasks and domains - into a single, more capable model. However, most existing model merging approaches assume that all experts are available simultaneously. In reality, new tasks and domains emerge progressively over time, requiring strategies to integrate the knowledge of expert models as they become available: a process we call temporal model merging. The temporal dimension introduces unique challenges not addressed in prior work, raising new questions such as: when training for a new task, should the expert model start from the merged past experts or from the original base model? Should we merge all models at each time step? Which merging techniques are best suited for temporal merging? Should different strategies be used to initialize the training and deploy the model? To answer these questions, we propose a unified framework called TIME - Temporal Integration of Model Expertise - which defines temporal model merging across three axes: (1) Initialization Phase, (2) Deployment Phase, and (3) Merging Technique. Using TIME, we study temporal model merging across model sizes, compute budgets, and learning horizons on the FoMo-in-Flux benchmark. Our comprehensive suite of experiments across TIME allows us to uncover key insights for temporal model merging, offering a better understanding of current challenges and best practices for effective temporal model merging.
Authors: Adhiraj Ghosh, Sebastian Dziadzio, Ameya Prabhu, Vishaal Udandarao, Samuel Albanie, Matthias Bethge
Abstract: Traditional fixed test sets fall short in evaluating open-ended capabilities of foundation models. To address this, we propose ONEBench(OpeN-Ended Benchmarking), a new testing paradigm that consolidates individual evaluation datasets into a unified, ever-expanding sample pool. ONEBench allows users to generate custom, open-ended evaluation benchmarks from this pool, corresponding to specific capabilities of interest. By aggregating samples across test sets, ONEBench enables the assessment of diverse capabilities beyond those covered by the original test sets, while mitigating overfitting and dataset bias. Most importantly, it frames model evaluation as a collective process of selecting and aggregating sample-level tests. The shift from task-specific benchmarks to ONEBench introduces two challenges: (1)heterogeneity and (2)incompleteness. Heterogeneity refers to the aggregation over diverse metrics, while incompleteness describes comparing models evaluated on different data subsets. To address these challenges, we explore algorithms to aggregate sparse measurements into reliable model scores. Our aggregation algorithm ensures identifiability(asymptotically recovering ground-truth scores) and rapid convergence, enabling accurate model ranking with less data. On homogenous datasets, we show our aggregation algorithm provides rankings that highly correlate with those produced by average scores. We also demonstrate robustness to ~95% of measurements missing, reducing evaluation cost by up to 20x with little-to-no change in model rankings. We introduce ONEBench-LLM for language models and ONEBench-LMM for vision-language models, unifying evaluations across these domains. Overall, we present a technique for open-ended evaluation, which can aggregate over incomplete, heterogeneous sample-level measurements to continually grow a benchmark alongside the rapidly developing foundation models.
Authors: Meera Hahn, Wenjun Zeng, Nithish Kannen, Rich Galt, Kartikeya Badola, Been Kim, Zi Wang
Abstract: User prompts for generative AI models are often underspecified, leading to sub-optimal responses. This problem is particularly evident in text-to-image (T2I) generation, where users commonly struggle to articulate their precise intent. This disconnect between the user's vision and the model's interpretation often forces users to painstakingly and repeatedly refine their prompts. To address this, we propose a design for proactive T2I agents equipped with an interface to (1) actively ask clarification questions when uncertain, and (2) present their understanding of user intent as an understandable belief graph that a user can edit. We build simple prototypes for such agents and verify their effectiveness through both human studies and automated evaluation. We observed that at least 90% of human subjects found these agents and their belief graphs helpful for their T2I workflow. Moreover, we develop a scalable automated evaluation approach using two agents, one with a ground truth image and the other tries to ask as few questions as possible to align with the ground truth. On DesignBench, a benchmark we created for artists and designers, the COCO dataset (Lin et al., 2014), and ImageInWords (Garg et al., 2024), we observed that these T2I agents were able to ask informative questions and elicit crucial information to achieve successful alignment with at least 2 times higher VQAScore (Lin et al., 2024) than the standard single-turn T2I generation. Demo: https://github.com/google-deepmind/proactive_t2i_agents.
URLs: https://github.com/google-deepmind/proactive_t2i_agents.
Authors: Zhefei Gong, Pengxiang Ding, Shangke Lyu, Siteng Huang, Mingyang Sun, Wei Zhao, Zhaoxin Fan, Donglin Wang
Abstract: In robotic visuomotor policy learning, diffusion-based models have achieved significant success in improving the accuracy of action trajectory generation compared to traditional autoregressive models. However, they suffer from inefficiency due to multiple denoising steps and limited flexibility from complex constraints. In this paper, we introduce Coarse-to-Fine AutoRegressive Policy (CARP), a novel paradigm for visuomotor policy learning that redefines the autoregressive action generation process as a coarse-to-fine, next-scale approach. CARP decouples action generation into two stages: first, an action autoencoder learns multi-scale representations of the entire action sequence; then, a GPT-style transformer refines the sequence prediction through a coarse-to-fine autoregressive process. This straightforward and intuitive approach produces highly accurate and smooth actions, matching or even surpassing the performance of diffusion-based policies while maintaining efficiency on par with autoregressive policies. We conduct extensive evaluations across diverse settings, including single-task and multi-task scenarios on state-based and image-based simulation benchmarks, as well as real-world tasks. CARP achieves competitive success rates, with up to a 10% improvement, and delivers 10x faster inference compared to state-of-the-art policies, establishing a high-performance, efficient, and flexible paradigm for action generation in robotic tasks.
Authors: Mara Levy, Siddhant Haldar, Lerrel Pinto, Abhinav Shirivastava
Abstract: Developing generalizable robot policies that can robustly handle varied environmental conditions and object instances remains a fundamental challenge in robot learning. While considerable efforts have focused on collecting large robot datasets and developing policy architectures to learn from such data, naively learning from visual inputs often results in brittle policies that fail to transfer beyond the training data. This work presents Prescriptive Point Priors for Policies or P3-PO, a novel framework that constructs a unique state representation of the environment leveraging recent advances in computer vision and robot learning to achieve improved out-of-distribution generalization for robot manipulation. This representation is obtained through two steps. First, a human annotator prescribes a set of semantically meaningful points on a single demonstration frame. These points are then propagated through the dataset using off-the-shelf vision models. The derived points serve as an input to state-of-the-art policy architectures for policy learning. Our experiments across four real-world tasks demonstrate an overall 43% absolute improvement over prior methods when evaluated in identical settings as training. Further, P3-PO exhibits 58% and 80% gains across tasks for new object instances and more cluttered environments respectively. Videos illustrating the robot's performance are best viewed at point-priors.github.io.
Authors: Laurent Valentin Jospin, Farid Boussaid, Hamid Laga, Mohammed Bennamoun
Abstract: Cost-based image patch matching is at the core of various techniques in computer vision, photogrammetry and remote sensing. When the subpixel disparity between the reference patch in the source and target images is required, either the cost function or the target image have to be interpolated. While cost-based interpolation is the easiest to implement, multiple works have shown that image based interpolation can increase the accuracy of the subpixel matching, but usually at the cost of expensive search procedures. This, however, is problematic, especially for very computation intensive applications such as stereo matching or optical flow computation. In this paper, we show that closed form formulae for subpixel disparity computation for the case of one dimensional matching, e.g., in the case of rectified stereo images where the search space is of one dimension, exists when using the standard NCC, SSD and SAD cost functions. We then demonstrate how to generalize the proposed formulae to the case of high dimensional search spaces, which is required for unrectified stereo matching and optical flow extraction. We also compare our results with traditional cost volume interpolation formulae as well as with state-of-the-art cost-based refinement methods, and show that the proposed formulae bring a small improvement over the state-of-the-art cost-based methods in the case of one dimensional search spaces, and a significant improvement when the search space is two dimensional.
Authors: Zhi-Song Liu, Li-Wen Wang, Wan-Chi Siu, Vicky Kalogeiton
Abstract: Image style transfer has attracted widespread attention in the past few years. Despite its remarkable results, it requires additional style images available as references, making it less flexible and inconvenient. Using text is the most natural way to describe the style. More importantly, text can describe implicit abstract styles, like styles of specific artists or art movements. In this paper, we propose a text-driven image style transfer (TxST) that leverages advanced image-text encoders to control arbitrary style transfer. We introduce a contrastive training strategy to effectively extract style descriptions from the image-text model (i.e., CLIP), which aligns stylization with the text description. To this end, we also propose a novel and efficient attention module that explores cross-attentions to fuse style and content features. Finally, we achieve an arbitrary artist-aware image style transfer to learn and transfer specific artistic characters such as Picasso, oil painting, or a rough sketch. Extensive experiments demonstrate that our approach outperforms the state-of-the-art methods on both image and textual styles. Moreover, it can mimic the styles of one or many artists to achieve attractive results, thus highlighting a promising direction in image style transfer.
Authors: Qing Guo, Ruofei Wang, Rui Huang, Shuifa Sun, Yuxiang Zhang
Abstract: Change detection (CD) aims to detect change regions within an image pair captured at different times, playing a significant role in diverse real-world applications. Nevertheless, most of the existing works focus on designing advanced network architectures to map the feature difference to the final change map while ignoring the influence of the quality of the feature difference. In this paper, we study the CD from a different perspective, i.e., how to optimize the feature difference to highlight changes and suppress unchanged regions, and propose a novel module denoted as iterative difference-enhanced transformers (IDET). IDET contains three transformers: two transformers for extracting the long-range information of the two images and one transformer for enhancing the feature difference. In contrast to the previous transformers, the third transformer takes the outputs of the first two transformers to guide the enhancement of the feature difference iteratively. To achieve more effective refinement, we further propose the multi-scale IDET-based change detection that uses multi-scale representations of the images for multiple feature difference refinements and proposes a coarse-to-fine fusion strategy to combine all refinements. Our final CD method outperforms seven state-of-the-art methods on six large-scale datasets under diverse application scenarios, which demonstrates the importance of feature difference enhancements and the effectiveness of IDET.
Authors: Xuecheng Wu, Mengmeng Tian, Lanhang Zhai
Abstract: With the fast development of artificial intelligence and short videos, emotion recognition in short videos has become one of the most important research topics in human-computer interaction. At present, most emotion recognition methods still stay in a single modality. However, in daily life, human beings will usually disguise their real emotions, which leads to the problem that the accuracy of single modal emotion recognition is relatively terrible. Moreover, it is not easy to distinguish similar emotions. Therefore, we propose a new approach denoted as ICANet to achieve multimodal short video emotion recognition by employing three different modalities of audio, video and optical flow, making up for the lack of a single modality and then improving the accuracy of emotion recognition in short videos. ICANet has a better accuracy of 80.77% on the IEMOCAP benchmark, exceeding the SOTA methods by 15.89%.
Authors: Xuecheng Wu, Mengmeng Tian, Lanhang Zhai
Abstract: Recently, the domestic COVID-19 epidemic situation is serious, but in public places, some people do not wear masks or wear masks incorrectly, which requires the relevant staff to instantly remind and supervise them to wear masks correctly. However, in the face of such an important and complicated work, it is very necessary to carry out automated mask-wearing detection in public places. This paper proposes a new mask-wearing detection method based on improved YOLOv4. Specifically, firstly, we add the Coordinate Attention Module to the backbone to coordinate feature fusion and representation. Secondly, we conduct a series of network structural improvements to enhance the model performance and robustness. Thirdly, we adaptively deploy the K-means clustering algorithm to make the nine anchor boxes more suitable for our NPMD dataset. The experiments show that the improved YOLOv4 performs better, exceeding the baseline by 4.06\% AP with a comparable speed of 64.37 FPS.
Authors: Xuqian Ren, Shaopeng Yang, Saihui Hou, Chunshui Cao, Xu Liu, Yongzhen Huang
Abstract: Previous gait recognition methods primarily trained on labeled datasets, which require painful labeling effort. However, using a pre-trained model on a new dataset without fine-tuning can lead to significant performance degradation. So to make the pre-trained gait recognition model able to be fine-tuned on unlabeled datasets, we propose a new task: Unsupervised Gait Recognition (UGR). We introduce a new cluster-based baseline to solve UGR with cluster-level contrastive learning. But we further find more challenges this task meets. First, sequences of the same person in different clothes tend to cluster separately due to the significant appearance changes. Second, sequences taken from 0{\deg} and 180{\deg} views lack walking postures and do not cluster with sequences taken from other views. To address these challenges, we propose a Selective Fusion method, which includes Selective Cluster Fusion (SCF) and Selective Sample Fusion (SSF). With SCF, we merge matched clusters of the same person wearing different clothes by updating the cluster-level memory bank with a multi-cluster update strategy. And in SSF, we merge sequences taken from front/back views gradually with curriculum learning. Extensive experiments show the effectiveness of our method in improving the rank-1 accuracy in walking with different coats condition and front/back views conditions.
Authors: Zixuan Chen, Xiaohua Xie, Lingxiao Yang, Jianhuang Lai
Abstract: Anomaly detectors are widely used in industrial manufacturing to detect and localize unknown defects in query images. These detectors are trained on anomaly-free samples and have successfully distinguished anomalies from most normal samples. However, hard-normal examples are scattered and far apart from most normal samples, and thus they are often mistaken for anomalies by existing methods. To address this issue, we propose Hard-normal Example-aware Template Mutual Matching (HETMM), an efficient framework to build a robust prototype-based decision boundary. Specifically, HETMM employs the proposed Affine-invariant Template Mutual Matching (ATMM) to mitigate the affection brought by the affine transformations and easy-normal examples. By mutually matching the pixel-level prototypes within the patch-level search spaces between query and template set, ATMM can accurately distinguish between hard-normal examples and anomalies, achieving low false-positive and missed-detection rates. In addition, we also propose PTS to compress the original template set for speed-up. PTS selects cluster centres and hard-normal examples to preserve the original decision boundary, allowing this tiny set to achieve comparable performance to the original one. Extensive experiments demonstrate that HETMM outperforms state-of-the-art methods, while using a 60-sheet tiny set can achieve competitive performance and real-time inference speed (around 26.1 FPS) on a Quadro 8000 RTX GPU. HETMM is training-free and can be hot-updated by directly inserting novel samples into the template set, which can promptly address some incremental learning issues in industrial manufacturing.
Authors: Rui Shao, Tianxing Wu, Liqiang Nie, Ziwei Liu
Abstract: Existing deepfake detection methods fail to generalize well to unseen or degraded samples, which can be attributed to the over-fitting of low-level forgery patterns. Here we argue that high-level semantics are also indispensable recipes for generalizable forgery detection. Recently, large pre-trained Vision Transformers (ViTs) have shown promising generalization capability. In this paper, we propose the first parameter-efficient tuning approach for deepfake detection, namely DeepFake-Adapter, to effectively and efficiently adapt the generalizable high-level semantics from large pre-trained ViTs to aid deepfake detection. Given large pre-trained models but limited deepfake data, DeepFake-Adapter introduces lightweight yet dedicated dual-level adapter modules to a ViT while keeping the model backbone frozen. Specifically, to guide the adaptation process to be aware of both global and local forgery cues of deepfake data, 1) we not only insert Globally-aware Bottleneck Adapters in parallel to MLP layers of ViT, 2) but also actively cross-attend Locally-aware Spatial Adapters with features from ViT. Unlike existing deepfake detection methods merely focusing on low-level forgery patterns, the forgery detection process of our model can be regularized by generalizable high-level semantics from a pre-trained ViT and adapted by global and local low-level forgeries of deepfake data. Extensive experiments on several standard deepfake detection benchmarks validate the effectiveness of our approach. Notably, DeepFake-Adapter demonstrates a convincing advantage under cross-dataset and cross-manipulation settings. The code has been released at https://github.com/rshaojimmy/DeepFake-Adapter.
Authors: Roi Peleg, Yair Smadar, Teddy Lazebnik, Assaf Hoogi
Abstract: The performance of deep learning models is critically dependent on sophisticated optimization strategies. While existing optimizers have shown promising results, many rely on first-order Exponential Moving Average (EMA) techniques, which often limit their ability to track complex gradient trends accurately. This fact can lead to a significant lag in trend identification and suboptimal optimization, particularly in highly dynamic gradient behavior. To address this fundamental limitation, we introduce Fast Adaptive Moment Estimation (FAME), a novel optimization technique that leverages the power of Triple Exponential Moving Average. By incorporating an advanced tracking mechanism, FAME enhances responsiveness to data dynamics, mitigates trend identification lag, and optimizes learning efficiency. Our comprehensive evaluation encompasses different computer vision tasks including image classification, object detection, and semantic segmentation, integrating FAME into 30 distinct architectures ranging from lightweight CNNs to Vision Transformers. Through rigorous benchmarking against state-of-the-art optimizers, FAME demonstrates superior accuracy and robustness. Notably, it offers high scalability, delivering substantial improvements across diverse model complexities, architectures, tasks, and benchmarks.
Authors: Runmin Cong, Mengyao Sun, Sanyi Zhang, Xiaofei Zhou, Wei Zhang, Yao Zhao
Abstract: Camouflaged object detection (COD) aims to accurately detect objects hidden in the surrounding environment. However, the existing COD methods mainly locate camouflaged objects in the RGB domain, their performance has not been fully exploited in many challenging scenarios. Considering that the features of the camouflaged object and the background are more discriminative in the frequency domain, we propose a novel learnable and separable frequency perception mechanism driven by the semantic hierarchy in the frequency domain. Our entire network adopts a two-stage model, including a frequency-guided coarse localization stage and a detail-preserving fine localization stage. With the multi-level features extracted by the backbone, we design a flexible frequency perception module based on octave convolution for coarse positioning. Then, we design the correction fusion module to step-by-step integrate the high-level features through the prior-guided correction and cross-layer feature channel association, and finally combine them with the shallow features to achieve the detailed correction of the camouflaged objects. Compared with the currently existing models, our proposed method achieves competitive performance in three popular benchmark datasets both qualitatively and quantitatively.
Authors: Runmin Cong, Hongyu Liu, Chen Zhang, Wei Zhang, Feng Zheng, Ran Song, Sam Kwong
Abstract: By integrating complementary information from RGB image and depth map, the ability of salient object detection (SOD) for complex and challenging scenes can be improved. In recent years, the important role of Convolutional Neural Networks (CNNs) in feature extraction and cross-modality interaction has been fully explored, but it is still insufficient in modeling global long-range dependencies of self-modality and cross-modality. To this end, we introduce CNNs-assisted Transformer architecture and propose a novel RGB-D SOD network with Point-aware Interaction and CNN-induced Refinement (PICR-Net). On the one hand, considering the prior correlation between RGB modality and depth modality, an attention-triggered cross-modality point-aware interaction (CmPI) module is designed to explore the feature interaction of different modalities with positional constraints. On the other hand, in order to alleviate the block effect and detail destruction problems brought by the Transformer naturally, we design a CNN-induced refinement (CNNR) unit for content refinement and supplementation. Extensive experiments on five RGB-D SOD datasets show that the proposed network achieves competitive results in both quantitative and qualitative comparisons.
Authors: Runmin Cong, Yuchen Guan, Jinpeng Chen, Wei Zhang, Yao Zhao, Sam Kwong
Abstract: Despite significant progress in shadow detection, current methods still struggle with the adverse impact of background color, which may lead to errors when shadows are present on complex backgrounds. Drawing inspiration from the human visual system, we treat the input shadow image as a composition of a background layer and a shadow layer, and design a Style-guided Dual-layer Disentanglement Network (SDDNet) to model these layers independently. To achieve this, we devise a Feature Separation and Recombination (FSR) module that decomposes multi-level features into shadow-related and background-related components by offering specialized supervision for each component, while preserving information integrity and avoiding redundancy through the reconstruction constraint. Moreover, we propose a Shadow Style Filter (SSF) module to guide the feature disentanglement by focusing on style differentiation and uniformization. With these two modules and our overall pipeline, our model effectively minimizes the detrimental effects of background color, yielding superior performance on three public datasets with a real-time inference speed of 32 FPS.
Authors: Arturo Miguel Russell Bernal, Walter Scheirer, Jane Cleland-Huang
Abstract: With the increasing reliance on small Unmanned Aerial Systems (sUAS) for Emergency Response Scenarios, such as Search and Rescue, the integration of computer vision capabilities has become a key factor in mission success. Nevertheless, computer vision performance for detecting humans severely degrades when shifting from ground to aerial views. Several aerial datasets have been created to mitigate this problem, however, none of them has specifically addressed the issue of occlusion, a critical component in Emergency Response Scenarios. Natural, Occluded, Multi-scale Aerial Dataset (NOMAD) presents a benchmark for human detection under occluded aerial views, with five different aerial distances and rich imagery variance. NOMAD is composed of 100 different Actors, all performing sequences of walking, laying and hiding. It includes 42,825 frames, extracted from 5.4k resolution videos, and manually annotated with a bounding box and a label describing 10 different visibility levels, categorized according to the percentage of the human body visible inside the bounding box. This allows computer vision models to be evaluated on their detection performance across different ranges of occlusion. NOMAD is designed to improve the effectiveness of aerial search and rescue and to enhance collaboration between sUAS and humans, by providing a new benchmark dataset for human detection under occluded aerial views. Full dataset can be found at: https://github.com/ArtRuss/NOMAD.
Authors: Rui Shao, Tianxing Wu, Ziwei Liu
Abstract: Since photorealistic faces can be readily generated by facial manipulation technologies nowadays, potential malicious abuse of these technologies has drawn great concerns. Numerous deepfake detection methods are thus proposed. However, existing methods only focus on detecting one-step facial manipulation. As the emergence of easy-accessible facial editing applications, people can easily manipulate facial components using multi-step operations in a sequential manner. This new threat requires us to detect a sequence of facial manipulations, which is vital for both detecting deepfake media and recovering original faces afterwards. Motivated by this observation, we emphasize the need and propose a novel research problem called Detecting Sequential DeepFake Manipulation (Seq-DeepFake). Unlike the existing deepfake detection task only demanding a binary label prediction, detecting Seq-DeepFake manipulation requires correctly predicting a sequential vector of facial manipulation operations. To support a large-scale investigation, we construct the first Seq-DeepFake dataset, where face images are manipulated sequentially with corresponding annotations of sequential facial manipulation vectors. Based on this new dataset, we cast detecting Seq-DeepFake manipulation as a specific image-to-sequence task and propose a concise yet effective Seq-DeepFake Transformer (SeqFakeFormer). To better reflect real-world deepfake data distributions, we further apply various perturbations on the original Seq-DeepFake dataset and construct the more challenging Sequential DeepFake dataset with perturbations (Seq-DeepFake-P). To exploit deeper correlation between images and sequences when facing Seq-DeepFake-P, a dedicated Seq-DeepFake Transformer with Image-Sequence Reasoning (SeqFakeFormer++) is devised, which builds stronger correspondence between image-sequence pairs for more robust Seq-DeepFake detection.
Authors: Gyuseong Lee, Wooseok Jang, Jinhyeon Kim, Jaewoo Jung, Seungryong Kim
Abstract: Learning robust vision models that perform well in out-of-distribution (OOD) situations is an important task for model deployment in real-world settings. Despite extensive research in this field, many proposed methods have only shown minor performance improvements compared to the simplest empirical risk minimization (ERM) approach, which was evaluated on a benchmark with a limited hyperparameter search space. Our focus in this study is on leveraging the knowledge of large pretrained models to improve handling of OOD scenarios and tackle domain generalization problems. However, prior research has revealed that naively fine-tuning a large pretrained model can impair OOD robustness. Thus, we employ parameter-efficient fine-tuning (PEFT) techniques to effectively preserve OOD robustness while working with large models. Our extensive experiments and analysis confirm that the most effective approaches involve ensembling diverse models and increasing the scale of pretraining. As a result, we achieve state-of-the-art performance in domain generalization tasks. Our code and project page are available at: https://cvlab-kaist.github.io/MoA
Authors: Yong Du, Jiahui Zhan, Xinzhe Li, Junyu Dong, Sheng Chen, Ming-Hsuan Yang, Shengfeng He
Abstract: In this paper, we propose a novel translation model, UniTranslator, for transforming representations between visually distinct domains under conditions of limited training data and significant visual differences. The main idea behind our approach is leveraging the domain-neutral capabilities of CLIP as a bridging mechanism, while utilizing a separate module to extract abstract, domain-agnostic semantics from the embeddings of both the source and target realms. Fusing these abstract semantics with target-specific semantics results in a transformed embedding within the CLIP space. To bridge the gap between the disparate worlds of CLIP and StyleGAN, we introduce a new non-linear mapper, the CLIP2P mapper. Utilizing CLIP embeddings, this module is tailored to approximate the latent distribution in the StyleGAN's latent space, effectively acting as a connector between these two spaces. The proposed UniTranslator is versatile and capable of performing various tasks, including style mixing, stylization, and translations, even in visually challenging scenarios across different visual domains. Notably, UniTranslator generates high-quality translations that showcase domain relevance, diversity, and improved image quality. UniTranslator surpasses the performance of existing general-purpose models and performs well against specialized models in representative tasks. The source code and trained models will be released to the public.
Authors: Rafi Ibn Sultan, Chengyin Li, Hui Zhu, Prashant Khanduri, Marco Brocanelli, Dongxiao Zhu
Abstract: In geographical image segmentation, performance is often constrained by the limited availability of training data and a lack of generalizability, particularly for segmenting mobility infrastructure such as roads, sidewalks, and crosswalks. Vision foundation models like the Segment Anything Model (SAM), pre-trained on millions of natural images, have demonstrated impressive zero-shot segmentation performance, providing a potential solution. However, SAM struggles with geographical images, such as aerial and satellite imagery, due to its training being confined to natural images and the narrow features and textures of these objects blending into their surroundings. To address these challenges, we propose Geographical SAM (GeoSAM), a SAM-based framework that fine-tunes SAM with automatically generated multi-modal prompts, combining point prompts from a pre-trained task-specific model as primary visual guidance and text prompts from a large language model as secondary semantic guidance to enhance model comprehension. GeoSAM outperforms existing approaches for mobility infrastructure segmentation in both familiar and completely unseen regions by at least 5\% in mIoU, representing a significant leap in leveraging foundation models to segment mobility infrastructure, including both road and pedestrian infrastructure in geographical images. The source code can be found in this GitHub Repository: https://github.com/rafiibnsultan/GeoSAM.
Authors: Bowen Jiang, Zhijun Zhuang, Shreyas S. Shivakumar, Camillo J. Taylor
Abstract: This work introduces an enhanced approach to generating scene graphs by incorporating both a relationship hierarchy and commonsense knowledge. Specifically, we begin by proposing a hierarchical relation head that exploits an informative hierarchical structure. It jointly predicts the relation super-category between object pairs in an image, along with detailed relations under each super-category. Following this, we implement a robust commonsense validation pipeline that harnesses foundation models to critique the results from the scene graph prediction system, removing nonsensical predicates even with a small language-only model. Extensive experiments on Visual Genome and OpenImage V6 datasets demonstrate that the proposed modules can be seamlessly integrated as plug-and-play enhancements to existing scene graph generation algorithms. The results show significant improvements with an extensive set of reasonable predictions beyond dataset annotations. Codes are available at https://github.com/bowen-upenn/scene_graph_commonsense.
URLs: https://github.com/bowen-upenn/scene_graph_commonsense.
Authors: Takehiko Ohkawa, Takuma Yagi, Taichi Nishimura, Ryosuke Furuta, Atsushi Hashimoto, Yoshitaka Ushiku, Yoichi Sato
Abstract: We propose a novel benchmark for cross-view knowledge transfer of dense video captioning, adapting models from web instructional videos with exocentric views to an egocentric view. While dense video captioning (predicting time segments and their captions) is primarily studied with exocentric videos (e.g., YouCook2), benchmarks with egocentric videos are restricted due to data scarcity. To overcome the limited video availability, transferring knowledge from abundant exocentric web videos is demanded as a practical approach. However, learning the correspondence between exocentric and egocentric views is difficult due to their dynamic view changes. The web videos contain shots showing either full-body or hand regions, while the egocentric view is constantly shifting. This necessitates the in-depth study of cross-view transfer under complex view changes. To this end, we first create a real-life egocentric dataset (EgoYC2) whose captions follow the definition of YouCook2 captions, enabling transfer learning between these datasets with access to their ground-truth. To bridge the view gaps, we propose a view-invariant learning method using adversarial training, which consists of pre-training and fine-tuning stages. Our experiments confirm the effectiveness of overcoming the view change problem and knowledge transfer to egocentric views. Our benchmark pushes the study of cross-view transfer into a new task domain of dense video captioning and envisions methodologies that describe egocentric videos in natural language.
Authors: Xuecheng Wu, Heli Sun, Junxiao Xue, Jiayu Nie, Xiangyan Kong, Ruofan Zhai, Liang He
Abstract: Nowadays, short-form videos (SVs) are essential to web information acquisition and sharing in our daily life. The prevailing use of SVs to spread emotions leads to the necessity of conducting video emotion analysis (VEA) towards SVs. Considering the lack of SVs emotion data, we introduce a large-scale dataset named eMotions, comprising 27,996 videos. Meanwhile, we alleviate the impact of subjectivities on labeling quality by emphasizing better personnel allocations and multi-stage annotations. In addition, we provide the category-balanced and test-oriented variants through targeted data sampling. Some commonly used videos, such as facial expressions, have been well studied. However, it is still challenging to analysis the emotions in SVs. Since the broader content diversity brings more distinct semantic gaps and difficulties in learning emotion-related features, and there exists local biases and collective information gaps caused by the emotion inconsistence under the prevalently audio-visual co-expressions. To tackle these challenges, we present an end-to-end audio-visual baseline AV-CANet which employs the video transformer to better learn semantically relevant representations. We further design the Local-Global Fusion Module to progressively capture the correlations of audio-visual features. The EP-CE Loss is then introduced to guide model optimization. Extensive experimental results on seven datasets demonstrate the effectiveness of AV-CANet, while providing broad insights for future works. Besides, we investigate the key components of AV-CANet by ablation studies. Datasets and code will be fully open soon.
Authors: Shuo Chen, Zhen Han, Bailan He, Jianzhe Liu, Mark Buckley, Yao Qin, Philip Torr, Volker Tresp, Jindong Gu
Abstract: Large Language Models (LLMs) with in-context learning (ICL) ability can quickly adapt to a specific context given a few demonstrations (demos). Recently, Multimodal Large Language Models (MLLMs) built upon LLMs have also shown multimodal ICL ability, i.e., responding to queries given a few multimodal demos, including images, queries, and answers. While ICL has been extensively studied on LLMs, its research on MLLMs remains limited. One essential question is whether these MLLMs can truly conduct multimodal ICL, or if only the textual modality is necessary. We investigate this question by examining two primary factors that influence ICL: 1) Demo content, i.e., understanding the influences of demo content in different modalities. 2) Demo selection strategy, i.e., how to select better multimodal demos for improved performance. Experiments revealed that multimodal ICL is predominantly driven by the textual content whereas the visual information in the demos has little influence. Interestingly, visual content is still necessary and useful for selecting demos to increase performance. Motivated by our analysis, we propose a simple yet effective approach, termed Mixed Modality In-Context Example Selection (MMICES), which considers both visual and language modalities when selecting demos. Extensive experiments are conducted to support our findings and verify the improvement brought by our method. Code is available at \url{https://chenxshuo.github.io/m-icl/}.
Authors: Juhyeon Park, Seokhyeon Jeong, Taesup Moon
Abstract: An image classifier may depend on incidental features stemming from a strong correlation between the feature and the classification target in the training dataset. Recently, Last Layer Retraining (LLR) with group-balanced datasets is shown to be efficient in mitigating the spurious correlation of classifiers. However, the acquisition of image-based group-balanced datasets is costly, which hinders the general applicability of the LLR method. In this work, we propose to perform LLR based on text datasets built with large language models to debias a general image classifier. To that end, we demonstrate that text can generally be a proxy for its corresponding image beyond the image-text joint embedding space, which is achieved with a linear projector that ensures orthogonality between its weight and the modality gap of the joint embedding space. In addition, we propose a systematic validation procedure that checks whether the generated words are compatible with the embedding space of CLIP and the image classifier, which is shown to be effective for improving debiasing performance. We dub these procedures as TLDR (Text-based Last layer retraining for Debiasing image classifieRs) and show our method achieves the performance that is competitive with the LLR methods that require group-balanced image dataset for retraining. Furthermore, TLDR outperforms other baselines that involve training the last layer without any group annotated dataset. Codes: https://github.com/beotborry/TLDR
Authors: Rohith Peddi, Shivvrat Arya, Bharath Challa, Likhitha Pallapothula, Akshay Vyas, Bhavya Gouripeddi, Jikai Wang, Qifan Zhang, Vasundhara Komaragiri, Eric Ragan, Nicholas Ruozzi, Yu Xiang, Vibhav Gogate
Abstract: Following step-by-step procedures is an essential component of various activities carried out by individuals in their daily lives. These procedures serve as a guiding framework that helps to achieve goals efficiently, whether it is assembling furniture or preparing a recipe. However, the complexity and duration of procedural activities inherently increase the likelihood of making errors. Understanding such procedural activities from a sequence of frames is a challenging task that demands an accurate interpretation of visual information and the ability to reason about the structure of the activity. To this end, we collect a new egocentric 4D dataset, CaptainCook4D, comprising 384 recordings (94.5 hours) of people performing recipes in real kitchen environments. This dataset consists of two distinct types of activity: one in which participants adhere to the provided recipe instructions and another in which they deviate and induce errors. We provide 5.3K step annotations and 10K fine-grained action annotations and benchmark the dataset for the following tasks: supervised error recognition, multistep localization, and procedure learning
Authors: Ahmed Ghita, Bj{\o}rk Antoniussen, Walter Zimmer, Ross Greer, Christian Cre{\ss}, Andreas M{\o}gelmose, Mohan M. Trivedi, Alois C. Knoll
Abstract: The curation of large-scale datasets is still costly and requires much time and resources. Data is often manually labeled, and the challenge of creating high-quality datasets remains. In this work, we fill the research gap using active learning for multi-modal 3D object detection. We propose ActiveAnno3D, an active learning framework to select data samples for labeling that are of maximum informativeness for training. We explore various continuous training methods and integrate the most efficient method regarding computational demand and detection performance. Furthermore, we perform extensive experiments and ablation studies with BEVFusion and PV-RCNN on the nuScenes and TUM Traffic Intersection dataset. We show that we can achieve almost the same performance with PV-RCNN and the entropy-based query strategy when using only half of the training data (77.25 mAP compared to 83.50 mAP) of the TUM Traffic Intersection dataset. BEVFusion achieved an mAP of 64.31 when using half of the training data and 75.0 mAP when using the complete nuScenes dataset. We integrate our active learning framework into the proAnno labeling tool to enable AI-assisted data selection and labeling and minimize the labeling costs. Finally, we provide code, weights, and visualization results on our website: https://active3d-framework.github.io/active3d-framework.
URLs: https://active3d-framework.github.io/active3d-framework.
Authors: Youngsik Yun, Jihie Kim
Abstract: Image Captioning generates descriptive sentences from images using Vision-Language Pre-trained models (VLPs) such as BLIP, which has improved greatly. However, current methods lack the generation of detailed descriptive captions for the cultural elements depicted in the images, such as the traditional clothing worn by people from Asian cultural groups. In this paper, we propose a new framework, Culturally-aware Image Captioning (CIC), that generates captions and describes cultural elements extracted from cultural visual elements in images representing cultures. Inspired by methods combining visual modality and Large Language Models (LLMs) through appropriate prompts, our framework (1) generates questions based on cultural categories from images, (2) extracts cultural visual elements from Visual Question Answering (VQA) using generated questions, and (3) generates culturally-aware captions using LLMs with the prompts. Our human evaluation conducted on 45 participants from 4 different cultural groups with a high understanding of the corresponding culture shows that our proposed framework generates more culturally descriptive captions when compared to the image captioning baseline based on VLPs. Resources can be found at https://shane3606.github.io/cic..
Authors: Blagoj Mitrevski, Arina Rak, Julian Schnitzler, Chengkun Li, Andrii Maksai, Jesse Berent, Claudiu Musat
Abstract: Digital note-taking is gaining popularity, offering a durable, editable, and easily indexable way of storing notes in a vectorized form, known as digital ink. However, a substantial gap remains between this way of note-taking and traditional pen-and-paper note-taking, a practice that is still favored by a vast majority. Our work InkSight, aims to bridge the gap by empowering physical note-takers to effortlessly convert their work (offline handwriting) to digital ink (online handwriting), a process we refer to as derendering. Prior research on the topic has focused on the geometric properties of images, resulting in limited generalization beyond their training domains. Our approach combines reading and writing priors, allowing training a model in the absence of large amounts of paired samples, which are difficult to obtain. To our knowledge, this is the first work that effectively derenders handwritten text in arbitrary photos with diverse visual characteristics and backgrounds. Furthermore, it generalizes beyond its training domain into simple sketches. Our human evaluation reveals that 87% of the samples produced by our model on the challenging HierText dataset are considered as a valid tracing of the input image and 67% look like a pen trajectory traced by a human.
Authors: Mai Gamal, Mohamed Rashad, Eman Ehab, Seif Eldawlatly, Mennatullah Siam
Abstract: Extensive literature has drawn comparisons between recordings of biological neurons in the brain and deep neural networks. This comparative analysis aims to advance and interpret deep neural networks and enhance our understanding of biological neural systems. However, previous works did not consider the time aspect and how the encoding of video and dynamics in deep networks relate to the biological neural systems within a large-scale comparison. Towards this end, we propose the first large-scale study focused on comparing video understanding models with respect to the visual cortex recordings using video stimuli. The study encompasses more than two million regression fits, examining image vs. video understanding, convolutional vs. transformer-based and fully vs. self-supervised models. Additionally, we propose a novel neural encoding scheme to better encode biological neural systems. We provide key insights on how video understanding models predict visual cortex responses; showing video understanding better than image understanding models, convolutional models are better in the early-mid visual cortical regions than transformer based ones except for multiscale transformers, and that two-stream models are better than single stream. Furthermore, we propose a novel neural encoding scheme that is built on top of the best performing video understanding models, while incorporating inter-intra region connectivity across the visual cortex. Our neural encoding leverages the encoded dynamics from video stimuli, through utilizing two-stream networks and multiscale transformers, while taking connectivity priors into consideration. Our results show that merging both intra and inter-region connectivity priors increases the encoding performance over each one of them standalone or no connectivity priors. It also shows the necessity for encoding dynamics to fully benefit from such connectivity priors.
Authors: Runmin Cong, Hang Xiong, Jinpeng Chen, Wei Zhang, Qingming Huang, Yao Zhao
Abstract: Previous Few-Shot Segmentation (FSS) approaches exclusively utilize support features for prototype generation, neglecting the specific requirements of the query. To address this, we present the Query-guided Prototype Evolution Network (QPENet), a new method that integrates query features into the generation process of foreground and background prototypes, thereby yielding customized prototypes attuned to specific queries. The evolution of the foreground prototype is accomplished through a \textit{support-query-support} iterative process involving two new modules: Pseudo-prototype Generation (PPG) and Dual Prototype Evolution (DPE). The PPG module employs support features to create an initial prototype for the preliminary segmentation of the query image, resulting in a pseudo-prototype reflecting the unique needs of the current query. Subsequently, the DPE module performs reverse segmentation on support images using this pseudo-prototype, leading to the generation of evolved prototypes, which can be considered as custom solutions. As for the background prototype, the evolution begins with a global background prototype that represents the generalized features of all training images. We also design a Global Background Cleansing (GBC) module to eliminate potential adverse components mirroring the characteristics of the current foreground class. Experimental results on the PASCAL-$5^i$ and COCO-$20^i$ datasets attest to the substantial enhancements achieved by QPENet over prevailing state-of-the-art techniques, underscoring the validity of our ideas.
Authors: Runmin Cong, Ronghui Sheng, Hao Wu, Yulan Guo, Yunchao Wei, Wangmeng Zuo, Yao Zhao, Sam Kwong
Abstract: Color information is the most commonly used prior knowledge for depth map super-resolution (DSR), which can provide high-frequency boundary guidance for detail restoration. However, its role and functionality in DSR have not been fully developed. In this paper, we rethink the utilization of color information and propose a hierarchical color guidance network to achieve DSR. On the one hand, the low-level detail embedding module is designed to supplement high-frequency color information of depth features in a residual mask manner at the low-level stages. On the other hand, the high-level abstract guidance module is proposed to maintain semantic consistency in the reconstruction process by using a semantic mask that encodes the global guidance information. The color information of these two dimensions plays a role in the front and back ends of the attention-based feature projection (AFP) module in a more comprehensive form. Simultaneously, the AFP module integrates the multi-scale content enhancement block and adaptive attention projection block to make full use of multi-scale information and adaptively project critical restoration information in an attention manner for DSR. Compared with the state-of-the-art methods on four benchmark datasets, our method achieves more competitive performance both qualitatively and quantitatively.
Authors: Yusuke Marumo, Kazuhiko Kawamoto, Satomi Tanaka, Shigenobu Hirano, Hiroshi Kera
Abstract: Not identical but similar objects are ubiquitous in our world, ranging from four-legged animals such as dogs and cats to cars of different models and flowers of various colors. This study addresses a novel task of matching such non-identical objects at the pixel level. We propose a weighting scheme of descriptors that incorporates semantic information from object detectors into existing sparse image matching methods, extending their targets from identical objects captured from different perspectives to semantically similar objects. The experiments show successful matching between non-identical objects in various cases, including in-class design variations, class discrepancy, and domain shifts (e.g., photo--drawing and image corruptions).
Authors: Lorenzo Agnolucci, Leonardo Galteri, Marco Bertini
Abstract: No-Reference Image Quality Assessment (NR-IQA) focuses on designing methods to measure image quality in alignment with human perception when a high-quality reference image is unavailable. The reliance on human-annotated Mean Opinion Score (MOS) in the majority of state-of-the-art NR-IQA approaches limits their scalability and broader applicability to real-world scenarios. To overcome this limitation, we propose QualiCLIP (Quality-aware CLIP), a CLIP-based self-supervised opinion-unaware method that does not require MOS. In particular, we introduce a quality-aware image-text alignment strategy to make CLIP generate quality-aware image representations. Starting from pristine images, we synthetically degrade them with increasing levels of intensity. Then, we train CLIP to rank these degraded images based on their similarity to quality-related antonym text prompts. At the same time, we force CLIP to generate consistent representations for images with similar content and the same level of degradation. Our method significantly outperforms other opinion-unaware approaches on several datasets with authentic distortions. Moreover, despite not requiring MOS, QualiCLIP achieves state-of-the-art performance even when compared with supervised methods in cross-dataset experiments, thus proving to be suitable for application in real-world scenarios. The code and the model are publicly available at https://github.com/miccunifi/QualiCLIP.
Authors: Yutong He, Alexander Robey, Naoki Murata, Yiding Jiang, Joshua Nathaniel Williams, George J. Pappas, Hamed Hassani, Yuki Mitsufuji, Ruslan Salakhutdinov, J. Zico Kolter
Abstract: Prompt engineering is effective for controlling the output of text-to-image (T2I) generative models, but it is also laborious due to the need for manually crafted prompts. This challenge has spurred the development of algorithms for automated prompt generation. However, these methods often struggle with transferability across T2I models, require white-box access to the underlying model, and produce non-intuitive prompts. In this work, we introduce PRISM, an algorithm that automatically identifies human-interpretable and transferable prompts that can effectively generate desired concepts given only black-box access to T2I models. Inspired by large language model (LLM) jailbreaking, PRISM leverages the in-context learning ability of LLMs to iteratively refine the candidate prompts distribution for given reference images. Our experiments demonstrate the versatility and effectiveness of PRISM in generating accurate prompts for objects, styles and images across multiple T2I models, including Stable Diffusion, DALL-E, and Midjourney.
Authors: Avinash Paliwal, Wei Ye, Jinhui Xiong, Dmytro Kotovenko, Rakesh Ranjan, Vikas Chandra, Nima Khademi Kalantari
Abstract: The field of 3D reconstruction from images has rapidly evolved in the past few years, first with the introduction of Neural Radiance Field (NeRF) and more recently with 3D Gaussian Splatting (3DGS). The latter provides a significant edge over NeRF in terms of the training and inference speed, as well as the reconstruction quality. Although 3DGS works well for dense input images, the unstructured point-cloud like representation quickly overfits to the more challenging setup of extremely sparse input images (e.g., 3 images), creating a representation that appears as a jumble of needles from novel views. To address this issue, we propose regularized optimization and depth-based initialization. Our key idea is to introduce a structured Gaussian representation that can be controlled in 2D image space. We then constraint the Gaussians, in particular their position, and prevent them from moving independently during optimization. Specifically, we introduce single and multiview constraints through an implicit convolutional decoder and a total variation loss, respectively. With the coherency introduced to the Gaussians, we further constrain the optimization through a flow-based loss function. To support our regularized optimization, we propose an approach to initialize the Gaussians using monocular depth estimates at each input view. We demonstrate significant improvements compared to the state-of-the-art sparse-view NeRF-based approaches on a variety of scenes.
Authors: SeungJeh Chung, JooHyun Park, HyeongYeop Kang
Abstract: 3D stylization, the application of specific styles to three-dimensional objects, offers substantial commercial potential by enabling the creation of uniquely styled 3D objects tailored to diverse scenes. Recent advancements in artificial intelligence and text-driven manipulation methods have made the stylization process increasingly intuitive and automated. While these methods reduce human costs by minimizing reliance on manual labor and expertise, they predominantly focus on holistic stylization, neglecting the application of desired styles to individual components of a 3D object. This limitation restricts the fine-grained controllability. To address this gap, we introduce 3DStyleGLIP, a novel framework specifically designed for text-driven, part-tailored 3D stylization. Given a 3D mesh and a text prompt, 3DStyleGLIP utilizes the vision-language embedding space of the Grounded Language-Image Pre-training (GLIP) model to localize individual parts of the 3D mesh and modify their appearance to match the styles specified in the text prompt. 3DStyleGLIP effectively integrates part localization and stylization guidance within GLIP's shared embedding space through an end-to-end process, enabled by part-level style loss and two complementary learning techniques. This neural methodology meets the user's need for fine-grained style editing and delivers high-quality part-specific stylization results, opening new possibilities for customization and flexibility in 3D content creation. Our code and results are available at https://github.com/sj978/3DStyleGLIP.
Authors: Anwesa Choudhuri, Girish Chowdhary, Alexander G. Schwing
Abstract: We propose the new task 'open-world video instance segmentation and captioning'. It requires to detect, segment, track and describe with rich captions never before seen objects. This challenging task can be addressed by developing "abstractors" which connect a vision model and a language foundation model. Concretely, we connect a multi-scale visual feature extractor and a large language model (LLM) by developing an object abstractor and an object-to-text abstractor. The object abstractor, consisting of a prompt encoder and transformer blocks, introduces spatially-diverse open-world object queries to discover never before seen objects in videos. An inter-query contrastive loss further encourages the diversity of object queries. The object-to-text abstractor is augmented with masked cross-attention and acts as a bridge between the object queries and a frozen LLM to generate rich and descriptive object-centric captions for each detected object. Our generalized approach surpasses the baseline that jointly addresses the tasks of open-world video instance segmentation and dense video object captioning by 13% on never before seen objects, and by 10% on object-centric captions.
Authors: Yifan Shen, Zhengyuan Li, Gang Wang
Abstract: Segment Anything Models (SAM) have made significant advancements in image segmentation, allowing users to segment target portions of an image with a single click (i.e., user prompt). Given its broad applications, the robustness of SAM against adversarial attacks is a critical concern. While recent works have explored adversarial attacks against a pre-defined prompt/click, their threat model is not yet realistic: (1) they often assume the user-click position is known to the attacker (point-based attack), and (2) they often operate under a white-box setting with limited transferability. In this paper, we propose a more practical region-level attack where attackers do not need to know the precise user prompt. The attack remains effective as the user clicks on any point on the target object in the image, hiding the object from SAM. Also, by adapting a spectrum transformation method, we make the attack more transferable under a black-box setting. Both control experiments and testing against real-world SAM services confirm its effectiveness.
Authors: Xi Wang, Yichen Peng, Heng Fang, Yilin Wang, Haoran Xie, Xi Yang, Chuntao Li
Abstract: In controllable generation tasks, flexibly manipulating the generated images to attain a desired appearance or structure based on a single input image cue remains a critical and longstanding challenge. Achieving this requires the effective decoupling of key attributes within the input image data to achieve representations accurately. Previous works have concentrated predominantly on disentangling image attributes within feature space. However, the complex distribution present in real-world data often makes the application of such decoupling algorithms to other datasets challenging. Moreover, the granularity of control over feature encoding frequently fails to meet specific task requirements. Upon scrutinizing the characteristics of various generative models, we have observed that the input sensitivity and dynamic evolution properties of the diffusion model can be effectively fused with the explicit decomposition operation in pixel space. This allows the operation that we design and use in pixel space to achieve the desired control effect on the specific representation in the generated results. Therefore, we propose FilterPrompt, an approach to enhance the effect of controllable generation. It can be universally applied to any diffusion model, allowing users to adjust the representation of specific image features in accordance with task requirements, thereby facilitating more precise and controllable generation outcomes. In particular, our designed experiments demonstrate that the FilterPrompt optimizes feature correlation, mitigates content conflicts during the generation process, and enhances the effect of controllable generation.
Authors: Panagiotis Sapoutzoglou, Georgios Giapitzakis, Georgios Floros, George Terzakis, Maria Pateraki
Abstract: We propose a generic procedure for assessing 6D object pose estimates. Our approach relies on the evaluation of discrepancies in the geometry of the observed object, in particular its respective estimated back-projection in 3D, against a putative functional shape representation comprising mixtures of Gaussian Processes, that act as a template. Each Gaussian Process is trained to yield a fragment of the object's surface in a radial fashion with respect to designated reference points. We further define a pose confidence measure as the average probability of pixel back-projections in the Gaussian mixture. The goal of our experiments is two-fold. a) We demonstrate that our functional representation is sufficiently accurate as a shape template on which the probability of back-projected object points can be evaluated, and, b) we show that the resulting confidence scores based on these probabilities are indeed a consistent quality measure of pose.
Authors: Colby Banbury, Emil Njor, Andrea Mattia Garavagno, Matthew Stewart, Pete Warden, Manjunath Kudlur, Nat Jeffries, Xenofon Fafoutis, Vijay Janapa Reddi
Abstract: Tiny machine learning (TinyML) for low-power devices lacks robust datasets for development. We present Wake Vision, a large-scale dataset for person detection that contains over 6 million quality-filtered images. We provide two variants: Wake Vision (Large) and Wake Vision (Quality), leveraging the large variant for pretraining and knowledge distillation, while the higher-quality labels drive final model performance. The manually labeled validation and test sets reduce error rates from 7.8% to 2.2% compared to previous standards. In addition, we introduce five detailed benchmark sets to evaluate model performance in real-world scenarios, including varying lighting, camera distances, and demographic characteristics. Training with Wake Vision improves accuracy by 1.93% over existing datasets, demonstrating the importance of dataset quality for low-capacity models and dataset size for high-capacity models. The dataset, benchmarks, code, and models are available under the CC-BY 4.0 license, maintained by the Edge AI Foundation.
Authors: Quang-Huy Che, Le-Chuong Nguyen, Duc-Tuan Luu, Vinh-Tiep Nguyen
Abstract: Person re-identification (Re-ID) is a challenging task that involves identifying the same person across different camera views in surveillance systems. Current methods usually rely on features from single-camera views, which can be limiting when dealing with multiple cameras and challenges such as changing viewpoints and occlusions. In this paper, a new approach is introduced that enhances the capability of ReID models through the Uncertain Feature Fusion Method (UFFM) and Auto-weighted Measure Combination (AMC). UFFM generates multi-view features using features extracted independently from multiple images to mitigate view bias. However, relying only on similarity based on multi-view features is limited because these features ignore the details represented in single-view features. Therefore, we propose the AMC method to generate a more robust similarity measure by combining various measures. Our method significantly improves Rank@1 accuracy and Mean Average Precision (mAP) when evaluated on person re-identification datasets. Combined with the BoT Baseline on challenging datasets, we achieve impressive results, with a 7.9% improvement in Rank@1 and a 12.1% improvement in mAP on the MSMT17 dataset. On the Occluded-DukeMTMC dataset, our method increases Rank@1 by 22.0% and mAP by 18.4%. Code is available: \url{https://github.com/chequanghuy/Enhancing-Person-Re-Identification-via-UFFM-and-AMC}
URLs: https://github.com/chequanghuy/Enhancing-Person-Re-Identification-via-UFFM-and-AMC
Authors: Yiheng Huang, Hui Yang, Chuanchen Luo, Yuxi Wang, Shibiao Xu, Zhaoxiang Zhang, Man Zhang, Junran Peng
Abstract: Thanks to the powerful generative capacity of diffusion models, recent years have witnessed rapid progress in human motion generation. Existing diffusion-based methods employ disparate network architectures and training strategies. The effect of the design of each component is still unclear. In addition, the iterative denoising process consumes considerable computational overhead, which is prohibitive for real-time scenarios such as virtual characters and humanoid robots. For this reason, we first conduct a comprehensive investigation into network architectures, training strategies, and inference processs. Based on the profound analysis, we tailor each component for efficient high-quality human motion generation. Despite the promising performance, the tailored model still suffers from foot skating which is an ubiquitous issue in diffusion-based solutions. To eliminate footskate, we identify foot-ground contact and correct foot motions along the denoising process. By organically combining these well-designed components together, we present StableMoFusion, a robust and efficient framework for human motion generation. Extensive experimental results show that our StableMoFusion performs favorably against current state-of-the-art methods. Project page: https://h-y1heng.github.io/StableMoFusion-page/
Authors: Lu Dong, Lipisha Chaudhary, Fei Xu, Xiao Wang, Mason Lary, Ifeoma Nwogu
Abstract: Achieving expressive 3D motion reconstruction and automatic generation for isolated sign words can be challenging, due to the lack of real-world 3D sign-word data, the complex nuances of signing motions, and the cross-modal understanding of sign language semantics. To address these challenges, we introduce SignAvatar, a framework capable of both word-level sign language reconstruction and generation. SignAvatar employs a transformer-based conditional variational autoencoder architecture, effectively establishing relationships across different semantic modalities. Additionally, this approach incorporates a curriculum learning strategy to enhance the model's robustness and generalization, resulting in more realistic motions. Furthermore, we contribute the ASL3DWord dataset, composed of 3D joint rotation data for the body, hands, and face, for unique sign words. We demonstrate the effectiveness of SignAvatar through extensive experiments, showcasing its superior reconstruction and automatic generation capabilities. The code and dataset are available on the project page.
Authors: Ahmad Bdeir, Niels Landwehr
Abstract: Hyperbolic deep learning has become a growing research direction in computer vision for the unique properties afforded by the alternate embedding space. The negative curvature and exponentially growing distance metric provide a natural framework for capturing hierarchical relationships between datapoints and allowing for finer separability between their embeddings. However, these methods are still computationally expensive and prone to instability, especially when attempting to learn the negative curvature that best suits the task and the data. Current Riemannian optimizers do not account for changes in the manifold which greatly harms performance and forces lower learning rates to minimize projection errors. Our paper focuses on curvature learning by introducing an improved schema for popular learning algorithms and providing a novel normalization approach to constrain embeddings within the variable representative radius of the manifold. Additionally, we introduce a novel formulation for Riemannian AdamW, and alternative hybrid encoder techniques and foundational formulations for current convolutional hyperbolic operations, greatly reducing the computational penalty of the hyperbolic embedding space. Our approach demonstrates consistent performance improvements across both direct classification and hierarchical metric learning tasks while allowing for larger hyperbolic models.
Authors: Yusuke Akamatsu, Terumi Umematsu, Hitoshi Imaoka, Shizuko Gomi, Hideo Tsurushima
Abstract: Daily monitoring of intra-personal facial changes associated with health and emotional conditions has great potential to be useful for medical, healthcare, and emotion recognition fields. However, the approach for capturing intra-personal facial changes is relatively unexplored due to the difficulty of collecting temporally changing face images. In this paper, we propose a facial representation learning method using synthetic images for comparing faces, called ComFace, which is designed to capture intra-personal facial changes. For effective representation learning, ComFace aims to acquire two feature representations, i.e., inter-personal facial differences and intra-personal facial changes. The key point of our method is the use of synthetic face images to overcome the limitations of collecting real intra-personal face images. Facial representations learned by ComFace are transferred to three extensive downstream tasks for comparing faces: estimating facial expression changes, weight changes, and age changes from two face images of the same individual. Our ComFace, trained using only synthetic data, achieves comparable to or better transfer performance than general pre-training and state-of-the-art representation learning methods trained using real images.
Authors: Xiangyu Chen, Zhenzhen Liu, Katie Z Luo, Siddhartha Datta, Adhitya Polavaram, Yan Wang, Yurong You, Boyi Li, Marco Pavone, Wei-Lun Chao, Mark Campbell, Bharath Hariharan, Kilian Q. Weinberger
Abstract: Ensuring robust 3D object detection and localization is crucial for many applications in robotics and autonomous driving. Recent models, however, face difficulties in maintaining high performance when applied to domains with differing sensor setups or geographic locations, often resulting in poor localization accuracy due to domain shift. To overcome this challenge, we introduce a novel diffusion-based box refinement approach. This method employs a domain-agnostic diffusion model, conditioned on the LiDAR points surrounding a coarse bounding box, to simultaneously refine the box's location, size, and orientation. We evaluate this approach under various domain adaptation settings, and our results reveal significant improvements across different datasets, object classes and detectors. Our PyTorch implementation is available at \href{https://github.com/cxy1997/DiffuBox}{https://github.com/cxy1997/DiffuBox}.
URLs: https://github.com/cxy1997/DiffuBox, https://github.com/cxy1997/DiffuBox
Authors: Shugo Yamashita, Masaaki Ikehara
Abstract: Removing rain degradations in images is recognized as a significant issue. In this field, deep learning-based approaches, such as Convolutional Neural Networks (CNNs) and Transformers, have succeeded. Recently, State Space Models (SSMs) have exhibited superior performance across various tasks in both natural language processing and image processing due to their ability to model long-range dependencies. This study introduces SSM to image deraining with deraining-specific enhancements and proposes a Deraining Frequency-Enhanced State Space Model (DFSSM). To effectively remove rain streaks, which produce high-intensity frequency components in specific directions, we employ frequency domain processing concurrently with SSM. Additionally, we develop a novel mixed-scale gated-convolutional block, which uses convolutions with multiple kernel sizes to capture various scale degradations effectively and integrates a gating mechanism to manage the flow of information. Finally, experiments on synthetic and real-world rainy image datasets show that our method surpasses state-of-the-art methods. Code is available at https://github.com/ShugoYamashita/DFSSM.
Authors: Chinedu Innocent Nwoye, Nicolas Padoy
Abstract: Accurate tool tracking is essential for the success of computer-assisted intervention. Previous efforts often modeled tool trajectories rigidly, overlooking the dynamic nature of surgical procedures, especially tracking scenarios like out-of-body and out-of-camera views. Addressing this limitation, the new CholecTrack20 dataset provides detailed labels that account for multiple tool trajectories in three perspectives: (1) intraoperative, (2) intracorporeal, and (3) visibility, representing the different types of temporal duration of tool tracks. These fine-grained labels enhance tracking flexibility but also increase the task complexity. Re-identifying tools after occlusion or re-insertion into the body remains challenging due to high visual similarity, especially among tools of the same category. This work recognizes the critical role of the tool operators in distinguishing tool track instances, especially those belonging to the same tool category. The operators' information are however not explicitly captured in surgical videos. We therefore propose SurgiTrack, a novel deep learning method that leverages YOLOv7 for precise tool detection and employs an attention mechanism to model the originating direction of the tools, as a proxy to their operators, for tool re-identification. To handle diverse tool trajectory perspectives, SurgiTrack employs a harmonizing bipartite matching graph, minimizing conflicts and ensuring accurate tool identity association. Experimental results on CholecTrack20 demonstrate SurgiTrack's effectiveness, outperforming baselines and state-of-the-art methods with real-time inference capability. This work sets a new standard in surgical tool tracking, providing dynamic trajectories for more adaptable and precise assistance in minimally invasive surgeries.
Authors: Nurit Spingarn-Eliezer, Tomer Michaeli
Abstract: We study the possibility of imitating the functionality of an image-to-image translation model by observing input-output pairs. We focus on cases where training the model from scratch is impossible, either because training data are unavailable or because the model architecture is unknown. This is the case, for example, with commercial models for biological applications. Since the development of these models requires large investments, their owners commonly keep them confidential, and reveal only a few input-output examples on the company's website or in an academic paper. Surprisingly, we find that even a single example typically suffices for learning to imitate the model's functionality, and that this can be achieved using a simple distillation approach. We present an extensive ablation study encompassing a wide variety of model architectures, datasets and tasks, to characterize the factors affecting vulnerability to functionality imitation, and provide a preliminary theoretical discussion on the reasons for this unwanted behavior.
Authors: Chun-Hung Wu, Shih-Hong Chen, Chih-Yao Hu, Hsin-Yu Wu, Kai-Hsin Chen, Yu-You Chen, Chih-Hai Su, Chih-Kuo Lee, Yu-Lun Liu
Abstract: This paper presents Deformable Neural Vessel Representations (DeNVeR), an unsupervised approach for vessel segmentation in X-ray angiography videos without annotated ground truth. DeNVeR utilizes optical flow and layer separation techniques, enhancing segmentation accuracy and adaptability through test-time training. Key contributions include a novel layer separation bootstrapping technique, a parallel vessel motion loss, and the integration of Eulerian motion fields for modeling complex vessel dynamics. A significant component of this research is the introduction of the XACV dataset, the first X-ray angiography coronary video dataset with high-quality, manually labeled segmentation ground truth. Extensive evaluations on both XACV and CADICA datasets demonstrate that DeNVeR outperforms current state-of-the-art methods in vessel segmentation accuracy and generalization capability while maintaining temporal coherency. See our project page for video results at https://kirito878.github.io/DeNVeR/.
Authors: Tero Karras, Miika Aittala, Tuomas Kynk\"a\"anniemi, Jaakko Lehtinen, Timo Aila, Samuli Laine
Abstract: The primary axes of interest in image-generating diffusion models are image quality, the amount of variation in the results, and how well the results align with a given condition, e.g., a class label or a text prompt. The popular classifier-free guidance approach uses an unconditional model to guide a conditional model, leading to simultaneously better prompt alignment and higher-quality images at the cost of reduced variation. These effects seem inherently entangled, and thus hard to control. We make the surprising observation that it is possible to obtain disentangled control over image quality without compromising the amount of variation by guiding generation using a smaller, less-trained version of the model itself rather than an unconditional model. This leads to significant improvements in ImageNet generation, setting record FIDs of 1.01 for 64x64 and 1.25 for 512x512, using publicly available networks. Furthermore, the method is also applicable to unconditional diffusion models, drastically improving their quality.
Authors: Zhongpai Gao, Benjamin Planche, Meng Zheng, Xiao Chen, Terrence Chen, Ziyan Wu
Abstract: Digitally reconstructed radiographs (DRRs) are simulated 2D X-ray images generated from 3D CT volumes, widely used in preoperative settings but limited in intraoperative applications due to computational bottlenecks, especially for accurate but heavy physics-based Monte Carlo methods. While analytical DRR renderers offer greater efficiency, they overlook anisotropic X-ray image formation phenomena, such as Compton scattering. We present a novel approach that marries realistic physics-inspired X-ray simulation with efficient, differentiable DRR generation using 3D Gaussian splatting (3DGS). Our direction-disentangled 3DGS (DDGS) method separates the radiosity contribution into isotropic and direction-dependent components, approximating complex anisotropic interactions without intricate runtime simulations. Additionally, we adapt the 3DGS initialization to account for tomography data properties, enhancing accuracy and efficiency. Our method outperforms state-of-the-art techniques in image accuracy. Furthermore, our DDGS shows promise for intraoperative applications and inverse problems such as pose registration, delivering superior registration accuracy and runtime performance compared to analytical DRR methods.
Authors: Ghjulia Sialelli, Torben Peters, Jan D. Wegner, Konrad Schindler
Abstract: Accurate estimates of Above Ground Biomass (AGB) are essential in addressing two of humanity's biggest challenges, climate change and biodiversity loss. Existing datasets for AGB estimation from satellite imagery are limited. Either they focus on specific, local regions at high resolution, or they offer global coverage at low resolution. There is a need for a machine learning-ready, globally representative, high-resolution benchmark. Our findings indicate significant variability in biomass estimates across different vegetation types, emphasizing the necessity for a dataset that accurately captures global diversity. To address these gaps, we introduce a comprehensive new dataset that is globally distributed, covers a range of vegetation types, and spans several years. This dataset combines AGB reference data from the GEDI mission with data from Sentinel-2 and PALSAR-2 imagery. Additionally, it includes pre-processed high-level features such as a dense canopy height map, an elevation map, and a land-cover classification map. We also produce a dense, high-resolution (10m) map of AGB predictions for the entire area covered by the dataset. Rigorously tested, our dataset is accompanied by several benchmark models and is publicly available. It can be easily accessed using a single line of code, offering a solid basis for efforts towards global AGB estimation. The GitHub repository github.com/ghjuliasialelli/AGBD serves as a one-stop shop for all code and data.
Authors: Ge Ya Luo, Zhi Hao Luo, Anthony Gosselin, Alexia Jolicoeur-Martineau, Christopher Pal
Abstract: Controllable video generation has attracted significant attention, largely due to advances in video diffusion models. In domains such as autonomous driving, it is essential to develop highly accurate predictions for object motions. This paper tackles a crucial challenge of how to exert precise control over object motion for realistic video synthesis. To accomplish this, we 1) control object movements using bounding boxes and extend this control to the renderings of 2D or 3D boxes in pixel space, 2) employ a distinct, specialized model to forecast the trajectories of object bounding boxes based on their previous and, if desired, future positions, and 3) adapt and enhance a separate video diffusion network to create video content based on these high quality trajectory forecasts. Our method, Ctrl-V, leverages modified and fine-tuned Stable Video Diffusion (SVD) models to solve both trajectory and video generation. Extensive experiments conducted on the KITTI, Virtual-KITTI 2, BDD100k, and nuScenes datasets validate the effectiveness of our approach in producing realistic and controllable video generation.
Authors: Jiangshan Wang, Yue Ma, Jiayi Guo, Yicheng Xiao, Gao Huang, Xiu Li
Abstract: Video editing is an emerging task, in which most current methods adopt the pre-trained text-to-image (T2I) diffusion model to edit the source video in a zero-shot manner. Despite extensive efforts, maintaining the temporal consistency of edited videos remains challenging due to the lack of temporal constraints in the regular T2I diffusion model. To address this issue, we propose COrrespondence-guided Video Editing (COVE), leveraging the inherent diffusion feature correspondence to achieve high-quality and consistent video editing. Specifically, we propose an efficient sliding-window-based strategy to calculate the similarity among tokens in the diffusion features of source videos, identifying the tokens with high correspondence across frames. During the inversion and denoising process, we sample the tokens in noisy latent based on the correspondence and then perform self-attention within them. To save GPU memory usage and accelerate the editing process, we further introduce the temporal-dimensional token merging strategy, which can effectively reduce redundancy. COVE can be seamlessly integrated into the pre-trained T2I diffusion model without the need for extra training or optimization. Extensive experiment results demonstrate that COVE achieves the start-of-the-art performance in various video editing scenarios, outperforming existing methods both quantitatively and qualitatively. The code will be release at https://github.com/wangjiangshan0725/COVE.
Authors: Yunsong Zhou, Michael Simon, Zhenghao Peng, Sicheng Mo, Hongzi Zhu, Minyi Guo, Bolei Zhou
Abstract: Controllable synthetic data generation can substantially lower the annotation cost of training data. Prior works use diffusion models to generate driving images conditioned on the 3D object layout. However, those models are trained on small-scale datasets like nuScenes, which lack appearance and layout diversity. Moreover, overfitting often happens, where the trained models can only generate images based on the layout data from the validation set of the same dataset. In this work, we introduce a simulator-conditioned scene generation framework called SimGen that can learn to generate diverse driving scenes by mixing data from the simulator and the real world. It uses a novel cascade diffusion pipeline to address challenging sim-to-real gaps and multi-condition conflicts. A driving video dataset DIVA is collected to enhance the generative diversity of SimGen, which contains over 147.5 hours of real-world driving videos from 73 locations worldwide and simulated driving data from the MetaDrive simulator. SimGen achieves superior generation quality and diversity while preserving controllability based on the text prompt and the layout pulled from a simulator. We further demonstrate the improvements brought by SimGen for synthetic data augmentation on the BEV detection and segmentation task and showcase its capability in safety-critical data generation.
Authors: Sheng-Yu Wang, Aaron Hertzmann, Alexei A. Efros, Jun-Yan Zhu, Richard Zhang
Abstract: The goal of data attribution for text-to-image models is to identify the training images that most influence the generation of a new image. Influence is defined such that, for a given output, if a model is retrained from scratch without the most influential images, the model would fail to reproduce the same output. Unfortunately, directly searching for these influential images is computationally infeasible, since it would require repeatedly retraining models from scratch. In our work, we propose an efficient data attribution method by simulating unlearning the synthesized image. We achieve this by increasing the training loss on the output image, without catastrophic forgetting of other, unrelated concepts. We then identify training images with significant loss deviations after the unlearning process and label these as influential. We evaluate our method with a computationally intensive but "gold-standard" retraining from scratch and demonstrate our method's advantages over previous methods.
Authors: Junha Hyung, Susung Hong, Sungwon Hwang, Jaeseong Lee, Jaegul Choo, Jin-Hwa Kim
Abstract: 3D reconstruction from multi-view images is one of the fundamental challenges in computer vision and graphics. Recently, 3D Gaussian Splatting (3DGS) has emerged as a promising technique capable of real-time rendering with high-quality 3D reconstruction. This method utilizes 3D Gaussian representation and tile-based splatting techniques, bypassing the expensive neural field querying. Despite its potential, 3DGS encounters challenges such as needle-like artifacts, suboptimal geometries, and inaccurate normals caused by the Gaussians converging into anisotropic shapes with one dominant variance. We propose using the effective rank analysis to examine the shape statistics of 3D Gaussian primitives, and identify the Gaussians indeed converge into needle-like shapes with the effective rank 1. To address this, we introduce the effective rank as a regularization, which constrains the structure of the Gaussians. Our new regularization method enhances normal and geometry reconstruction while reducing needle-like artifacts. The approach can be integrated as an add-on module to other 3DGS variants, improving their quality without compromising visual fidelity. The project page is available at https://junhahyung.github.io/erankgs.github.io.
Authors: Jiarui Xing, Nivetha Jayakumar, Nian Wu, Yu Wang, Frederick H. Epstein, Miaomiao Zhang
Abstract: Motion and deformation analysis of cardiac magnetic resonance (CMR) imaging videos is crucial for assessing myocardial strain of patients with abnormal heart functions. Recent advances in deep learning-based image registration algorithms have shown promising results in predicting motion fields from routinely acquired CMR sequences. However, their accuracy often diminishes in regions with subtle appearance changes, with errors propagating over time. Advanced imaging techniques, such as displacement encoding with stimulated echoes (DENSE) CMR, offer highly accurate and reproducible motion data but require additional image acquisition, which poses challenges in busy clinical flows. In this paper, we introduce a novel Latent Motion Diffusion model (LaMoD) to predict highly accurate DENSE motions from standard CMR videos. More specifically, our method first employs an encoder from a pre-trained registration network that learns latent motion features (also considered as deformation-based shape features) from image sequences. Supervised by the ground-truth motion provided by DENSE, LaMoD then leverages a probabilistic latent diffusion model to reconstruct accurate motion from these extracted features. Experimental results demonstrate that our proposed method, LaMoD, significantly improves the accuracy of motion analysis in standard CMR images; hence improving myocardial strain analysis in clinical settings for cardiac patients. Our code is publicly available at https://github.com/jr-xing/LaMoD.
Authors: Jungho Lee, Donghyeong Kim, Dogyoon Lee, Suhwan Cho, Minhyeok Lee, Sangyoun Lee
Abstract: 3D Gaussian Splatting (3DGS) has gained significant attention for their high-quality novel view rendering, motivating research to address real-world challenges. A critical issue is the camera motion blur caused by movement during exposure, which hinders accurate 3D scene reconstruction. In this study, we propose CRiM-GS, a \textbf{C}ontinuous \textbf{Ri}gid \textbf{M}otion-aware \textbf{G}aussian \textbf{S}platting that reconstructs precise 3D scenes from motion-blurred images while maintaining real-time rendering speed. Considering the complex motion patterns inherent in real-world camera movements, we predict continuous camera trajectories using neural ordinary differential equations (ODE). To ensure accurate modeling, we employ rigid body transformations with proper regularization, preserving object shape and size. Additionally, we introduce an adaptive distortion-aware transformation to compensate for potential nonlinear distortions, such as rolling shutter effects, and unpredictable camera movements. By revisiting fundamental camera theory and leveraging advanced neural training techniques, we achieve precise modeling of continuous camera trajectories. Extensive experiments demonstrate state-of-the-art performance both quantitatively and qualitatively on benchmark datasets.
Authors: Zehui Zhao, Laith Alzubaidi, Jinglan Zhang, Ye Duan, Usman Naseem, Yuantong Gu
Abstract: Recently, transfer learning and self-supervised learning have gained significant attention within the medical field due to their ability to mitigate the challenges posed by limited data availability, improve model generalisation, and reduce computational expenses. Transfer learning and self-supervised learning hold immense potential for advancing medical research. However, it is crucial to recognise that transfer learning and self-supervised learning architectures exhibit distinct advantages and limitations, manifesting variations in accuracy, training speed, and robustness. This paper compares the performance and robustness of transfer learning and self-supervised learning in the medical field. Specifically, we pre-trained two models using the same source domain datasets with different pre-training methods and evaluated them on small-sized medical datasets to identify the factors influencing their final performance. We tested data with several common issues in medical domains, such as data imbalance, data scarcity, and domain mismatch, through comparison experiments to understand their impact on specific pre-trained models. Finally, we provide recommendations to help users apply transfer learning and self-supervised learning methods in medical areas, and build more convenient and efficient deployment strategies.
Authors: Rong Ma, Jie Chen, Xiangyang Xue, Jian Pu
Abstract: Deep supervised models possess significant capability to assimilate extensive training data, thereby presenting an opportunity to enhance model performance through training on multiple datasets. However, conflicts arising from different label spaces among datasets may adversely affect model performance. In this paper, we propose a novel approach to automatically construct a unified label space across multiple datasets using graph neural networks. This enables semantic segmentation models to be trained simultaneously on multiple datasets, resulting in performance improvements. Unlike existing methods, our approach facilitates seamless training without the need for additional manual reannotation or taxonomy reconciliation. This significantly enhances the efficiency and effectiveness of multi-dataset segmentation model training. The results demonstrate that our method significantly outperforms other multi-dataset training methods when trained on seven datasets simultaneously, and achieves state-of-the-art performance on the WildDash 2 benchmark.
Authors: Shuaixian Wang, Haoran Xu, Yaokun Li, Jiwei Chen, Guang Tan
Abstract: We present a novel approach for synthesizing realistic novel views using Neural Radiance Fields (NeRF) with uncontrolled photos in the wild. While NeRF has shown impressive results in controlled settings, it struggles with transient objects commonly found in dynamic and time-varying scenes. Our framework called \textit{Inpainting Enhanced NeRF}, or \ours, enhances the conventional NeRF by drawing inspiration from the technique of image inpainting. Specifically, our approach extends the Multi-Layer Perceptrons (MLP) of NeRF, enabling it to simultaneously generate intrinsic properties (static color, density) and extrinsic transient masks. We introduce an inpainting module that leverages the transient masks to effectively exclude occlusions, resulting in improved volume rendering quality. Additionally, we propose a new training strategy with frequency regularization to address the sparsity issue of low-frequency transient components. We evaluate our approach on internet photo collections of landmarks, demonstrating its ability to generate high-quality novel views and achieve state-of-the-art performance.
Authors: Guoliang You, Xiaomeng Chu, Yifan Duan, Wenyu Zhang, Xingchen Li, Sha Zhang, Yao Li, Jianmin Ji, Yanyong Zhang
Abstract: When planning for autonomous driving, it is crucial to consider essential traffic elements such as lanes, intersections, traffic regulations, and dynamic agents. However, they are often overlooked by the traditional end-to-end planning methods, likely leading to inefficiencies and non-compliance with traffic regulations. In this work, we endeavor to integrate the perception of these elements into the planning task. To this end, we propose Perception Helps Planning (PHP), a novel framework that reconciles lane-level planning with perception. This integration ensures that planning is inherently aligned with traffic constraints, thus facilitating safe and efficient driving. Specifically, PHP focuses on both edges of a lane for planning and perception purposes, taking into consideration the 3D positions of both lane edges and attributes for lane intersections, lane directions, lane occupancy, and planning. In the algorithmic design, the process begins with the transformer encoding multi-camera images to extract the above features and predicting lane-level perception results. Next, the hierarchical feature early fusion module refines the features for predicting planning attributes. Finally, the double-edge interpreter utilizes a late-fusion process specifically designed to integrate lane-level perception and planning information, culminating in the generation of vehicle control signals. Experiments on three Carla benchmarks show significant improvements in driving score of 27.20%, 33.47%, and 15.54% over existing algorithms, respectively, achieving the state-of-the-art performance, with the system operating up to 22.57 FPS.
Authors: Vasco Ramos, Yonatan Bitton, Michal Yarom, Idan Szpektor, Joao Magalhaes
Abstract: Generated video scenes for action-centric sequence descriptions, such as recipe instructions and do-it-yourself projects, often include non-linear patterns, where the next video may need to be visually consistent not with the immediately preceding video but with earlier ones. Current multi-scene video synthesis approaches fail to meet these consistency requirements. To address this, we propose a contrastive sequential video diffusion method that selects the most suitable previously generated scene to guide and condition the denoising process of the next scene. The result is a multi-scene video that is grounded in the scene descriptions and coherent w.r.t. the scenes that require visual consistency. Experiments with action-centered data from the real world demonstrate the practicality and improved consistency of our model compared to previous work.
Authors: Robin Hesse, Simone Schaub-Meyer, Stefan Roth
Abstract: Attribution maps are one of the most established tools to explain the functioning of computer vision models. They assign importance scores to input features, indicating how relevant each feature is for the prediction of a deep neural network. While much research has gone into proposing new attribution methods, their proper evaluation remains a difficult challenge. In this work, we propose a novel evaluation protocol that overcomes two fundamental limitations of the widely used incremental-deletion protocol, i.e., the out-of-domain issue and lacking inter-model comparisons. This allows us to evaluate 23 attribution methods and how different design choices of popular vision backbones affect their attribution quality. We find that intrinsically explainable models outperform standard models and that raw attribution values exhibit a higher attribution quality than what is known from previous work. Further, we show consistent changes in the attribution quality when varying the network design, indicating that some standard design choices promote attribution quality.
Authors: Zhongpai Gao, Abhishek Sharma, Meng Zheng, Benjamin Planche, Terrence Chen, Ziyan Wu
Abstract: Positioning patients for scanning and interventional procedures is a critical task that requires high precision and accuracy. The conventional workflow involves manually adjusting the patient support to align the center of the target body part with the laser projector or other guiding devices. This process is not only time-consuming but also prone to inaccuracies. In this work, we propose an automated patient positioning system that utilizes a camera to detect specific hand gestures from technicians, allowing users to indicate the target patient region to the system and initiate automated positioning. Our approach relies on a novel multi-stage pipeline to recognize and interpret the technicians' gestures, translating them into precise motions of medical devices. We evaluate our proposed pipeline during actual MRI scanning procedures, using RGB-Depth cameras to capture the process. Results show that our system achieves accurate and precise patient positioning with minimal technician intervention. Furthermore, we validate our method on HaGRID, a large-scale hand gesture dataset, demonstrating its effectiveness in hand detection and gesture recognition.
Authors: Jingtai He, Gehao Zhang, Tingting Liu, Songlin Du
Abstract: Recent advancements in keypoint detection and descriptor extraction have shown impressive performance in local feature learning tasks. However, existing methods generally exhibit suboptimal performance under extreme conditions such as significant appearance changes and domain shifts. In this study, we introduce a multi-level feature aggregation network that incorporates two pivotal components to facilitate the learning of robust and accurate features with domain adaptation. First, we employ domain adaptation supervision to align high-level feature distributions across different domains to achieve invariant domain representations. Second, we propose a Transformer-based booster that enhances descriptor robustness by integrating visual and geometric information through wave position encoding concepts, effectively handling complex conditions. To ensure the accuracy and robustness of features, we adopt a hierarchical architecture to capture comprehensive information and apply meticulous targeted supervision to keypoint detection, descriptor extraction, and their coupled processing. Extensive experiments demonstrate that our method, RADA, achieves excellent results in image matching, camera pose estimation, and visual localization tasks.
Authors: Shougao Zhang, Mengqi Zhou, Yuxi Wang, Chuanchen Luo, Rongyu Wang, Yiwei Li, Zhaoxiang Zhang, Junran Peng
Abstract: Urban areas, as the primary human habitat in modern civilization, accommodate a broad spectrum of social activities. With the surge of embodied intelligence, recent years have witnessed an increasing presence of physical agents in urban areas, such as autonomous vehicles and delivery robots. As a result, practitioners significantly value crafting authentic, simulation-ready 3D cities to facilitate the training and verification of such agents. However, this task is quite challenging. Current generative methods fall short in either diversity, controllability, or fidelity. In this work, we resort to the procedural content generation (PCG) technique for high-fidelity generation. It assembles superior assets according to empirical rules, ultimately leading to industrial-grade outcomes. To ensure diverse and self contained creation, we design a management protocol to accommodate extensive PCG plugins with distinct functions and interfaces. Based on this unified PCG library, we develop a multi-agent framework to transform multi-modal instructions, including OSM, semantic maps, and satellite images, into executable programs. The programs coordinate relevant plugins to construct the 3D city consistent with the control condition. A visual feedback scheme is introduced to further refine the initial outcomes. Our method, named CityX, demonstrates its superiority in creating diverse, controllable, and realistic 3D urban scenes. The synthetic scenes can be seamlessly deployed as a real-time simulator and an infinite data generator for embodied intelligence research. Our project page: https://cityx-lab.github.io.
Authors: Ezequiel Perez-Zarate, Oscar Ramos-Soto, Chunxiao Liu, Diego Oliva, Marco Perez-Cisneros
Abstract: Low-light image enhancement is vital for improving the visibility and quality of images captured under suboptimal lighting conditions. Traditional methods often fail to adequately capture local lighting variations and enhance both textural and chromatic details. Recent deep learning-based approaches, while effective, still struggle with generalization across diverse datasets, leading to noise amplification and unnatural color saturation. To address these challenges, the Adaptive Light Enhancement Network (ALEN) is introduced, a novel method that utilizes a classification mechanism to determine whether local or global illumination enhancement is required. ALEN integrates the Swin Light-Classification Transformer (SLCformer) for illuminance categorization, complemented by the Single-Channel Network (SCNet) and Multi-Channel Network (MCNet) for precise illumination and color estimation, respectively. Extensive experiments on publicly available datasets demonstrate ALEN's robust generalization capabilities, outperforming state-of-the-art methods in both quantitative metrics and qualitative assessments. Furthermore, ALEN not only enhances image quality but also improves the performance of high-level vision tasks such as semantic segmentation, showcasing its broader applicability and potential impact. The code for this method and the datasets are available at https://github.com/xingyumex/ALEN}{https://github.com/xingyumex/ALEN
URLs: https://github.com/xingyumex/ALEN, https://github.com/xingyumex/ALEN
Authors: Chen Liu, Can Han, Chengfeng Zhou, Crystal Cai, Dahong Qian
Abstract: Gesture recognition based on surface electromyography (sEMG) has achieved significant progress in human-machine interaction (HMI), especially in prosthetic control and movement rehabilitation. However, accurately recognizing predefined gestures within a closed set is still inadequate in practice; a robust open-set system needs to effectively reject unknown gestures while correctly classifying known ones, which is rarely explored in the field of myoelectric gesture recognition. To handle this challenge, we first report a significant distinction in prediction inconsistency discovered for unknown classes, which arises from different perspectives and can substantially enhance open-set recognition performance. Based on this insight, we propose a novel dual-perspective inconsistency learning approach, PredIN, to explicitly magnify the prediction inconsistency by enhancing the inconsistency of class feature distribution within different perspectives. Specifically, PredIN maximizes the class feature distribution inconsistency among the dual perspectives to enhance their differences. Meanwhile, it optimizes inter-class separability within an individual perspective to maintain individual performance. Comprehensive experiments on various benchmark datasets demonstrate that the PredIN outperforms state-of-the-art methods by a clear margin. Our proposed method simultaneously achieves accurate closed-set classification for predefined gestures and effective rejection for unknown gestures, exhibiting its efficacy and superiority in open-set gesture recognition based on sEMG.
Authors: Abhishek Vivekanandan, J. Marius Z\"ollner
Abstract: Representing diverse and plausible future trajectories is critical for motion forecasting in autonomous driving. However, efficiently capturing these trajectories in a compact set remains challenging. This study introduces a novel approach for generating scene-specific trajectory sets tailored to different contexts, such as intersections and straight roads, by leveraging map information and actor dynamics. A deterministic goal sampling algorithm identifies relevant map regions, while our Recursive In-Distribution Subsampling (RIDS) method enhances trajectory plausibility by condensing redundant representations. Experiments on the Argoverse 2 dataset demonstrate that our method achieves up to a 10% improvement in Driving Area Compliance (DAC) compared to baseline methods while maintaining competitive displacement errors. Our work highlights the benefits of mining such scene-aware trajectory sets and how they could capture the complex and heterogeneous nature of actor behavior in real-world driving scenarios.
Authors: Wenzhe Tian, Haijin Zeng, Yin-Ping Zhao, Yongyong Chen, Zhen Wang, Xuelong Li
Abstract: Snapshot Compressive Imaging (SCI) enables fast spectral imaging but requires effective decoding algorithms for hyperspectral image (HSI) reconstruction from compressed measurements. Current CNN-based methods are limited in modeling long-range dependencies, while Transformer-based models face high computational complexity. Although recent Mamba models outperform CNNs and Transformers in RGB tasks concerning computational efficiency or accuracy, they are not specifically optimized to fully leverage the local spatial and spectral correlations inherent in HSIs. To address this, we propose the Cross-Scanning Mamba, named CS-Mamba, that employs a Spatial-Spectral SSM for global-local balanced context encoding and cross-channel interaction promotion. Besides, while current reconstruction algorithms perform increasingly well in simulation scenarios, they exhibit suboptimal performance on real data due to limited generalization capability. During the training process, the model may not capture the inherent features of the images but rather learn the parameters to mitigate specific noise and loss, which may lead to a decline in reconstruction quality when faced with real scenes. To overcome this challenge, we propose a masked training method to enhance the generalization ability of models. Experiment results show that our CS-Mamba achieves state-of-the-art performance and the masked training method can better reconstruct smooth features to improve the visual quality.
Authors: Joanna Kaleta, Kacper Kania, Tomasz Trzcinski, Marek Kowalski
Abstract: Decoupling lighting from geometry using unconstrained photo collections is notoriously challenging. Solving it would benefit many users as creating complex 3D assets takes days of manual labor. Many previous works have attempted to address this issue, often at the expense of output fidelity, which questions the practicality of such methods. We introduce LumiGauss - a technique that tackles 3D reconstruction of scenes and environmental lighting through 2D Gaussian Splatting. Our approach yields high-quality scene reconstructions and enables realistic lighting synthesis under novel environment maps. We also propose a method for enhancing the quality of shadows, common in outdoor scenes, by exploiting spherical harmonics properties. Our approach facilitates seamless integration with game engines and enables the use of fast precomputed radiance transfer. We validate our method on the NeRF-OSR dataset, demonstrating superior performance over baseline methods. Moreover, LumiGauss can synthesize realistic images for unseen environment maps. Our code: https://github.com/joaxkal/lumigauss.
Authors: Yuxiao Wang, Qiwei Xiong, Yu Lei, Weiying Xue, Qi Liu, Zhenao Wei
Abstract: Human-object interaction (HOI) detection plays a key role in high-level visual understanding, facilitating a deep comprehension of human activities. Specifically, HOI detection aims to locate the humans and objects involved in interactions within images or videos and classify the specific interactions between them. The success of this task is influenced by several key factors, including the accurate localization of human and object instances, as well as the correct classification of object categories and interaction relationships. This paper systematically summarizes and discusses the recent work in image-based HOI detection. First, the mainstream datasets involved in HOI relationship detection are introduced. Furthermore, starting with two-stage methods and end-to-end one-stage detection approaches, this paper comprehensively discusses the current developments in image-based HOI detection, analyzing the strengths and weaknesses of these two methods. Additionally, the advancements of zero-shot learning, weakly supervised learning, and the application of large-scale language models in HOI detection are discussed. Finally, the current challenges in HOI detection are outlined, and potential research directions and future trends are explored.
Authors: Yuhao Chen, Jiangpeng He, Gautham Vinod, Siddeshwar Raghavan, Chris Czarnecki, Jinge Ma, Talha Ibn Mahmud, Bruce Coburn, Dayou Mao, Saeejith Nair, Pengcheng Xi, Alexander Wong, Edward Delp, Fengqing Zhu
Abstract: Food computing is both important and challenging in computer vision (CV). It significantly contributes to the development of CV algorithms due to its frequent presence in datasets across various applications, ranging from classification and instance segmentation to 3D reconstruction. The polymorphic shapes and textures of food, coupled with high variation in forms and vast multimodal information, including language descriptions and nutritional data, make food computing a complex and demanding task for modern CV algorithms. 3D food modeling is a new frontier for addressing food related problems, due to its inherent capability to deal with random camera views and its straightforward representation for calculating food portion size. However, the primary hurdle in the development of algorithms for food object analysis is the lack of nutrition values in existing 3D datasets. Moreover, in the broader field of 3D research, there is a critical need for domain-specific test datasets. To bridge the gap between general 3D vision and food computing research, we introduce MetaFood3D. This dataset consists of 743 meticulously scanned and labeled 3D food objects across 131 categories, featuring detailed nutrition information, weight, and food codes linked to a comprehensive nutrition database. Our MetaFood3D dataset emphasizes intra-class diversity and includes rich modalities such as textured mesh files, RGB-D videos, and segmentation masks. Experimental results demonstrate our dataset's strong capabilities in enhancing food portion estimation algorithms, highlight the gap between video captures and 3D scanned data, and showcase the strengths of MetaFood3D in generating synthetic eating occasion data and 3D food objects.
Authors: Tianlong Xu, Yi-Fan Zhang, Zhendong Chu, Shen Wang, Qingsong Wen
Abstract: Students frequently make mistakes while solving mathematical problems, and traditional error correction methods are both time-consuming and labor-intensive. This paper introduces an innovative \textbf{V}irtual \textbf{A}I \textbf{T}eacher system designed to autonomously analyze and correct student \textbf{E}rrors (VATE). Leveraging advanced large language models (LLMs), the system uses student drafts as a primary source for error analysis, which enhances understanding of the student's learning process. It incorporates sophisticated prompt engineering and maintains an error pool to reduce computational overhead. The AI-driven system also features a real-time dialogue component for efficient student interaction. Our approach demonstrates significant advantages over traditional and machine learning-based error correction methods, including reduced educational costs, high scalability, and superior generalizability. The system has been deployed on the Squirrel AI learning platform for elementary mathematics education, where it achieves 78.3\% accuracy in error analysis and shows a marked improvement in student learning efficiency. Satisfaction surveys indicate a strong positive reception, highlighting the system's potential to transform educational practices.
Authors: Ivan DeAndres-Tame, Muhammad Faisal, Ruben Tolosana, Rouqaiah Al-Refai, Ruben Vera-Rodriguez, Philipp Terh\"orst
Abstract: Face Recognition (FR) has advanced significantly with the development of deep learning, achieving high accuracy in several applications. However, the lack of interpretability of these systems raises concerns about their accountability, fairness, and reliability. In the present study, we propose an interactive framework to enhance the explainability of FR models by combining model-agnostic Explainable Artificial Intelligence (XAI) and Natural Language Processing (NLP) techniques. The proposed framework is able to accurately answer various questions of the user through an interactive chatbot. In particular, the explanations generated by our proposed method are in the form of natural language text and visual representations, which for example can describe how different facial regions contribute to the similarity measure between two faces. This is achieved through the automatic analysis of the output's saliency heatmaps of the face images and a BERT question-answering model, providing users with an interface that facilitates a comprehensive understanding of the FR decisions. The proposed approach is interactive, allowing the users to ask questions to get more precise information based on the user's background knowledge. More importantly, in contrast to previous studies, our solution does not decrease the face recognition performance. We demonstrate the effectiveness of the method through different experiments, highlighting its potential to make FR systems more interpretable and user-friendly, especially in sensitive applications where decision-making transparency is crucial.
Authors: Mark Weber, Lijun Yu, Qihang Yu, Xueqing Deng, Xiaohui Shen, Daniel Cremers, Liang-Chieh Chen
Abstract: Masked transformer models for class-conditional image generation have become a compelling alternative to diffusion models. Typically comprising two stages - an initial VQGAN model for transitioning between latent space and image space, and a subsequent Transformer model for image generation within latent space - these frameworks offer promising avenues for image synthesis. In this study, we present two primary contributions: Firstly, an empirical and systematic examination of VQGANs, leading to a modernized VQGAN. Secondly, a novel embedding-free generation network operating directly on bit tokens - a binary quantized representation of tokens with rich semantics. The first contribution furnishes a transparent, reproducible, and high-performing VQGAN model, enhancing accessibility and matching the performance of current state-of-the-art methods while revealing previously undisclosed details. The second contribution demonstrates that embedding-free image generation using bit tokens achieves a new state-of-the-art FID of 1.52 on the ImageNet 256x256 benchmark, with a compact generator model of mere 305M parameters. The code for this project is available on https://github.com/markweberdev/maskbit.
Authors: Yong Xien Chng, Xuchong Qiu, Yizeng Han, Kai Ding, Wan Ding, Gao Huang
Abstract: Despite extensive research, open-vocabulary segmentation methods still struggle to generalize across diverse domains. To reduce the computational cost of adapting Vision-Language Models (VLMs) while preserving their pre-trained knowledge, most methods freeze the VLMs for mask classification and train only the mask generator. However, our comprehensive analysis reveals a surprising insight: open-vocabulary segmentation is primarily bottlenecked by mask classification, not mask generation. This discovery prompts us to rethink the existing paradigm and explore an alternative approach. Instead of freezing the VLM, we propose to freeze the pre-trained mask generator and focus on optimizing the mask classifier. Building on the observation that VLMs pre-trained on global-pooled image-text features often fail to capture fine-grained semantics necessary for effective mask classification, we propose a novel Fine-grained Semantic Adaptation (FISA) method to address this limitation. FISA enhances the extracted visual features with fine-grained semantic awareness by explicitly integrating this crucial semantic information early in the visual encoding process. As our method strategically optimizes only a small portion of the VLM's parameters, it enjoys the efficiency of adapting to new data distributions while largely preserving the valuable VLM pre-trained knowledge. Extensive ablation studies confirm the superiority of our approach. Notably, FISA achieves new state-of-the-art results across multiple representative benchmarks, improving performance by up to +1.0 PQ and +3.0 mIoU and reduces training costs by nearly 5x compared to previous best methods. Our code and data will be made public.
Authors: George R. Nahass, Ghasem Yazdanpanah, Madison Cheung, Alex Palacios, Jeffrey C. Peterson, Kevin Heinze, Sasha Hubschman, Chad A. Purnell, Pete Setabutr, Ann Q. Tran, Darvin Yi
Abstract: Periorbital distances and features around the eyes and lids hold valuable information for disease quantification and monitoring of surgical and medical intervention. These distances are commonly measured manually, a process that is both subjective and highly time-consuming. Here, we set out to developed three deep-learning methods for segmentation and periorbital distance prediction, and also evaluate the utility of periorbital distances for disease classification. The MAE of our deep learning predicted distances was less than or very close to the error observed between trained human annotators. We compared our models to the current state-of-the-art (SOTA) method for periorbital distance prediction and found that our methods outperformed SOTA on all of our datasets on all but one periorbital measurement. We also show that robust segmentation can be achieved on diseased eyes using models trained on open-source, healthy eyes, and that periorbital distances have can be used as high-quality features in downstream classification models. Leveraging segmentation networks as intermediary steps in classification has broad implications for increasing the generalizability of classification models in ophthalmic plastic and craniofacial surgery by avoiding the out-of-distribution problem observed in traditional convolutional neural networks.
Authors: George R. Nahass, Emma Koehler, Nicholas Tomaras, Danny Lopez, Madison Cheung, Alexander Palacios, Jeffrey C. Peterson, Sasha Hubschman, Kelsey Green, Chad A. Purnell, Pete Setabutr, Ann Q. Tran, Darvin Yi
Abstract: Periorbital segmentation and distance prediction using deep learning allows for the objective quantification of disease state, treatment monitoring, and remote medicine. However, there are currently no reports of segmentation datasets for the purposes of training deep learning models with sub mm accuracy on the regions around the eyes. All images (n=2842) had the iris, sclera, lid, caruncle, and brow segmented by five trained annotators. Here, we validate this dataset through intra and intergrader reliability tests and show the utility of the data in training periorbital segmentation networks. All the annotations are publicly available for free download. Having access to segmentation datasets designed specifically for oculoplastic surgery will permit more rapid development of clinically useful segmentation networks which can be leveraged for periorbital distance prediction and disease classification. In addition to the annotations, we also provide an open-source toolkit for periorbital distance prediction from segmentation masks. The weights of all models have also been open-sourced and are publicly available for use by the community.
Authors: Huy Nguyen, Kien Nguyen, Akila Pemasiri, Sridha Sridharan, Clinton Fookes
Abstract: This study introduces a new framework for 3D person re-identification (re-ID) that leverages readily available high-resolution texture data in 3D reconstruction to improve the performance and explainability of the person re-ID task. We propose a method to emphasize texture in 3D person re-ID models by incorporating UVTexture mapping, which better differentiates human subjects. Our approach uniquely combines UVTexture and its heatmaps with 3D models to visualize and explain the person re-ID process. In particular, the visualization and explanation are achieved through activation maps and attribute-based attention maps, which highlight the important regions and features contributing to the person re-ID decision. Our contributions include: (1) a novel technique for emphasizing texture in 3D models using UVTexture processing, (2) an innovative method for explicating person re-ID matches through a combination of 3D models and UVTexture mapping, and (3) achieving state-of-the-art performance in 3D person re-ID. We ensure the reproducibility of our results by making all data, codes, and models publicly available.
Authors: Chengkai Hou, Zhengrong Xue, Bingyang Zhou, Jinghan Ke, Lin Shao, Huazhe Xu
Abstract: Detecting 3D keypoints with semantic consistency is widely used in many scenarios such as pose estimation, shape registration and robotics. Currently, most unsupervised 3D keypoint detection methods focus on the rigid-body objects. However, when faced with deformable objects, the keypoints they identify do not preserve semantic consistency well. In this paper, we introduce an innovative unsupervised keypoint detector Key-Grid for both the rigid-body and deformable objects, which is an autoencoder framework. The encoder predicts keypoints and the decoder utilizes the generated keypoints to reconstruct the objects. Unlike previous work, we leverage the identified keypoint in formation to form a 3D grid feature heatmap called grid heatmap, which is used in the decoder section. Grid heatmap is a novel concept that represents the latent variables for grid points sampled uniformly in the 3D cubic space, where these variables are the shortest distance between the grid points and the skeleton connected by keypoint pairs. Meanwhile, we incorporate the information from each layer of the encoder into the decoder section. We conduct an extensive evaluation of Key-Grid on a list of benchmark datasets. Key-Grid achieves the state-of-the-art performance on the semantic consistency and position accuracy of keypoints. Moreover, we demonstrate the robustness of Key-Grid to noise and downsampling. In addition, we achieve SE-(3) invariance of keypoints though generalizing Key-Grid to a SE(3)-invariant backbone.
Authors: Dohun Lee, Bryan S Kim, Geon Yeong Park, Jong Chul Ye
Abstract: Text-to-image (T2I) diffusion models have revolutionized visual content creation, but extending these capabilities to text-to-video (T2V) generation remains a challenge, particularly in preserving temporal consistency. Existing methods that aim to improve consistency often cause trade-offs such as reduced imaging quality and impractical computational time. To address these issues we introduce VideoGuide, a novel framework that enhances the temporal consistency of pretrained T2V models without the need for additional training or fine-tuning. Instead, VideoGuide leverages any pretrained video diffusion model (VDM) or itself as a guide during the early stages of inference, improving temporal quality by interpolating the guiding model's denoised samples into the sampling model's denoising process. The proposed method brings about significant improvement in temporal consistency and image fidelity, providing a cost-effective and practical solution that synergizes the strengths of various video diffusion models. Furthermore, we demonstrate prior distillation, revealing that base models can achieve enhanced text coherence by utilizing the superior data prior of the guiding model through the proposed method. Project Page: https://dohunlee1.github.io/videoguide.github.io/
Authors: Yiming Zuo, Karhan Kayan, Maggie Wang, Kevin Jeon, Jia Deng, Thomas L. Griffiths
Abstract: Building a foundation model for 3D vision is a complex challenge that remains unsolved. Towards that goal, it is important to understand the 3D reasoning capabilities of current models as well as identify the gaps between these models and humans. Therefore, we construct a new 3D visual understanding benchmark named UniQA-3D. UniQA-3D covers fundamental 3D vision tasks in the Visual Question Answering (VQA) format. We evaluate state-of-the-art Vision-Language Models (VLMs), specialized models, and human subjects on it. Our results show that VLMs generally perform poorly, while the specialized models are accurate but not robust, failing under geometric perturbations. In contrast, human vision continues to be the most reliable 3D visual system. We further demonstrate that neural networks align more closely with human 3D vision mechanisms compared to classical computer vision methods, and Transformer-based networks such as ViT align more closely with human 3D vision mechanisms than CNNs. We hope our study will benefit the future development of foundation models for 3D vision. Code is available at https://github.com/princeton-vl/UniQA-3D .
Authors: Shizhen Zhao, Xin Wen, Jiahui Liu, Chuofan Ma, Chunfeng Yuan, Xiaojuan Qi
Abstract: Balancing training on long-tail data distributions remains a long-standing challenge in deep learning. While methods such as re-weighting and re-sampling help alleviate the imbalance issue, limited sample diversity continues to hinder models from learning robust and generalizable feature representations, particularly for tail classes. In contrast to existing methods, we offer a novel perspective on long-tail learning, inspired by an observation: datasets with finer granularity tend to be less affected by data imbalance. In this paper, we investigate this phenomenon through both quantitative and qualitative studies, showing that increased granularity enhances the generalization of learned features in tail categories. Motivated by these findings, we propose a method to increase dataset granularity through category extrapolation. Specifically, we introduce open-set auxiliary classes that are visually similar to existing ones, aiming to enhance representation learning for both head and tail classes. This forms the core contribution and insight of our approach. To automate the curation of auxiliary data, we leverage large language models (LLMs) as knowledge bases to search for auxiliary categories and retrieve relevant images through web crawling. To prevent the overwhelming presence of auxiliary classes from disrupting training, we introduce a neighbor-silencing loss that encourages the model to focus on class discrimination within the target dataset. During inference, the classifier weights for auxiliary categories are masked out, leaving only the target class weights for use. Extensive experiments and ablation studies on three standard long-tail benchmarks demonstrate the effectiveness of our approach, notably outperforming strong baseline methods that use the same amount of data. The code will be made publicly available.
Authors: Cheng-De Fan, Chen-Wei Chang, Yi-Ruei Liu, Jie-Ying Lee, Jiun-Long Huang, Yu-Chee Tseng, Yu-Lun Liu
Abstract: We present SpectroMotion, a novel approach that combines 3D Gaussian Splatting (3DGS) with physically-based rendering (PBR) and deformation fields to reconstruct dynamic specular scenes. Previous methods extending 3DGS to model dynamic scenes have struggled to represent specular surfaces accurately. Our method addresses this limitation by introducing a residual correction technique for accurate surface normal computation during deformation, complemented by a deformable environment map that adapts to time-varying lighting conditions. We implement a coarse-to-fine training strategy that significantly enhances scene geometry and specular color prediction. It is the only existing 3DGS method capable of synthesizing photorealistic real-world dynamic specular scenes, outperforming state-of-the-art methods in rendering complex, dynamic, and specular scenes. See our project page for video results at https://cdfan0627.github.io/spectromotion/.
Authors: Haiyang Zheng, Nan Pu, Wenjing Li, Nicu Sebe, Zhun Zhong
Abstract: In this paper, we study a practical yet challenging task, On-the-fly Category Discovery (OCD), aiming to online discover the newly-coming stream data that belong to both known and unknown classes, by leveraging only known category knowledge contained in labeled data. Previous OCD methods employ the hash-based technique to represent old/new categories by hash codes for instance-wise inference. However, directly mapping features into low-dimensional hash space not only inevitably damages the ability to distinguish classes and but also causes "high sensitivity" issue, especially for fine-grained classes, leading to inferior performance. To address these issues, we propose a novel Prototypical Hash Encoding (PHE) framework consisting of Category-aware Prototype Generation (CPG) and Discriminative Category Encoding (DCE) to mitigate the sensitivity of hash code while preserving rich discriminative information contained in high-dimension feature space, in a two-stage projection fashion. CPG enables the model to fully capture the intra-category diversity by representing each category with multiple prototypes. DCE boosts the discrimination ability of hash code with the guidance of the generated category prototypes and the constraint of minimum separation distance. By jointly optimizing CPG and DCE, we demonstrate that these two components are mutually beneficial towards an effective OCD. Extensive experiments show the significant superiority of our PHE over previous methods, e.g., obtaining an improvement of +5.3% in ALL ACC averaged on all datasets. Moreover, due to the nature of the interpretable prototypes, we visually analyze the underlying mechanism of how PHE helps group certain samples into either known or unknown categories. Code is available at https://github.com/HaiyangZheng/PHE.
Authors: Naga VS Raviteja Chappa, Page Daniel Dobbs, Bhiksha Raj, Khoa Luu
Abstract: The proliferation of tobacco-related content on social media platforms poses significant challenges for public health monitoring and intervention. This paper introduces a novel multi-modal deep learning framework named Flow-Attention Adaptive Semantic Hierarchical Fusion (FLAASH) designed to analyze tobacco-related video content comprehensively. FLAASH addresses the complexities of integrating visual and textual information in short-form videos by leveraging a hierarchical fusion mechanism inspired by flow network theory. Our approach incorporates three key innovations, including a flow-attention mechanism that captures nuanced interactions between visual and textual modalities, an adaptive weighting scheme that balances the contribution of different hierarchical levels, and a gating mechanism that selectively emphasizes relevant features. This multi-faceted approach enables FLAASH to effectively process and analyze diverse tobacco-related content, from product showcases to usage scenarios. We evaluate FLAASH on the Multimodal Tobacco Content Analysis Dataset (MTCAD), a large-scale collection of tobacco-related videos from popular social media platforms. Our results demonstrate significant improvements over existing methods, outperforming state-of-the-art approaches in classification accuracy, F1 score, and temporal consistency. The proposed method also shows strong generalization capabilities when tested on standard video question-answering datasets, surpassing current models. This work contributes to the intersection of public health and artificial intelligence, offering an effective tool for analyzing tobacco promotion in digital media.
Authors: Yitong Dong, Yijin Li, Zhaoyang Huang, Weikang Bian, Jingbo Liu, Hujun Bao, Zhaopeng Cui, Hongsheng Li, Guofeng Zhang
Abstract: In this paper, we propose a novel multi-view stereo (MVS) framework that gets rid of the depth range prior. Unlike recent prior-free MVS methods that work in a pair-wise manner, our method simultaneously considers all the source images. Specifically, we introduce a Multi-view Disparity Attention (MDA) module to aggregate long-range context information within and across multi-view images. Considering the asymmetry of the epipolar disparity flow, the key to our method lies in accurately modeling multi-view geometric constraints. We integrate pose embedding to encapsulate information such as multi-view camera poses, providing implicit geometric constraints for multi-view disparity feature fusion dominated by attention. Additionally, we construct corresponding hidden states for each source image due to significant differences in the observation quality of the same pixel in the reference frame across multiple source frames. We explicitly estimate the quality of the current pixel corresponding to sampled points on the epipolar line of the source image and dynamically update hidden states through the uncertainty estimation module. Extensive results on the DTU dataset and Tanks&Temple benchmark demonstrate the effectiveness of our method. The code is available at our project page: https://zju3dv.github.io/GD-PoseMVS/.
Authors: Jiyul Ham, Yonggon Jung, Jun-Geol Baek
Abstract: Zero-shot anomaly detection (ZSAD) is crucial for detecting anomalous patterns in target datasets without using training samples, specifically in scenarios where there are distributional differences between the target domain and training data or where data scarcity arises because of restricted access. Although recently pretrained vision-language models demonstrate strong zero-shot performance across various visual tasks, they focus on learning class semantics, which makes their direct application to ZSAD challenging. To address this scenario, we propose GlocalCLIP, which uniquely separates global and local prompts and jointly optimizes them. This approach enables the object-agnostic glocal semantic prompt to effectively capture general normal and anomalous patterns without dependency on specific objects in the image. We refine the text prompts for more precise adjustments by utilizing deep-text prompt tuning in the text encoder. In the vision encoder, we apply V-V attention layers to capture detailed local image features. Finally, we introduce glocal contrastive learning to improve the complementary learning of global and local prompts, effectively detecting anomalous patterns across various domains. The generalization performance of GlocalCLIP in ZSAD was demonstrated on 15 real-world datasets from both the industrial and medical domains, achieving superior performance compared to existing methods. Code will be made available at https://github.com/YUL-git/GlocalCLIP.
Authors: Xuannan Liu, Xing Cui, Peipei Li, Zekun Li, Huaibo Huang, Shuhan Xia, Miaoxuan Zhang, Yueying Zou, Ran He
Abstract: The rapid evolution of multimodal foundation models has led to significant advancements in cross-modal understanding and generation across diverse modalities, including text, images, audio, and video. However, these models remain susceptible to jailbreak attacks, which can bypass built-in safety mechanisms and induce the production of potentially harmful content. Consequently, understanding the methods of jailbreak attacks and existing defense mechanisms is essential to ensure the safe deployment of multimodal generative models in real-world scenarios, particularly in security-sensitive applications. To provide comprehensive insight into this topic, this survey reviews jailbreak and defense in multimodal generative models. First, given the generalized lifecycle of multimodal jailbreak, we systematically explore attacks and corresponding defense strategies across four levels: input, encoder, generator, and output. Based on this analysis, we present a detailed taxonomy of attack methods, defense mechanisms, and evaluation frameworks specific to multimodal generative models. Additionally, we cover a wide range of input-output configurations, including modalities such as Any-to-Text, Any-to-Vision, and Any-to-Any within generative systems. Finally, we highlight current research challenges and propose potential directions for future research. The open-source repository corresponding to this work can be found at https://github.com/liuxuannan/Awesome-Multimodal-Jailbreak.
URLs: https://github.com/liuxuannan/Awesome-Multimodal-Jailbreak.
Authors: Ryoma Yataka, Adriano Cardace, Pu Perry Wang, Petros Boufounos, Ryuhei Takahashi
Abstract: Indoor radar perception has seen rising interest due to affordable costs driven by emerging automotive imaging radar developments and the benefits of reduced privacy concerns and reliability under hazardous conditions (e.g., fire and smoke). However, existing radar perception pipelines fail to account for distinctive characteristics of the multi-view radar setting. In this paper, we propose Radar dEtection TRansformer (RETR), an extension of the popular DETR architecture, tailored for multi-view radar perception. RETR inherits the advantages of DETR, eliminating the need for hand-crafted components for object detection and segmentation in the image plane. More importantly, RETR incorporates carefully designed modifications such as 1) depth-prioritized feature similarity via a tunable positional encoding (TPE); 2) a tri-plane loss from both radar and camera coordinates; and 3) a learnable radar-to-camera transformation via reparameterization, to account for the unique multi-view radar setting. Evaluated on two indoor radar perception datasets, our approach outperforms existing state-of-the-art methods by a margin of 15.38+ AP for object detection and 11.91+ IoU for instance segmentation, respectively.
Authors: Mahmudul Hasan
Abstract: In computer vision tasks, the ability to focus on relevant regions within an image is crucial for improving model performance, particularly when key features are small, subtle, or spatially dispersed. Convolutional neural networks (CNNs) typically treat all regions of an image equally, which can lead to inefficient feature extraction. To address this challenge, I have introduced Vision Eagle Attention, a novel attention mechanism that enhances visual feature extraction using convolutional spatial attention. The model applies convolution to capture local spatial features and generates an attention map that selectively emphasizes the most informative regions of the image. This attention mechanism enables the model to focus on discriminative features while suppressing irrelevant background information. I have integrated Vision Eagle Attention into a lightweight ResNet-18 architecture, demonstrating that this combination results in an efficient and powerful model. I have evaluated the performance of the proposed model on three widely used benchmark datasets: FashionMNIST, Intel Image Classification, and OracleMNIST, with a primary focus on image classification. Experimental results show that the proposed approach improves classification accuracy. Additionally, this method has the potential to be extended to other vision tasks, such as object detection, segmentation, and visual tracking, offering a computationally efficient solution for a wide range of vision-based applications. Code is available at: https://github.com/MahmudulHasan11085/Vision-Eagle-Attention.git
URLs: https://github.com/MahmudulHasan11085/Vision-Eagle-Attention.git
Authors: Anthony Nguyen
Abstract: Graphical User Interface (GUI) grounding plays a crucial role in enhancing the capabilities of Vision-Language Model (VLM) agents. While general VLMs, such as GPT-4V, demonstrate strong performance across various tasks, their proficiency in GUI grounding remains suboptimal. Recent studies have focused on fine-tuning these models specifically for one-shot GUI grounding, yielding significant improvements over baseline performance. We introduce a visual prompting framework that employs an iterative narrowing mechanism to further improve the performance of both general and fine-tuned models in GUI grounding. For evaluation, we tested our method on a comprehensive benchmark comprising various UI platforms and provided the code to reproduce our results.
Authors: Hugo Carneiro, Stefan Wermter
Abstract: Active speaker detection (ASD) in multimodal environments is crucial for various applications, from video conferencing to human-robot interaction. This paper introduces FabuLight-ASD, an advanced ASD model that integrates facial, audio, and body pose information to enhance detection accuracy and robustness. Our model builds upon the existing Light-ASD framework by incorporating human pose data, represented through skeleton graphs, which minimises computational overhead. Using the Wilder Active Speaker Detection (WASD) dataset, renowned for reliable face and body bounding box annotations, we demonstrate FabuLight-ASD's effectiveness in real-world scenarios. Achieving an overall mean average precision (mAP) of 94.3%, FabuLight-ASD outperforms Light-ASD, which has an overall mAP of 93.7% across various challenging scenarios. The incorporation of body pose information shows a particularly advantageous impact, with notable improvements in mAP observed in scenarios with speech impairment, face occlusion, and human voice background noise. Furthermore, efficiency analysis indicates only a modest increase in parameter count (27.3%) and multiply-accumulate operations (up to 2.4%), underscoring the model's efficiency and feasibility. These findings validate the efficacy of FabuLight-ASD in enhancing ASD performance through the integration of body pose data. FabuLight-ASD's code and model weights are available at https://github.com/knowledgetechnologyuhh/FabuLight-ASD.
URLs: https://github.com/knowledgetechnologyuhh/FabuLight-ASD.
Authors: Chaoyou Fu, Yi-Fan Zhang, Shukang Yin, Bo Li, Xinyu Fang, Sirui Zhao, Haodong Duan, Xing Sun, Ziwei Liu, Liang Wang, Caifeng Shan, Ran He
Abstract: As a prominent direction of Artificial General Intelligence (AGI), Multimodal Large Language Models (MLLMs) have garnered increased attention from both industry and academia. Building upon pre-trained LLMs, this family of models further develops multimodal perception and reasoning capabilities that are impressive, such as writing code given a flow chart or creating stories based on an image. In the development process, evaluation is critical since it provides intuitive feedback and guidance on improving models. Distinct from the traditional train-eval-test paradigm that only favors a single task like image classification, the versatility of MLLMs has spurred the rise of various new benchmarks and evaluation methods. In this paper, we aim to present a comprehensive survey of MLLM evaluation, discussing four key aspects: 1) the summarised benchmarks types divided by the evaluation capabilities, including foundation capabilities, model self-analysis, and extented applications; 2) the typical process of benchmark counstruction, consisting of data collection, annotation, and precautions; 3) the systematic evaluation manner composed of judge, metric, and toolkit; 4) the outlook for the next benchmark. This work aims to offer researchers an easy grasp of how to effectively evaluate MLLMs according to different needs and to inspire better evaluation methods, thereby driving the progress of MLLM research.
Authors: Niranjan Rajesh, Georgin Jacob, SP Arun
Abstract: Despite the rapid pace at which deep networks are improving on standardized vision benchmarks, they are still outperformed by humans on real-world vision tasks. This paradoxical lack of generalization could be addressed by making deep networks more brain-like. Although several benchmarks have compared the ability of deep networks to predict brain responses to natural images, they do not capture subtle but important brain-like emergent properties. To resolve this issue, we report several well-known perceptual and neural emergent properties that can be tested on deep networks. To evaluate how various design factors impact brain-like properties, we systematically evaluated over 30 state-of-the-art networks with varying network architectures, training datasets and training regimes. Our main findings are as follows. First, network architecture had the strongest impact on brain-like properties compared to dataset and training regime variations. Second, networks varied widely in their alignment to the brain with no single network outperforming all others. Taken together, our results complement existing benchmarks by revealing brain-like properties that are either emergent or lacking in state-of-the-art deep networks.
Authors: Ziang Xu, Bin Li, Yang Hu, Chenyu Zhang, James East, Sharib Ali, Jens Rittscher
Abstract: Accurate 3D mapping in endoscopy enables quantitative, holistic lesion characterization within the gastrointestinal (GI) tract, requiring reliable depth and pose estimation. However, endoscopy systems are monocular, and existing methods relying on synthetic datasets or complex models often lack generalizability in challenging endoscopic conditions. We propose a robust self-supervised monocular depth and pose estimation framework that incorporates a Generative Latent Bank and a Variational Autoencoder (VAE). The Generative Latent Bank leverages extensive depth scenes from natural images to condition the depth network, enhancing realism and robustness of depth predictions through latent feature priors. For pose estimation, we reformulate it within a VAE framework, treating pose transitions as latent variables to regularize scale, stabilize z-axis prominence, and improve x-y sensitivity. This dual refinement pipeline enables accurate depth and pose predictions, effectively addressing the GI tract's complex textures and lighting. Extensive evaluations on SimCol and EndoSLAM datasets confirm our framework's superior performance over published self-supervised methods in endoscopic depth and pose estimation.
Authors: Zhang Cheng, Yanxia Wang
Abstract: Using lightweight models as backbone networks in gaze estimation tasks often results in significant performance degradation. The main reason is that the number of feature channels in lightweight networks is usually small, which makes the model expression ability limited. In order to improve the performance of lightweight models in gaze estimation tasks, a network model named Multitask-Gaze is proposed. The main components of Multitask-Gaze include Unidirectional Convolution (UC), Spatial and Channel Attention (SCA), Global Convolution Module (GCM), and Multi-task Regression Module(MRM). UC not only significantly reduces the number of parameters and FLOPs, but also extends the receptive field and improves the long-distance modeling capability of the model, thereby improving the model performance. SCA highlights gaze-related features and suppresses gaze-irrelevant features. The GCM replaces the pooling layer and avoids the performance degradation due to information loss. MRM improves the accuracy of individual tasks and strengthens the connections between tasks for overall performance improvement. The experimental results show that compared with the State-of-the-art method SUGE, the performance of Multitask-Gaze on MPIIFaceGaze and Gaze360 datasets is improved by 1.71% and 2.75%, respectively, while the number of parameters and FLOPs are significantly reduced by 75.5% and 86.88%.
Authors: Xue Song, Jiequan Cui, Hanwang Zhang, Jiaxin Shi, Jingjing Chen, Chi Zhang, Yu-Gang Jiang
Abstract: In this paper, we propose the LoRA of Change (LoC) framework for image editing with visual instructions, i.e., before-after image pairs. Compared to the ambiguities, insufficient specificity, and diverse interpretations of natural language, visual instructions can accurately reflect users' intent. Building on the success of LoRA in text-based image editing and generation, we dynamically learn an instruction-specific LoRA to encode the "change" in a before-after image pair, enhancing the interpretability and reusability of our model. Furthermore, generalizable models for image editing with visual instructions typically require quad data, i.e., a before-after image pair, along with query and target images. Due to the scarcity of such quad data, existing models are limited to a narrow range of visual instructions. To overcome this limitation, we introduce the LoRA Reverse optimization technique, enabling large-scale training with paired data alone. Extensive qualitative and quantitative experiments demonstrate that our model produces high-quality images that align with user intent and support a broad spectrum of real-world visual instructions.
Authors: Herbert Wright, Weiming Zhi, Matthew Johnson-Roberson, Tucker Hermans
Abstract: Constructing 3D representations of object geometry is critical for many downstream robotics tasks, particularly tabletop manipulation problems. These representations must be built from potentially noisy partial observations. In this work, we focus on the problem of reconstructing a multi-object scene from a single RGBD image, generally from a fixed camera in the scene. Traditional scene representation methods generally cannot infer the geometry of unobserved regions of the objects from the image. Attempts have been made to leverage deep learning to train on a dataset of observed objects and representations, and then generalize to new observations. However, this can be brittle to noisy real-world observations and objects not contained in the dataset, and cannot reason about their confidence. We propose BRRP, a reconstruction method that leverages preexisting mesh datasets to build an informative prior during robust probabilistic reconstruction. In order to make our method more efficient, we introduce the concept of retrieval-augmented prior, where we retrieve relevant components of our prior distribution during inference. The prior is used to estimate the geometry of occluded portions of the in-scene objects. Our method produces a distribution over object shape that can be used for reconstruction or measuring uncertainty. We evaluate our method in both simulated scenes and in the real world. We demonstrate the robustness of our method against deep learning-only approaches while being more accurate than a method without an informative prior.
Authors: Yu Wang, Xiaofei Zhou, Yichen Wang, Geyuan Zhang, Tianxing He
Abstract: With the significant advancement of Large Vision-Language Models (VLMs), concerns about their potential misuse and abuse have grown rapidly. Previous studies have highlighted VLMs' vulnerability to jailbreak attacks, where carefully crafted inputs can lead the model to produce content that violates ethical and legal standards. However, existing methods struggle against state-of-the-art VLMs like GPT-4o, due to the over-exposure of harmful content and lack of stealthy malicious guidance. In this work, we propose a novel jailbreak attack framework: Multi-Modal Linkage (MML) Attack. Drawing inspiration from cryptography, MML utilizes an encryption-decryption process across text and image modalities to mitigate over-exposure of malicious information. To align the model's output with malicious intent covertly, MML employs a technique called "evil alignment", framing the attack within a video game production scenario. Comprehensive experiments demonstrate MML's effectiveness. Specifically, MML jailbreaks GPT-4o with attack success rates of 97.80% on SafeBench, 98.81% on MM-SafeBench and 99.07% on HADES-Dataset. Our code is available at https://github.com/wangyu-ovo/MML
Authors: Shiyu Zhao, Zhenting Wang, Felix Juefei-Xu, Xide Xia, Miao Liu, Xiaofang Wang, Mingfu Liang, Ning Zhang, Dimitris N. Metaxas, Licheng Yu
Abstract: Prevailing Multimodal Large Language Models (MLLMs) encode the input image(s) as vision tokens and feed them into the language backbone, similar to how Large Language Models (LLMs) process the text tokens. However, the number of vision tokens increases quadratically as the image resolutions, leading to huge computational costs. In this paper, we consider improving MLLM's efficiency from two scenarios, (I) Reducing computational cost without degrading the performance. (II) Improving the performance with given budgets. We start with our main finding that the ranking of each vision token sorted by attention scores is similar in each layer except the first layer. Based on it, we assume that the number of essential top vision tokens does not increase along layers. Accordingly, for Scenario I, we propose a greedy search algorithm (G-Search) to find the least number of vision tokens to keep at each layer from the shallow to the deep. Interestingly, G-Search is able to reach the optimal reduction strategy based on our assumption. For Scenario II, based on the reduction strategy from G-Search, we design a parametric sigmoid function (P-Sigmoid) to guide the reduction at each layer of the MLLM, whose parameters are optimized by Bayesian Optimization. Extensive experiments demonstrate that our approach can significantly accelerate those popular MLLMs, e.g. LLaVA, and InternVL2 models, by more than $2 \times$ without performance drops. Our approach also far outperforms other token reduction methods when budgets are limited, achieving a better trade-off between efficiency and effectiveness.
Authors: Chensheng Peng, Ido Sobol, Masayoshi Tomizuka, Kurt Keutzer, Chenfeng Xu, Or Litany
Abstract: We introduce a diffusion model for Gaussian Splats, SplatDiffusion, to enable generation of three-dimensional structures from single images, addressing the ill-posed nature of lifting 2D inputs to 3D. Existing methods rely on deterministic, feed-forward predictions, which limit their ability to handle the inherent ambiguity of 3D inference from 2D data. Diffusion models have recently shown promise as powerful generative models for 3D data, including Gaussian splats; however, standard diffusion frameworks typically require the target signal and denoised signal to be in the same modality, which is challenging given the scarcity of 3D data. To overcome this, we propose a novel training strategy that decouples the denoised modality from the supervision modality. By using a deterministic model as a noisy teacher to create the noised signal and transitioning from single-step to multi-step denoising supervised by an image rendering loss, our approach significantly enhances performance compared to the deterministic teacher. Additionally, our method is flexible, as it can learn from various 3D Gaussian Splat (3DGS) teachers with minimal adaptation; we demonstrate this by surpassing the performance of two different deterministic models as teachers, highlighting the potential generalizability of our framework. Our approach further incorporates a guidance mechanism to aggregate information from multiple views, enhancing reconstruction quality when more than one view is available. Experimental results on object-level and scene-level datasets demonstrate the effectiveness of our framework.
Authors: Yizhou Wang, Tim Meinhardt, Orcun Cetintas, Cheng-Yen Yang, Sameer Satish Pusegaonkar, Benjamin Missaoui, Sujit Biswas, Zheng Tang, Laura Leal-Taix\'e
Abstract: Object perception from multi-view cameras is crucial for intelligent systems, particularly in indoor environments, e.g., warehouses, retail stores, and hospitals. Most traditional multi-target multi-camera (MTMC) detection and tracking methods rely on 2D object detection, single-view multi-object tracking (MOT), and cross-view re-identification (ReID) techniques, without properly handling important 3D information by multi-view image aggregation. In this paper, we propose a 3D object detection and tracking framework, named BEV-SUSHI, which first aggregates multi-view images with necessary camera calibration parameters to obtain 3D object detections in bird's-eye view (BEV). Then, we introduce hierarchical graph neural networks (GNNs) to track these 3D detections in BEV for MTMC tracking results. Unlike existing methods, BEV-SUSHI has impressive generalizability across different scenes and diverse camera settings, with exceptional capability for long-term association handling. As a result, our proposed BEV-SUSHI establishes the new state-of-the-art on the AICity'24 dataset with 81.22 HOTA, and 95.6 IDF1 on the WildTrack dataset.
Authors: Haodong Lu, Chongyang Zhao, Jason Xue, Lina Yao, Kristen Moore, Dong Gong
Abstract: We investigate whether the pre-trained knowledge of vision-language models (VLMs), such as CLIP, can be retained or even enhanced during continual learning (CL) while absorbing knowledge from a data stream. Existing methods often rely on additional reference data, isolated components for distribution or domain predictions, leading to high training costs, increased inference complexity, and limited improvement potential for pre-trained models. To address these challenges, we first comprehensively analyze the effects of parameter update locations and ranks on downstream adaptation and knowledge retention. Based on these insights, we propose Dynamic Rank-Selective Low Rank Adaptation (LoRA), a universal and efficient CL approach that adaptively assigns ranks to LoRA modules based on their relevance to the current data. Unlike prior methods, our approach continually enhances the pre-trained VLM by retaining both the pre-trained knowledge and the knowledge acquired during CL. Our approach eliminates the need for explicit domain or distribution prediction and additional reference data, enabling seamless integration of new tasks while preserving pre-trained capabilities. It also maintains the original architecture and deployment pipeline of the pre-trained model without incurring any additional inference overhead. Extensive experiments and analyses demonstrate that our method outperforms state-of-the-art approaches in continually absorbing knowledge of downstream tasks while retaining pre-trained knowledge.
Authors: Ziyang Yan, Lei Li, Yihua Shao, Siyu Chen, Zongkai Wu, Jenq-Neng Hwang, Hao Zhao, Fabio Remondino
Abstract: The creation of 3D scenes has traditionally been both labor-intensive and costly, requiring designers to meticulously configure 3D assets and environments. Recent advancements in generative AI, including text-to-3D and image-to-3D methods, have dramatically reduced the complexity and cost of this process. However, current techniques for editing complex 3D scenes continue to rely on generally interactive multi-step, 2D-to-3D projection methods and diffusion-based techniques, which often lack precision in control and hamper real-time performance. In this work, we propose 3DSceneEditor, a fully 3D-based paradigm for real-time, precise editing of intricate 3D scenes using Gaussian Splatting. Unlike conventional methods, 3DSceneEditor operates through a streamlined 3D pipeline, enabling direct manipulation of Gaussians for efficient, high-quality edits based on input prompts.The proposed framework (i) integrates a pre-trained instance segmentation model for semantic labeling; (ii) employs a zero-shot grounding approach with CLIP to align target objects with user prompts; and (iii) applies scene modifications, such as object addition, repositioning, recoloring, replacing, and deletion directly on Gaussians. Extensive experimental results show that 3DSceneEditor achieves superior editing precision and speed with respect to current SOTA 3D scene editing approaches, establishing a new benchmark for efficient and interactive 3D scene customization.
Authors: Anirudh S Chakravarthy, Shuai Kyle Zheng, Xin Huang, Sachithra Hemachandra, Xiao Zhang, Yuning Chai, Zhao Chen
Abstract: Fine-tuning pre-trained models has become invaluable in computer vision and robotics. Recent fine-tuning approaches focus on improving efficiency rather than accuracy by using a mixture of smaller learning rates or frozen backbones. To return the spotlight to model accuracy, we present PROFIT (Proximally Restricted Optimizer For Iterative Training), one of the first optimizers specifically designed for incrementally fine-tuning converged models on new tasks or datasets. Unlike traditional optimizers such as SGD or Adam, which make minimal assumptions due to random initialization, PROFIT leverages the structure of a converged model to regularize the optimization process, leading to improved results. By employing a simple temporal gradient orthogonalization process, PROFIT outperforms traditional fine-tuning methods across various tasks: image classification, representation learning, and large-scale motion prediction. Moreover, PROFIT is encapsulated within the optimizer logic, making it easily integrated into any training pipeline with minimal engineering effort. A new class of fine-tuning optimizers like PROFIT can drive advancements as fine-tuning and incremental training become increasingly prevalent, reducing reliance on costly model training from scratch.
Authors: Gustavo P. C. P. da Luz, Gabriel Massuyoshi Sato, Luis Fernando Gomez Gonzalez, Juliana Freitag Borin
Abstract: The increasing urbanization and the growing number of vehicles in cities have underscored the need for efficient parking management systems. Traditional smart parking solutions often rely on sensors or cameras for occupancy detection, each with its limitations. Recent advancements in deep learning have introduced new YOLO models (YOLOv8, YOLOv9, YOLOv10, and YOLOv11), but these models have not been extensively evaluated in the context of smart parking systems, particularly when combined with Region of Interest (ROI) selection for object detection. Existing methods still rely on fixed polygonal ROI selections or simple pixel-based modifications, which limit flexibility and precision. This work introduces a novel approach that integrates Internet of Things, Edge Computing, and Deep Learning concepts, by using the latest YOLO models for vehicle detection. By exploring both edge and cloud computing, it was found that inference times on edge devices ranged from 1 to 92 seconds, depending on the hardware and model version. Additionally, a new pixel-wise post-processing ROI selection method is proposed for accurately identifying regions of interest to count vehicles in parking lot images. The proposed system achieved 99.68% balanced accuracy on a custom dataset of 3,484 images, offering a cost-effective smart parking solution that ensures precise vehicle detection while preserving data privacy
Authors: Marzieh Mohammadi, Amir Salarpour
Abstract: This paper introduces Point-GN, a novel non-parametric network for efficient and accurate 3D point cloud classification. Unlike conventional deep learning models that rely on a large number of trainable parameters, Point-GN leverages non-learnable components-specifically, Farthest Point Sampling (FPS), k-Nearest Neighbors (k-NN), and Gaussian Positional Encoding (GPE)-to extract both local and global geometric features. This design eliminates the need for additional training while maintaining high performance, making Point-GN particularly suited for real-time, resource-constrained applications. We evaluate Point-GN on two benchmark datasets, ModelNet40 and ScanObjectNN, achieving classification accuracies of 85.29% and 85.89%, respectively, while significantly reducing computational complexity. Point-GN outperforms existing non-parametric methods and matches the performance of fully trained models, all with zero learnable parameters. Our results demonstrate that Point-GN is a promising solution for 3D point cloud classification in practical, real-time environments.
Authors: August Leander H{\o}eg, Sophia W. Bardenfleth, Hans Martin Kjer, Tim B. Dyrby, Vedrana Andersen Dahl, Anders Dahl
Abstract: Until now, it has been difficult for volumetric super-resolution to utilize the recent advances in transformer-based models seen in 2D super-resolution. The memory required for self-attention in 3D volumes limits the receptive field. Therefore, long-range interactions are not used in 3D to the extent done in 2D and the strength of transformers is not realized. We propose a multi-scale transformer-based model based on hierarchical attention blocks combined with carrier tokens at multiple scales to overcome this. Here information from larger regions at coarse resolution is sequentially carried on to finer-resolution regions to predict the super-resolved image. Using transformer layers at each resolution, our coarse-to-fine modeling limits the number of tokens at each scale and enables attention over larger regions than what has previously been possible. We experimentally compare our method, MTVNet, against state-of-the-art volumetric super-resolution models on five 3D datasets demonstrating the advantage of an increased receptive field. This advantage is especially pronounced for images that are larger than what is seen in popularly used 3D datasets. Our code is available at https://github.com/AugustHoeg/MTVNet
Authors: Ao Wang, Hui Chen, Jianchao Tan, Kefeng Zhang, Xunliang Cai, Zijia Lin, Jungong Han, Guiguang Ding
Abstract: Recently, large vision-language models (LVLMs) have rapidly gained popularity for their strong generation and reasoning capabilities given diverse multimodal inputs. However, these models incur significant computational and memory overhead during inference, which greatly hinders the efficient deployment in practical scenarios. The extensive key-value (KV) cache, necessitated by the lengthy input and output sequences, notably contributes to the high inference cost. Based on this, recent works have investigated ways to reduce the KV cache size for higher efficiency. Although effective, they generally overlook the distinct importance distributions of KV vectors across layers and maintain the same cache size for each layer during the next token prediction. This results in the significant contextual information loss for certain layers, leading to notable performance decline. To address this, we present PrefixKV. It reframes the challenge of determining KV cache sizes for all layers into the task of searching for the optimal global prefix configuration. With an adaptive layer-wise KV retention recipe based on binary search, the maximum contextual information can thus be preserved in each layer, facilitating the generation. Extensive experiments demonstrate that our method achieves the state-of-the-art performance compared with others. It exhibits superior inference efficiency and generation quality trade-offs, showing promising potential for practical applications. Code is available at \url{https://github.com/THU-MIG/PrefixKV}.
Authors: Hannan Lu, Xiaohe Wu, Shudong Wang, Xiameng Qin, Xinyu Zhang, Junyu Han, Wangmeng Zuo, Ji Tao
Abstract: Generating multi-view videos for autonomous driving training has recently gained much attention, with the challenge of addressing both cross-view and cross-frame consistency. Existing methods typically apply decoupled attention mechanisms for spatial, temporal, and view dimensions. However, these approaches often struggle to maintain consistency across dimensions, particularly when handling fast-moving objects that appear at different times and viewpoints. In this paper, we present CogDriving, a novel network designed for synthesizing high-quality multi-view driving videos. CogDriving leverages a Diffusion Transformer architecture with holistic-4D attention modules, enabling simultaneous associations across the spatial, temporal, and viewpoint dimensions. We also propose a lightweight controller tailored for CogDriving, i.e., Micro-Controller, which uses only 1.1% of the parameters of the standard ControlNet, enabling precise control over Bird's-Eye-View layouts. To enhance the generation of object instances crucial for autonomous driving, we propose a re-weighted learning objective, dynamically adjusting the learning weights for object instances during training. CogDriving demonstrates strong performance on the nuScenes validation set, achieving an FVD score of 37.8, highlighting its ability to generate realistic driving videos. The project can be found at https://luhannan.github.io/CogDrivingPage/.
Authors: Mahtab Bigverdi, Zelun Luo, Cheng-Yu Hsieh, Ethan Shen, Dongping Chen, Linda G. Shapiro, Ranjay Krishna
Abstract: Multimodal language models (MLMs) still face challenges in fundamental visual perception tasks where specialized models excel. Tasks requiring reasoning about 3D structures benefit from depth estimation, and reasoning about 2D object instances benefits from object detection. Yet, MLMs can not produce intermediate depth or boxes to reason over. Finetuning MLMs on relevant data doesn't generalize well and outsourcing computation to specialized vision tools is too compute-intensive and memory-inefficient. To address this, we introduce Perception Tokens, intrinsic image representations designed to assist reasoning tasks where language is insufficient. Perception tokens act as auxiliary reasoning tokens, akin to chain-of-thought prompts in language models. For example, in a depth-related task, an MLM augmented with perception tokens can reason by generating a depth map as tokens, enabling it to solve the problem effectively. We propose AURORA, a training method that augments MLMs with perception tokens for improved reasoning over visual inputs. AURORA leverages a VQVAE to transform intermediate image representations, such as depth maps into a tokenized format and bounding box tokens, which is then used in a multi-task training framework. AURORA achieves notable improvements across counting benchmarks: +10.8% on BLINK, +11.3% on CVBench, and +8.3% on SEED-Bench, outperforming finetuning approaches in generalization across datasets. It also improves on relative depth: over +6% on BLINK. With perception tokens, AURORA expands the scope of MLMs beyond language-based reasoning, paving the way for more effective visual reasoning capabilities.
Authors: Jayaprakash Sundararaj, Akhil Vyas, Benjamin Gonzalez-Maldonado
Abstract: Transforming mathematical expressions into LaTeX poses a significant challenge. In this paper, we examine the application of advanced transformer-based architectures to address the task of converting handwritten or digital mathematical expression images into corresponding LaTeX code. As a baseline, we utilize the current state-of-the-art CNN encoder and LSTM decoder. Additionally, we explore enhancements to the CNN-RNN architecture by replacing the CNN encoder with the pretrained ResNet50 model with modification to suite the grey scale input. Further, we experiment with vision transformer model and compare with Baseline and CNN-LSTM model. Our findings reveal that the vision transformer architectures outperform the baseline CNN-RNN framework, delivering higher overall accuracy and BLEU scores while achieving lower Levenshtein distances. Moreover, these results highlight the potential for further improvement through fine-tuning of model parameters. To encourage open research, we also provide the model implementation, enabling reproduction of our results and facilitating further research in this domain.
Authors: Th\'eo Sourget, Michelle Hestbek-M{\o}ller, Amelia Jim\'enez-S\'anchez, Jack Junchi Xu, Veronika Cheplygina
Abstract: The development of larger models for medical image analysis has led to increased performance. However, it also affected our ability to explain and validate model decisions. Models can use non-relevant parts of images, also called spurious correlations or shortcuts, to obtain high performance on benchmark datasets but fail in real-world scenarios. In this work, we challenge the capacity of convolutional neural networks (CNN) to classify chest X-rays and eye fundus images while masking out clinically relevant parts of the image. We show that all models trained on the PadChest dataset, irrespective of the masking strategy, are able to obtain an Area Under the Curve (AUC) above random. Moreover, the models trained on full images obtain good performance on images without the region of interest (ROI), even superior to the one obtained on images only containing the ROI. We also reveal a possible spurious correlation in the Chaksu dataset while the performances are more aligned with the expectation of an unbiased model. We go beyond the performance analysis with the usage of the explainability method SHAP and the analysis of embeddings. We asked a radiology resident to interpret chest X-rays under different masking to complement our findings with clinical knowledge. Our code is available at https://github.com/TheoSourget/MMC_Masking and https://github.com/TheoSourget/MMC_Masking_EyeFundus
URLs: https://github.com/TheoSourget/MMC_Masking, https://github.com/TheoSourget/MMC_Masking_EyeFundus
Authors: Yayuan Li, Zhi Cao, Jason J. Corso
Abstract: Despite the recent strides in video generation, state-of-the-art methods still struggle with elements of visual detail. One particularly challenging case is the class of egocentric instructional videos in which the intricate motion of the hand coupled with a mostly stable and non-distracting environment is necessary to convey the appropriate visual action instruction. To address these challenges, we introduce a new method for instructional video generation. Our diffusion-based method incorporates two distinct innovations. First, we propose an automatic method to generate the expected region of motion, guided by both the visual context and the action text. Second, we introduce a critical hand structure loss to guide the diffusion model to focus on smooth and consistent hand poses. We evaluate our method on augmented instructional datasets based on EpicKitchens and Ego4D, demonstrating significant improvements over state-of-the-art methods in terms of instructional clarity, especially of the hand motion in the target region, across diverse environments and actions. Video results can be found on the project webpage: https://excitedbutter.github.io/Instructional-Video-Generation/
URLs: https://excitedbutter.github.io/Instructional-Video-Generation/
Authors: Ziwei Huang, Wanggui He, Quanyu Long, Yandi Wang, Haoyuan Li, Zhelun Yu, Fangxun Shu, Long Chan, Hao Jiang, Leilei Gan, Fei Wu
Abstract: Evaluating the quality of synthesized images remains a significant challenge in the development of text-to-image (T2I) generation. Most existing studies in this area primarily focus on evaluating text-image alignment, image quality, and object composition capabilities, with comparatively fewer studies addressing the evaluation of the factuality of T2I models, particularly when the concepts involved are knowledge-intensive. To mitigate this gap, we present T2I-FactualBench in this work - the largest benchmark to date in terms of the number of concepts and prompts specifically designed to evaluate the factuality of knowledge-intensive concept generation. T2I-FactualBench consists of a three-tiered knowledge-intensive text-to-image generation framework, ranging from the basic memorization of individual knowledge concepts to the more complex composition of multiple knowledge concepts. We further introduce a multi-round visual question answering (VQA) based evaluation framework to assess the factuality of three-tiered knowledge-intensive text-to-image generation tasks. Experiments on T2I-FactualBench indicate that current state-of-the-art (SOTA) T2I models still leave significant room for improvement.
Authors: Trong-Tung Nguyen, Quang Nguyen, Khoi Nguyen, Anh Tran, Cuong Pham
Abstract: Recent advances in text-guided image editing enable users to perform image edits through simple text inputs, leveraging the extensive priors of multi-step diffusion-based text-to-image models. However, these methods often fall short of the speed demands required for real-world and on-device applications due to the costly multi-step inversion and sampling process involved. In response to this, we introduce SwiftEdit, a simple yet highly efficient editing tool that achieve instant text-guided image editing (in 0.23s). The advancement of SwiftEdit lies in its two novel contributions: a one-step inversion framework that enables one-step image reconstruction via inversion and a mask-guided editing technique with our proposed attention rescaling mechanism to perform localized image editing. Extensive experiments are provided to demonstrate the effectiveness and efficiency of SwiftEdit. In particular, SwiftEdit enables instant text-guided image editing, which is extremely faster than previous multi-step methods (at least 50 times faster) while maintain a competitive performance in editing results. Our project page is at: https://swift-edit.github.io/
Authors: Yassine Ouali, Adrian Bulat, Alexandros Xenos, Anestis Zaganidis, Ioannis Maniadis Metaxas, Brais Martinez, Georgios Tzimiropoulos
Abstract: Contrastively-trained Vision-Language Models (VLMs) like CLIP have become the de facto approach for discriminative vision-language representation learning. However, these models have limited language understanding, often exhibiting a "bag of words" behavior. At the same time, Large Vision-Language Models (LVLMs), which combine vision encoders with LLMs, have been shown capable of detailed vision-language reasoning, yet their autoregressive nature renders them less suitable for discriminative tasks. In this work, we propose to combine "the best of both worlds": a new training approach for discriminative fine-tuning of LVLMs that results in strong discriminative and compositional capabilities. Essentially, our approach converts a generative LVLM into a discriminative one, unlocking its capability for powerful image-text discrimination combined with enhanced language understanding. Our contributions include: (1) A carefully designed training/optimization framework that utilizes image-text pairs of variable length and granularity for training the model with both contrastive and next-token prediction losses. This is accompanied by ablation studies that justify the necessity of our framework's components. (2) A parameter-efficient adaptation method using a combination of soft prompting and LoRA adapters. (3) Significant improvements over state-of-the-art CLIP-like models of similar size, including standard image-text retrieval benchmarks and notable gains in compositionality.
Authors: Shota Sasaki, Jane Wu, Ko Nishino
Abstract: This paper introduces a novel clothed human model that can be learned from multiview RGB videos, with a particular emphasis on recovering physically accurate body and cloth movements. Our method, Position Based Dynamic Gaussians (PBDyG), realizes ``movement-dependent'' cloth deformation via physical simulation, rather than merely relying on ``pose-dependent'' rigid transformations. We model the clothed human holistically but with two distinct physical entities in contact: clothing modeled as 3D Gaussians, which are attached to a skinned SMPL body that follows the movement of the person in the input videos. The articulation of the SMPL body also drives physically-based simulation of the clothes' Gaussians to transform the avatar to novel poses. In order to run position based dynamics simulation, physical properties including mass and material stiffness are estimated from the RGB videos through Dynamic 3D Gaussian Splatting. Experiments demonstrate that our method not only accurately reproduces appearance but also enables the reconstruction of avatars wearing highly deformable garments, such as skirts or coats, which have been challenging to reconstruct using existing methods.
Authors: Zhengqi Li, Richard Tucker, Forrester Cole, Qianqian Wang, Linyi Jin, Vickie Ye, Angjoo Kanazawa, Aleksander Holynski, Noah Snavely
Abstract: We present a system that allows for accurate, fast, and robust estimation of camera parameters and depth maps from casual monocular videos of dynamic scenes. Most conventional structure from motion and monocular SLAM techniques assume input videos that feature predominantly static scenes with large amounts of parallax. Such methods tend to produce erroneous estimates in the absence of these conditions. Recent neural network-based approaches attempt to overcome these challenges; however, such methods are either computationally expensive or brittle when run on dynamic videos with uncontrolled camera motion or unknown field of view. We demonstrate the surprising effectiveness of a deep visual SLAM framework: with careful modifications to its training and inference schemes, this system can scale to real-world videos of complex dynamic scenes with unconstrained camera paths, including videos with little camera parallax. Extensive experiments on both synthetic and real videos demonstrate that our system is significantly more accurate and robust at camera pose and depth estimation when compared with prior and concurrent work, with faster or comparable running times. See interactive results on our project page: https://mega-sam.github.io/
Authors: Gaurav Shrivastava, Abhinav Shrivastava
Abstract: Diffusion models have made significant strides in image generation, mastering tasks such as unconditional image synthesis, text-image translation, and image-to-image conversions. However, their capability falls short in the realm of video prediction, mainly because they treat videos as a collection of independent images, relying on external constraints such as temporal attention mechanisms to enforce temporal coherence. In our paper, we introduce a novel model class, that treats video as a continuous multi-dimensional process rather than a series of discrete frames. We also report a reduction of 75\% sampling steps required to sample a new frame thus making our framework more efficient during the inference time. Through extensive experimentation, we establish state-of-the-art performance in video prediction, validated on benchmark datasets including KTH, BAIR, Human3.6M, and UCF101. Navigate to the project page https://www.cs.umd.edu/~gauravsh/cvp/supp/website.html for video results.
URLs: https://www.cs.umd.edu/
Authors: Gaurav Shrivastava, Ser-Nam Lim, Abhinav Shrivastava
Abstract: In the evolving landscape of video enhancement and editing methodologies, a majority of deep learning techniques often rely on extensive datasets of observed input and ground truth sequence pairs for optimal performance. Such reliance often falters when acquiring data becomes challenging, especially in tasks like video dehazing and relighting, where replicating identical motions and camera angles in both corrupted and ground truth sequences is complicated. Moreover, these conventional methodologies perform best when the test distribution closely mirrors the training distribution. Recognizing these challenges, this paper introduces a novel video decomposition prior `VDP' framework which derives inspiration from professional video editing practices. Our methodology does not mandate task-specific external data corpus collection, instead pivots to utilizing the motion and appearance of the input video. VDP framework decomposes a video sequence into a set of multiple RGB layers and associated opacity levels. These set of layers are then manipulated individually to obtain the desired results. We addresses tasks such as video object segmentation, dehazing, and relighting. Moreover, we introduce a novel logarithmic video decomposition formulation for video relighting tasks, setting a new benchmark over the existing methodologies. We observe the property of relighting emerge as we optimize for our novel relighting decomposition formulation. We evaluate our approach on standard video datasets like DAVIS, REVIDE, & SDSD and show qualitative results on a diverse array of internet videos. Project Page - https://www.cs.umd.edu/~gauravsh/video_decomposition/index.html for video results.
URLs: https://www.cs.umd.edu/
Authors: Lu Wang, Dong-xue Liang, Xiao-lei Yin, Jing Qiu, Zhi-yun Yang, Jun-hui Xing, Jian-zeng Dong, Zhao-yuan Ma
Abstract: The reconstruction of three-dimensional models of coronary arteries is of great significance for the localization, evaluation and diagnosis of stenosis and plaque in the arteries, as well as for the assisted navigation of interventional surgery. In the clinical practice, physicians use a few angles of coronary angiography to capture arterial images, so it is of great practical value to perform 3D reconstruction directly from coronary angiography images. However, this is a very difficult computer vision task due to the complex shape of coronary blood vessels, as well as the lack of data set and key point labeling. With the rise of deep learning, more and more work is being done to reconstruct 3D models of human organs from medical images using deep neural networks. We propose an adversarial and generative way to reconstruct three dimensional coronary artery models, from two different views of angiographic images of coronary arteries. With 3D fully supervised learning and 2D weakly supervised learning schemes, we obtained reconstruction accuracies that outperform state-of-art techniques.
Authors: Miao Li, Wenhao Ding, Ding Zhao
Abstract: The prominence of embodied Artificial Intelligence (AI), which empowers robots to navigate, perceive, and engage within virtual environments, has attracted significant attention, owing to the remarkable advances in computer vision and large language models. Privacy emerges as a pivotal concern within the realm of embodied AI, as the robot accesses substantial personal information. However, the issue of privacy leakage in embodied AI tasks, particularly concerning reinforcement learning algorithms, has not received adequate consideration in research. This paper aims to address this gap by proposing an attack on the training process of the value-based algorithm and the gradient-based algorithm, utilizing gradient inversion to reconstruct states, actions, and supervisory signals. The choice of using gradients for the attack is motivated by the fact that commonly employed federated learning techniques solely utilize gradients computed based on private user data to optimize models, without storing or transmitting the data to public servers. Nevertheless, these gradients contain sufficient information to potentially expose private data. To validate our approach, we conducted experiments on the AI2THOR simulator and evaluated our algorithm on active perception, a prevalent task in embodied AI. The experimental results demonstrate the effectiveness of our method in successfully reconstructing all information from the data in 120 room layouts. Check our website for videos.
Authors: Michael W. Spratling
Abstract: Reliable and robust evaluation methods are a necessary first step towards developing machine learning models that are themselves robust and reliable. Unfortunately, current evaluation protocols typically used to assess classifiers fail to comprehensively evaluate performance as they tend to rely on limited types of test data, and ignore others. For example, using the standard test data fails to evaluate the predictions made by the classifier to samples from classes it was not trained on. On the other hand, testing with data containing samples from unknown classes fails to evaluate how well the classifier can predict the labels for known classes. This article advocates bench-marking performance using a wide range of different types of data and using a single metric that can be applied to all such data types to produce a consistent evaluation of performance. Using such a benchmark it is found that current deep neural networks, including those trained with methods that are believed to produce state-of-the-art robustness, are extremely vulnerable to making mistakes on certain types of data. This means that such models will be unreliable in real-world scenarios where they may encounter data from many different domains, and that they are insecure as they can easily be fooled into making the wrong decisions. It is hoped that these results will motivate the wider adoption of more comprehensive testing methods that will, in turn, lead to the development of more robust machine learning methods in the future. Code is available at: https://codeberg.org/mwspratling/RobustnessEvaluation
Authors: Jee Seok Yoon, Kwanseok Oh, Yooseung Shin, Maciej A. Mazurowski, Heung-Il Suk
Abstract: Medical image analysis (MedIA) has become an essential tool in medicine and healthcare, aiding in disease diagnosis, prognosis, and treatment planning, and recent successes in deep learning (DL) have made significant contributions to its advances. However, deploying DL models for MedIA in real-world situations remains challenging due to their failure to generalize across the distributional gap between training and testing samples - a problem known as domain shift. Researchers have dedicated their efforts to developing various DL methods to adapt and perform robustly on unknown and out-of-distribution (OOD) data distributions. This article comprehensively reviews domain generalization (DG) studies specifically tailored for MedIA. We provide a holistic view of how DG techniques interact within the broader MedIA system, going beyond methodologies to consider the operational implications on the entire MedIA workflow. Specifically, we categorize DG methods into data-level, feature-level, model-level, and analysis-level methods. We show how those methods can be used in various stages of the MedIA workflow with DL equipped from data acquisition to model prediction and analysis. Furthermore, we critically analyze the strengths and weaknesses of various methods, unveiling future research opportunities.
Authors: Shivam Gupta, Aditya Parulekar, Eric Price, Zhiyang Xun
Abstract: Diffusion models have become the most popular approach to deep generative modeling of images, largely due to their empirical performance and reliability. From a theoretical standpoint, a number of recent works~\cite{chen2022,chen2022improved,benton2023linear} have studied the iteration complexity of sampling, assuming access to an accurate diffusion model. In this work, we focus on understanding the \emph{sample complexity} of training such a model; how many samples are needed to learn an accurate diffusion model using a sufficiently expressive neural network? Prior work~\cite{BMR20} showed bounds polynomial in the dimension, desired Total Variation error, and Wasserstein error. We show an \emph{exponential improvement} in the dependence on Wasserstein error and depth, along with improved dependencies on other relevant parameters.
Authors: Lorand Vatamany, Siamak Mehrkanoon
Abstract: Accurate precipitation nowcasting is crucial for applications such as flood prediction, disaster management, agriculture optimization, and transportation management. While many studies have approached this task using sequence-to-sequence models, most focus on single regions, ignoring correlations between disjoint areas. We reformulate precipitation nowcasting as a spatiotemporal graph sequence problem. Specifically, we propose Graph Dual-stream Convolutional Attention Fusion, a novel extension of the graph attention network. Our model's dual-stream design employs distinct attention mechanisms for spatial and temporal interactions, capturing their unique dynamics. A gated fusion module integrates both streams, leveraging spatial and temporal information for improved predictive accuracy. Additionally, our framework enhances graph attention by directly processing three-dimensional tensors within graph nodes, removing the need for reshaping. This capability enables handling complex, high-dimensional data and exploiting higher-order correlations between data dimensions. Depthwise-separable convolutions are also incorporated to refine local feature extraction and efficiently manage high-dimensional inputs. We evaluate our model using seven years of precipitation data from Copernicus Climate Change Services, covering Europe and neighboring regions. Experimental results demonstrate superior performance of our approach compared to other models. Moreover, visualizations of seasonal spatial and temporal attention scores provide insights into the most significant connections between regions and time steps.
Authors: Yumeng Liu, Yaxun Yang, Youzhuo Wang, Xiaofei Wu, Jiamin Wang, Yichen Yao, S\"oren Schwertfeger, Sibei Yang, Wenping Wang, Jingyi Yu, Xuming He, Yuexin Ma
Abstract: In this paper, we introduce RealDex, a pioneering dataset capturing authentic dexterous hand grasping motions infused with human behavioral patterns, enriched by multi-view and multimodal visual data. Utilizing a teleoperation system, we seamlessly synchronize human-robot hand poses in real time. This collection of human-like motions is crucial for training dexterous hands to mimic human movements more naturally and precisely. RealDex holds immense promise in advancing humanoid robot for automated perception, cognition, and manipulation in real-world scenarios. Moreover, we introduce a cutting-edge dexterous grasping motion generation framework, which aligns with human experience and enhances real-world applicability through effectively utilizing Multimodal Large Language Models. Extensive experiments have demonstrated the superior performance of our method on RealDex and other open datasets. The complete dataset and code will be made available upon the publication of this work.
Authors: Letitia Parcalabescu, Anette Frank
Abstract: Vision and language model (VLM) decoders are currently the best-performing architectures on multimodal tasks. Next to answers, they are able to produce natural language explanations, either in post-hoc or CoT settings. However, it is not clear to what extent they are using the input vision and text modalities when generating answers or explanations. In this work, we investigate if VLMs rely on their input modalities differently when they produce explanations as opposed to answers. We also evaluate the self-consistency of VLM decoders in both post-hoc and CoT explanation settings, by extending existing unimodal tests and measures to VLM decoders. We find that most tested VLMs are less self-consistent than LLMs. Text contributions in all tested VL decoders are more important than image contributions in all examined tasks. However, when comparing explanation generation to answer generation, the contributions of images are significantly stronger for generating explanations compared to answers. This difference is even larger in CoT compared to post-hoc explanations. Lastly, we provide an up-to-date benchmarking of state-of-the-art VL decoders on the VALSE benchmark, which before was restricted to VL encoders. We find that the tested VL decoders still struggle with most phenomena tested by VALSE.
Authors: Yongxian Wei, Zixuan Hu, Zhenyi Wang, Li Shen, Chun Yuan, Dacheng Tao
Abstract: Data-Free Meta-Learning (DFML) aims to extract knowledge from a collection of pre-trained models without requiring the original data, presenting practical benefits in contexts constrained by data privacy concerns. Current DFML methods primarily focus on the data recovery from these pre-trained models. However, they suffer from slow recovery speed and overlook gaps inherent in heterogeneous pre-trained models. In response to these challenges, we introduce the Faster and Better Data-Free Meta-Learning (FREE) framework, which contains: (i) a meta-generator for rapidly recovering training tasks from pre-trained models; and (ii) a meta-learner for generalizing to new unseen tasks. Specifically, within the module Faster Inversion via Meta-Generator, each pre-trained model is perceived as a distinct task. The meta-generator can rapidly adapt to a specific task in just five steps, significantly accelerating the data recovery. Furthermore, we propose Better Generalization via Meta-Learner and introduce an implicit gradient alignment algorithm to optimize the meta-learner. This is achieved as aligned gradient directions alleviate potential conflicts among tasks from heterogeneous pre-trained models. Empirical experiments on multiple benchmarks affirm the superiority of our approach, marking a notable speed-up (20$\times$) and performance enhancement (1.42%$\sim$4.78%) in comparison to the state-of-the-art.
Authors: Chen Feng, Duolikun Danier, Fan Zhang, Alex Mackin, Andrew Collins, David Bull
Abstract: Visual artifacts are often introduced into streamed video content, due to prevailing conditions during content production and delivery. Since these can degrade the quality of the user's experience, it is important to automatically and accurately detect them in order to enable effective quality measurement and enhancement. Existing detection methods often focus on a single type of artifact and/or determine the presence of an artifact through thresholding objective quality indices. Such approaches have been reported to offer inconsistent prediction performance and are also impractical for real-world applications where multiple artifacts co-exist and interact. In this paper, we propose a Multiple Visual Artifact Detector, MVAD, for video streaming which, for the first time, is able to detect multiple artifacts using a single framework that is not reliant on video quality assessment models. Our approach employs a new Artifact-aware Dynamic Feature Extractor (ADFE) to obtain artifact-relevant spatial features within each frame for multiple artifact types. The extracted features are further processed by a Recurrent Memory Vision Transformer (RMViT) module, which captures both short-term and long-term temporal information within the input video. The proposed network architecture is optimized in an end-to-end manner based on a new, large and diverse training database that is generated by simulating the video streaming pipeline and based on Adversarial Data Augmentation. This model has been evaluated on two video artifact databases, Maxwell and BVI-Artifact, and achieves consistent and improved prediction results for ten target visual artifacts when compared to seven existing single and multiple artifact detectors. The source code and training database will be available at https://chenfeng-bristol.github.io/MVAD/.
Authors: Shuhan Xiao, Lukas Klein, Jens Petersen, Philipp Vollmuth, Paul F. Jaeger, Klaus H. Maier-Hein
Abstract: Identifying predictive covariates, which forecast individual treatment effectiveness, is crucial for decision-making across different disciplines such as personalized medicine. These covariates, referred to as biomarkers, are extracted from pre-treatment data, often within randomized controlled trials, and should be distinguished from prognostic biomarkers, which are independent of treatment assignment. Our study focuses on discovering predictive imaging biomarkers, specific image features, by leveraging pre-treatment images to uncover new causal relationships. Unlike labor-intensive approaches relying on handcrafted features prone to bias, we present a novel task of directly learning predictive features from images. We propose an evaluation protocol to assess a model's ability to identify predictive imaging biomarkers and differentiate them from purely prognostic ones by employing statistical testing and a comprehensive analysis of image feature attribution. We explore the suitability of deep learning models originally developed for estimating the conditional average treatment effect (CATE) for this task, which have been assessed primarily for their precision of CATE estimation while overlooking the evaluation of imaging biomarker discovery. Our proof-of-concept analysis demonstrates the feasibility and potential of our approach in discovering and validating predictive imaging biomarkers from synthetic outcomes and real-world image datasets. Our code is available at \url{https://github.com/MIC-DKFZ/predictive_image_biomarker_analysis}.
URLs: https://github.com/MIC-DKFZ/predictive_image_biomarker_analysis
Authors: Pandiyaraju V, Shravan Venkatraman, Pavan Kumar S, Santhosh Malarvannan, Kannan A
Abstract: Breast cancer is one of the leading causes of death globally, and thus there is an urgent need for early and accurate diagnostic techniques. Although ultrasound imaging is a widely used technique for breast cancer screening, it faces challenges such as poor boundary delineation caused by variations in tumor morphology and reduced diagnostic accuracy due to inconsistent image quality. To address these challenges, we propose novel Deep Learning (DL) frameworks for breast lesion segmentation and classification. We introduce a precision mapping mechanism (PMM) for a precision mapping and attention-driven LinkNet (PMAD-LinkNet) segmentation framework that dynamically adapts spatial mappings through morphological variation analysis, enabling precise pixel-level refinement of tumor boundaries. Subsequently, we introduce a component-specific feature enhancement module (CSFEM) for a component-specific feature-enhanced classifier (CSFEC-Net). Through a multi-level attention approach, the CSFEM magnifies distinguishing features of benign, malignant, and normal tissues. The proposed frameworks are evaluated against existing literature and a diverse set of state-of-the-art Convolutional Neural Network (CNN) architectures. The obtained results show that our segmentation model achieves an accuracy of 98.1%, an IoU of 96.9%, and a Dice Coefficient of 97.2%. For the classification model, an accuracy of 99.2% is achieved with F1-score, precision, and recall values of 99.1%, 99.3%, and 99.1%, respectively.
Authors: Shraman Pramanick, Rama Chellappa, Subhashini Venugopalan
Abstract: Seeking answers to questions within long scientific research articles is a crucial area of study that aids readers in quickly addressing their inquiries. However, existing question-answering (QA) datasets based on scientific papers are limited in scale and focus solely on textual content. We introduce SPIQA (Scientific Paper Image Question Answering), the first large-scale QA dataset specifically designed to interpret complex figures and tables within the context of scientific research articles across various domains of computer science. Leveraging the breadth of expertise and ability of multimodal large language models (MLLMs) to understand figures, we employ automatic and manual curation to create the dataset. We craft an information-seeking task on interleaved images and text that involves multiple images covering plots, charts, tables, schematic diagrams, and result visualizations. SPIQA comprises 270K questions divided into training, validation, and three different evaluation splits. Through extensive experiments with 12 prominent foundational models, we evaluate the ability of current multimodal systems to comprehend the nuanced aspects of research articles. Additionally, we propose a Chain-of-Thought (CoT) evaluation strategy with in-context retrieval that allows fine-grained, step-by-step assessment and improves model performance. We further explore the upper bounds of performance enhancement with additional textual information, highlighting its promising potential for future research and the dataset's impact on revolutionizing how we interact with scientific literature.
Authors: Christoph Auer, Maksym Lysak, Ahmed Nassar, Michele Dolfi, Nikolaos Livathinos, Panos Vagenas, Cesar Berrospi Ramis, Matteo Omenetti, Fabian Lindlbauer, Kasper Dinkla, Lokesh Mishra, Yusik Kim, Shubham Gupta, Rafael Teixeira de Lima, Valery Weber, Lucas Morin, Ingmar Meijer, Viktor Kuropiatnyk, Peter W. J. Staar
Abstract: This technical report introduces Docling, an easy to use, self-contained, MIT-licensed open-source package for PDF document conversion. It is powered by state-of-the-art specialized AI models for layout analysis (DocLayNet) and table structure recognition (TableFormer), and runs efficiently on commodity hardware in a small resource budget. The code interface allows for easy extensibility and addition of new features and models.
Authors: Peter Neidlinger, Omar S. M. El Nahhas, Hannah Sophie Muti, Tim Lenz, Michael Hoffmeister, Hermann Brenner, Marko van Treeck, Rupert Langer, Bastian Dislich, Hans Michael Behrens, Christoph R\"ocken, Sebastian Foersch, Daniel Truhn, Antonio Marra, Oliver Lester Saldanha, Jakob Nikolas Kather
Abstract: Advancements in artificial intelligence have driven the development of numerous pathology foundation models capable of extracting clinically relevant information. However, there is currently limited literature independently evaluating these foundation models on truly external cohorts and clinically-relevant tasks to uncover adjustments for future improvements. In this study, we benchmarked 19 histopathology foundation models on 13 patient cohorts with 6,818 patients and 9,528 slides from lung, colorectal, gastric, and breast cancers. The models were evaluated on weakly-supervised tasks related to biomarkers, morphological properties, and prognostic outcomes. We show that a vision-language foundation model, CONCH, yielded the highest performance when compared to vision-only foundation models, with Virchow2 as close second. The experiments reveal that foundation models trained on distinct cohorts learn complementary features to predict the same label, and can be fused to outperform the current state of the art. An ensemble combining CONCH and Virchow2 predictions outperformed individual models in 55% of tasks, leveraging their complementary strengths in classification scenarios. Moreover, our findings suggest that data diversity outweighs data volume for foundation models. Our work highlights actionable adjustments to improve pathology foundation models.
Authors: Yi-Chia Chen, Wei-Hua Li, Cheng Sun, Yu-Chiang Frank Wang, Chu-Song Chen
Abstract: We introduce SAM4MLLM, an innovative approach which integrates the Segment Anything Model (SAM) with Multi-Modal Large Language Models (MLLMs) for pixel-aware tasks. Our method enables MLLMs to learn pixel-level location information without requiring excessive modifications to the existing model architecture or adding specialized tokens. We introduce an inquiry-based approach that can effectively find prompt points for SAM to perform segmentation based on MLLM. It combines detailed visual information with the powerful expressive capabilities of large language models in a unified language-based manner without additional computational overhead in learning. Experimental results on pubic benchmarks demonstrate the effectiveness of our approach.
Authors: Patrick Amadeus Irawan, Genta Indra Winata, Samuel Cahyawijaya, Ayu Purwarianti
Abstract: Natural Language Explanation (NLE) aims to elucidate the decision-making process by providing detailed, human-friendly explanations in natural language. It helps demystify the decision-making processes of large vision-language models (LVLMs) through the use of language models. While existing methods for creating a Vision Question-Answering with Natural Language Explanation (VQA-NLE) datasets can provide explanations, they heavily rely on human annotations that are time-consuming and costly. In this study, we propose a novel approach that leverages LVLMs to efficiently generate high-quality synthetic VQA-NLE datasets. By evaluating our synthetic data, we showcase how advanced prompting techniques can lead to the production of high-quality VQA-NLE data. Our findings indicate that this proposed method achieves up to 20x faster than human annotation, with only a minimal decrease in qualitative metrics, achieving robust quality that is nearly equivalent to human-annotated data. Furthermore, we show that incorporating visual prompts significantly enhances the relevance of text generation. Our study paves the way for a more efficient and robust automated generation of multi-modal NLE data, offering a promising solution to the problem.
Authors: Debargha Ganguly, Warren Morningstar, Andrew Yu, Vipin Chaudhary
Abstract: Generative models can now produce photorealistic synthetic data which is virtually indistinguishable from the real data used to train it. This is a significant evolution over previous models which could produce reasonable facsimiles of the training data, but ones which could be visually distinguished from the training data by human evaluation. Recent work on OOD detection has raised doubts that generative model likelihoods are optimal OOD detectors due to issues involving likelihood misestimation, entropy in the generative process, and typicality. We speculate that generative OOD detectors also failed because their models focused on the pixels rather than the semantic content of the data, leading to failures in near-OOD cases where the pixels may be similar but the information content is significantly different. We hypothesize that estimating typical sets using self-supervised learners leads to better OOD detectors. We introduce a novel approach that leverages representation learning, and informative summary statistics based on manifold estimation, to address all of the aforementioned issues. Our method outperforms other unsupervised approaches and achieves state-of-the art performance on well-established challenging benchmarks, and new synthetic data detection tasks.
Authors: Xiang Yue, Yueqi Song, Akari Asai, Seungone Kim, Jean de Dieu Nyandwi, Simran Khanuja, Anjali Kantharuban, Lintang Sutawika, Sathyanarayanan Ramamoorthy, Graham Neubig
Abstract: Despite recent advances in multimodal large language models (MLLMs), their development has predominantly focused on English- and western-centric datasets and tasks, leaving most of the world's languages and diverse cultural contexts underrepresented. This paper introduces Pangea, a multilingual multimodal LLM trained on PangeaIns, a diverse 6M instruction dataset spanning 39 languages. PangeaIns features: 1) high-quality English instructions, 2) carefully machine-translated instructions, and 3) culturally relevant multimodal tasks to ensure cross-cultural coverage. To rigorously assess models' capabilities, we introduce PangeaBench, a holistic evaluation suite encompassing 14 datasets covering 47 languages. Results show that Pangea significantly outperforms existing open-source models in multilingual settings and diverse cultural contexts. Ablation studies further reveal the importance of English data proportions, language popularity, and the number of multimodal training samples on overall performance. We fully open-source our data, code, and trained checkpoints, to facilitate the development of inclusive and robust multilingual MLLMs, promoting equity and accessibility across a broader linguistic and cultural spectrum.
Authors: Yang-Che Sun, Cheng Yu Yeo, Ernie Chu, Jun-Cheng Chen, Yu-Lun Liu
Abstract: In this work, we propose a unified representation for Super-Resolution (SR) and Image Compression, termed Factorized Fields, motivated by the shared principles between these two tasks. Both SISR and Image Compression require recovering and preserving fine image details--whether by enhancing resolution or reconstructing compressed data. Unlike previous methods that mainly focus on network architecture, our proposed approach utilizes a basis-coefficient decomposition to explicitly capture multi-scale visual features and structural components in images, addressing the core challenges of both tasks. We first derive our SR model, which includes a Coefficient Backbone and Basis Swin Transformer for generalizable Factorized Fields. Then, to further unify these two tasks, we leverage the strong information-recovery capabilities of the trained SR modules as priors in the compression pipeline, improving both compression efficiency and detail reconstruction. Additionally, we introduce a merged-basis compression branch that consolidates shared structures, further optimizing the compression process. Extensive experiments show that our unified representation delivers state-of-the-art performance, achieving an average relative improvement of 204.4% in PSNR over the baseline in Super-Resolution (SR) and 9.35% BD-rate reduction in Image Compression compared to the previous SOTA.
Authors: Daqian Bao, Alex Saad-Falcon, Justin Romberg
Abstract: Data acquisition in array signal processing (ASP) is costly because achieving high angular and range resolutions necessitates large antenna apertures and wide frequency bandwidths, respectively. The data requirements for ASP problems grow multiplicatively with the number of viewpoints and frequencies, significantly increasing the burden of data collection, even for simulation. Implicit Neural Representations (INRs) -- neural network-based models of 3D objects and scenes -- offer compact and continuous representations with minimal radar data. They can interpolate to unseen viewpoints and potentially address the sampling cost in ASP problems. In this work, we select Synthetic Aperture Radar (SAR) as a case from ASP and propose Radon Implicit Field Transform (RIFT). RIFT consists of two components: a classical forward model for radar (Generalized Radon Transform, GRT), and an INR based scene representation learned from radar signals. This method can be extended to other ASP problems by replacing the GRT with appropriate algorithms corresponding to different data modalities. In our experiments, we first synthesize radar data using the GRT. We then train the INR model on this synthetic data by minimizing the reconstruction error of the radar signal. After training, we render the scene using the trained INR and evaluate our scene representation against the ground truth scene. Due to the lack of existing benchmarks, we introduce two main new error metrics: phase-Root Mean Square Error (p-RMSE) for radar signal interpolation, and magnitude-Structural Similarity Index measure(m-SSIM) for scene reconstruction. These metrics adapt traditional error measures to account for the complex nature of radar signals. Compared to traditional scene models in radar signal processing, with only 10% data footprint, our RIFT model achieves up to 188% improvement in scene reconstruction.
Authors: Xiangyu Zhao, Zhiwang Zhou, Wenlong Zhang, Yihao Liu, Xiangyu Chen, Junchao Gong, Hao Chen, Ben Fei, Shiqi Chen, Wanli Ouyang, Xiao-Ming Wu, Lei Bai
Abstract: The Earth's weather system encompasses intricate weather data modalities and diverse weather understanding tasks, which hold significant value to human life. Existing data-driven models focus on single weather understanding tasks (e.g., weather forecasting). Although these models have achieved promising results, they fail to tackle various complex tasks within a single and unified model. Moreover, the paradigm that relies on limited real observations for a single scenario hinders the model's performance upper bound. In response to these limitations, we draw inspiration from the in-context learning paradigm employed in state-of-the-art visual foundation models and large language models. In this paper, we introduce the first generalist weather foundation model (WeatherGFM), designed to address a wide spectrum of weather understanding tasks in a unified manner. More specifically, we initially unify the representation and definition of the diverse weather understanding tasks. Subsequently, we devised weather prompt formats to manage different weather data modalities, namely single, multiple, and temporal modalities. Finally, we adopt a visual prompting question-answering paradigm for the training of unified weather understanding tasks. Extensive experiments indicate that our WeatherGFM can effectively handle up to ten weather understanding tasks, including weather forecasting, super-resolution, weather image translation, and post-processing. Our method also showcases generalization ability on unseen tasks.
Authors: Pranav Guruprasad, Harshvardhan Sikka, Jaewoo Song, Yangyue Wang, Paul Pu Liang
Abstract: Vision-language-action (VLA) models represent a promising direction for developing general-purpose robotic systems, demonstrating the ability to combine visual understanding, language comprehension, and action generation. However, systematic evaluation of these models across diverse robotic tasks remains limited. In this work, we present a comprehensive evaluation framework and benchmark suite for assessing VLA models. We profile three state-of-the-art VLM and VLAs - GPT-4o, OpenVLA, and JAT - across 20 diverse datasets from the Open-X-Embodiment collection, evaluating their performance on various manipulation tasks. Our analysis reveals several key insights: 1. current VLA models show significant variation in performance across different tasks and robot platforms, with GPT-4o demonstrating the most consistent performance through sophisticated prompt engineering, 2. all models struggle with complex manipulation tasks requiring multi-step planning, and 3. model performance is notably sensitive to action space characteristics and environmental factors. We release our evaluation framework and findings to facilitate systematic assessment of future VLA models and identify critical areas for improvement in the development of general purpose robotic systems.
Authors: Konstantinos Kontras, Thomas Strypsteen, Christos Chatzichristos, Paul Pu Liang, Matthew Blaschko, Maarten De Vos
Abstract: Multimodal learning can complete the picture of information extraction by uncovering key dependencies between data sources. However, current systems fail to fully leverage multiple modalities for optimal performance. This has been attributed to modality competition, where modalities strive for training resources, leaving some underoptimized. We show that current balancing methods struggle to train multimodal models that surpass even simple baselines, such as ensembles. This raises the question: how can we ensure that all modalities in multimodal training are sufficiently trained, and that learning from new modalities consistently improves performance? This paper proposes the Multimodal Competition Regularizer (MCR), a new loss component inspired by mutual information (MI) decomposition designed to prevent the adverse effects of competition in multimodal training. Our key contributions are: 1) Introducing game-theoretic principles in multimodal learning, where each modality acts as a player competing to maximize its influence on the final outcome, enabling automatic balancing of the MI terms. 2) Refining lower and upper bounds for each MI term to enhance the extraction of task-relevant unique and shared information across modalities. 3) Suggesting latent space permutations for conditional MI estimation, significantly improving computational efficiency. MCR outperforms all previously suggested training strategies and is the first to consistently improve multimodal learning beyond the ensemble baseline, clearly demonstrating that combining modalities leads to significant performance gains on both synthetic and large real-world datasets.
Authors: Ross Cutler, Babak Naderi, Vishak Gopal, Dharmendar Palle
Abstract: Photorealistic avatars are human avatars that look, move, and talk like real people. The performance of photorealistic avatars has significantly improved recently based on objective metrics such as PSNR, SSIM, LPIPS, FID, and FVD. However, recent photorealistic avatar publications do not provide subjective tests of the avatars to measure human usability factors. We provide an open source test framework to subjectively measure photorealistic avatar performance in ten dimensions: realism, trust, comfortableness using, comfortableness interacting with, appropriateness for work, creepiness, formality, affinity, resemblance to the person, and emotion accuracy. We show that the correlation of nine of these subjective metrics with PSNR, SSIM, LPIPS, FID, and FVD is weak, and moderate for emotion accuracy. The crowdsourced subjective test framework is highly reproducible and accurate when compared to a panel of experts. We analyze a wide range of avatars from photorealistic to cartoon-like and show that some photorealistic avatars are approaching real video performance based on these dimensions. We also find that for avatars above a certain level of realism, eight of these measured dimensions are strongly correlated. This means that avatars that are not as realistic as real video will have lower trust, comfortableness using, comfortableness interacting with, appropriateness for work, formality, and affinity, and higher creepiness compared to real video. In addition, because there is a strong linear relationship between avatar affinity and realism, there is no uncanny valley effect for photorealistic avatars in the telecommunication scenario. We provide several extensions of this test framework for future work and discuss design implications for telecommunication systems. The test framework is available at https://github.com/microsoft/P.910.
Authors: Meng Ye, Bingyu Xin, Bangwei Guo, Leon Axel, Dimitris Metaxas
Abstract: Multi-planar tagged MRI is the gold standard for regional heart wall motion evaluation. However, accurate recovery of the 3D true heart wall motion from a set of 2D apparent motion cues is challenging, due to incomplete sampling of the true motion and difficulty in information fusion from apparent motion cues observed on multiple imaging planes. To solve these challenges, we introduce a novel class of volumetric neural deformable models ($\upsilon$NDMs). Our $\upsilon$NDMs represent heart wall geometry and motion through a set of low-dimensional global deformation parameter functions and a diffeomorphic point flow regularized local deformation field. To learn such global and local deformation for 2D apparent motion mapping to 3D true motion, we design a hybrid point transformer, which incorporates both point cross-attention and self-attention mechanisms. While use of point cross-attention can learn to fuse 2D apparent motion cues into material point true motion hints, point self-attention hierarchically organised as an encoder-decoder structure can further learn to refine these hints and map them into 3D true motion. We have performed experiments on a large cohort of synthetic 3D regional heart wall motion dataset. The results demonstrated the high accuracy of our method for the recovery of dense 3D true motion from sparse 2D apparent motion cues. Project page is at https://github.com/DeepTag/VolumetricNeuralDeformableModels.
URLs: https://github.com/DeepTag/VolumetricNeuralDeformableModels.
Authors: Lehan He, Zeren Chen, Zhelun Shi, Tianyu Yu, Jing Shao, Lu Sheng
Abstract: Aligning the behaviors of Multimodal Large Language Models (MLLMs) with human preferences is crucial for developing robust and trustworthy AI systems. While recent attempts have employed human experts or powerful auxiliary AI systems to provide more accurate preference feedback, such as determining the preferable responses from MLLMs or directly rewriting hallucination-free responses, extensive resource overhead compromise the scalability of the feedback collection. In this work, we introduce Topic-level Preference Overwriting (TPO), a self-correctional approach that guide the model itself to mitigate its own hallucination at the topic level. Through a deconfounded strategy that replaces each topic within the response with the best or worst alternatives generated by the model itself, TPO creates more contrasting pairwise preference feedback, enhancing the feedback quality without human or proprietary model intervention. Notably, the experimental results demonstrate proposed TPO achieves state-of-the-art performance in trustworthiness, significantly reducing the object hallucinations by 92% and overall hallucinations by 38%. Code, model and dataset are available now.
Authors: Zhiyang Guo, Jinxu Xiang, Kai Ma, Wengang Zhou, Houqiang Li, Ran Zhang
Abstract: 3D characters are essential to modern creative industries, but making them animatable often demands extensive manual work in tasks like rigging and skinning. Existing automatic rigging tools face several limitations, including the necessity for manual annotations, rigid skeleton topologies, and limited generalization across diverse shapes and poses. An alternative approach is to generate animatable avatars pre-bound to a rigged template mesh. However, this method often lacks flexibility and is typically limited to realistic human shapes. To address these issues, we present Make-It-Animatable, a novel data-driven method to make any 3D humanoid model ready for character animation in less than one second, regardless of its shapes and poses. Our unified framework generates high-quality blend weights, bones, and pose transformations. By incorporating a particle-based shape autoencoder, our approach supports various 3D representations, including meshes and 3D Gaussian splats. Additionally, we employ a coarse-to-fine representation and a structure-aware modeling strategy to ensure both accuracy and robustness, even for characters with non-standard skeleton structures. We conducted extensive experiments to validate our framework's effectiveness. Compared to existing methods, our approach demonstrates significant improvements in both quality and speed.
Authors: Zhixuan Liang, Yao Mu, Yixiao Wang, Tianxing Chen, Wenqi Shao, Wei Zhan, Masayoshi Tomizuka, Ping Luo, Mingyu Ding
Abstract: Dexterous manipulation with contact-rich interactions is crucial for advanced robotics. While recent diffusion-based planning approaches show promise for simpler manipulation tasks, they often produce unrealistic ghost states (e.g., the object automatically moves without hand contact) or lack adaptability when handling complex sequential interactions. In this work, we introduce DexDiffuser, an interaction-aware diffusion planning framework for adaptive dexterous manipulation. DexDiffuser models joint state-action dynamics through a dual-phase diffusion process which consists of pre-interaction contact alignment and post-contact goal-directed control, enabling goal-adaptive generalizable dexterous manipulation. Additionally, we incorporate dynamics model-based dual guidance and leverage large language models for automated guidance function generation, enhancing generalizability for physical interactions and facilitating diverse goal adaptation through language cues. Experiments on physical interaction tasks such as door opening, pen and block re-orientation, and hammer striking demonstrate DexDiffuser's effectiveness on goals outside training distributions, achieving over twice the average success rate (59.2% vs. 29.5%) compared to existing methods. Our framework achieves 70.0% success on 30-degree door opening, 40.0% and 36.7% on pen and block half-side re-orientation respectively, and 46.7% on hammer nail half drive, highlighting its robustness and flexibility in contact-rich manipulation.
Authors: Yifan Zhu, Tianyi Xiang, Aaron Dollar, Zherong Pan
Abstract: Identifying predictive world models for robots in novel environments from sparse online observations is essential for robot task planning and execution in novel environments. However, existing methods that leverage differentiable simulators to identify world models are incapable of jointly optimizing the shape, appearance, and physical properties of the scene. In this work, we introduce a novel object representation that allows the joint identification of these properties. Our method employs a novel differentiable point-based object representation coupled with a grid-based appearance field, which allows differentiable object collision detection and rendering. Combined with a differentiable physical simulator, we achieve end-to-end optimization of world models, given the sparse visual and tactile observations of a physical motion sequence. Through a series of system identification tasks in simulated and real environments, we show that our method can learn both simulation- and rendering-ready world models from only one robot action sequence.
Authors: Razi Mahmood, Pingkun Yan, Diego Machado Reyes, Ge Wang, Mannudeep K. Kalra, Parisa Kaviani, Joy T. Wu, Tanveer Syeda-Mahmood
Abstract: Several evaluation metrics have been developed recently to automatically assess the quality of generative AI reports for chest radiographs based only on textual information using lexical, semantic, or clinical named entity recognition methods. In this paper, we develop a new method of report quality evaluation by first extracting fine-grained finding patterns capturing the location, laterality, and severity of a large number of clinical findings. We then performed phrasal grounding to localize their associated anatomical regions on chest radiograph images. The textual and visual measures are then combined to rate the quality of the generated reports. We present results that compare this evaluation metric with other textual metrics on a gold standard dataset derived from the MIMIC collection and show its robustness and sensitivity to factual errors.
Authors: Wenbo Zhang, Junyu Chen, Christopher Kanan
Abstract: Due to their large sizes, volumetric scans and whole-slide pathology images (WSIs) are often processed by extracting embeddings from local regions and then an aggregator makes predictions from this set. However, current methods require post-hoc visualization techniques (e.g., Grad-CAM) and often fail to localize small yet clinically crucial details. To address these limitations, we introduce INSIGHT, a novel weakly-supervised aggregator that integrates heatmap generation as an inductive bias. Starting from pre-trained feature maps, INSIGHT employs a detection module with small convolutional kernels to capture fine details and a context module with a broader receptive field to suppress local false positives. The resulting internal heatmap highlights diagnostically relevant regions. On CT and WSI benchmarks, INSIGHT achieves state-of-the-art classification results and high weakly-labeled semantic segmentation performance. Project website and code are available at: https://zhangdylan83.github.io/ewsmia/
Authors: Kuei-Chun Kao
Abstract: Recently, CLIP has become an important model for aligning images and text in multi-modal contexts. However, researchers have identified limitations in the ability of CLIP's text and image encoders to extract detailed knowledge from pairs of captions and images. In response, this paper presents Knowledge-CLIP, an innovative approach designed to improve CLIP's performance by integrating a new knowledge distillation (KD) method based on Llama 2. Our approach focuses on three key objectives: Text Embedding Distillation, Concept Learning, and Contrastive Learning. First, Text Embedding Distillation involves training the Knowledge-CLIP text encoder to mirror the teacher model, Llama 2. Next, Concept Learning assigns a soft concept label to each caption-image pair by employing offline K-means clustering on text data from Llama 2, enabling Knowledge-CLIP to learn from these soft concept labels. Lastly, Contrastive Learning aligns the text and image embeddings. Our experimental findings show that the proposed model improves the performance of both text and image encoders.
Authors: Enshen Zhou, Qi Su, Cheng Chi, Zhizheng Zhang, Zhongyuan Wang, Tiejun Huang, Lu Sheng, He Wang
Abstract: Automatic detection and prevention of open-set failures are crucial in closed-loop robotic systems. Recent studies often struggle to simultaneously identify unexpected failures reactively after they occur and prevent foreseeable ones proactively. To this end, we propose Code-as-Monitor (CaM), a novel paradigm leveraging the vision-language model (VLM) for both open-set reactive and proactive failure detection. The core of our method is to formulate both tasks as a unified set of spatio-temporal constraint satisfaction problems and use VLM-generated code to evaluate them for real-time monitoring. To enhance the accuracy and efficiency of monitoring, we further introduce constraint elements that abstract constraint-related entities or their parts into compact geometric elements. This approach offers greater generality, simplifies tracking, and facilitates constraint-aware visual programming by leveraging these elements as visual prompts. Experiments show that CaM achieves a 28.7% higher success rate and reduces execution time by 31.8% under severe disturbances compared to baselines across three simulators and a real-world setting. Moreover, CaM can be integrated with open-loop control policies to form closed-loop systems, enabling long-horizon tasks in cluttered scenes with dynamic environments.
Authors: Yixin Gao, Xin Li, Xiaohan Pan, Runsen Feng, Zongyu Guo, Yiting Lu, Yulin Ren, Zhibo Chen
Abstract: We present UniMIC, a universal multi-modality image compression framework, intending to unify the rate-distortion-perception (RDP) optimization for multiple image codecs simultaneously through excavating cross-modality generative priors. Unlike most existing works that need to design and optimize image codecs from scratch, our UniMIC introduces the visual codec repository, which incorporates amounts of representative image codecs and directly uses them as the basic codecs for various practical applications. Moreover, we propose multi-grained textual coding, where variable-length content prompt and compression prompt are designed and encoded to assist the perceptual reconstruction through the multi-modality conditional generation. In particular, a universal perception compensator is proposed to improve the perception quality of decoded images from all basic codecs at the decoder side by reusing text-assisted diffusion priors from stable diffusion. With the cooperation of the above three strategies, our UniMIC achieves a significant improvement of RDP optimization for different compression codecs, e.g., traditional and learnable codecs, and different compression costs, e.g., ultra-low bitrates. The code will be available in https://github.com/Amygyx/UniMIC .