Authors: Thomas Bashford-Rogers, Luis Paulo Santos
Abstract: Rendering algorithms typically integrate light paths over path space. However, integrating over this one unified space is not necessarily the most efficient approach, and we show that partitioning path space and integrating each of these partitioned spaces with a separate estimator can have advantages. We propose an approach for partitioning path space based on analyzing paths from a standard Monte Carlo estimator and integrating these partitioned path spaces using a Markov Chain Monte Carlo (MCMC) estimator. This also means that integration happens within a sparser subset of path space, so we propose the use of guided proposal distributions in image space to improve efficiency. We show that our method improves image quality over other MCMC integration approaches at the same number of samples.
Authors: Xinger Li, Zhiqiang Zhong, Bo Huang, Yang Yang
Abstract: This paper is the first-place solution for ICASSP MEIJU@2025 Track I, which focuses on low-resource multimodal emotion and intention recognition. How to effectively utilize a large amount of unlabeled data, while ensuring the mutual promotion of different difficulty levels tasks in the interaction stage, these two points become the key to the competition. In this paper, pseudo-label labeling is carried out on the model trained with labeled data, and samples with high confidence and their labels are selected to alleviate the problem of low resources. At the same time, the characteristic of easy represented ability of intention recognition found in the experiment is used to make mutually promote with emotion recognition under different attention heads, and higher performance of intention recognition is achieved through fusion. Finally, under the refined processing data, we achieve the score of 0.5532 in the Test set, and win the championship of the track.
Authors: Jan Hubi\v{c}ka, Linda Kimrov\'a, Melichar Kone\v{c}n\'y
Abstract: Dufaycolor, an additive color photography process produced from 1935 to the late 1950s, represents one of the most advanced iterations of this technique. This paper presents ongoing research and development of an open-source Color-Screen tool designed to reconstruct the original colors of additive color photographs. We discuss the incorporation of historical measurements of dyes used in the production of the color-screen filter (r\'eseau) to achieve accurate color recovery.
Authors: Xin Ding, Shijie Cao, Ting Cao, Zhibo Chen
Abstract: Vision generative models have recently made significant advancements along two primary paradigms: diffusion-style and language-style, both of which have demonstrated excellent scaling laws. Quantization is crucial for efficiently deploying these models, as it reduces memory and computation costs. In this work, we systematically investigate the impact of quantization on these two paradigms. Surprisingly, despite achieving comparable performance in full precision, language-style models consistently outperform diffusion-style models across various quantization settings. This observation suggests that language-style models have superior bit-level scaling laws, offering a better tradeoff between model quality and total bits. To dissect this phenomenon, we conduct extensive experiments and find that the primary reason is the discrete representation space of language-style models, which is more tolerant of information loss during quantization. Furthermore, our analysis indicates that improving the bit-level scaling law of quantized vision generative models is challenging, with model distillation identified as a highly effective approach. Specifically, we propose TopKLD to optimize the transfer of distilled knowledge by balancing ``implicit knowledge'' and ``explicit knowledge'' during the distillation process. This approach elevates the bit-level scaling laws by one level across both integer and floating-point quantization settings.
Authors: Phillip Maire, Samson G. King, Jonathan Andrew Cheung, Stefanie Walker, Samuel Andrew Hires
Abstract: The rodent vibrissal system is pivotal in advancing neuroscience research, particularly for studies of cortical plasticity, learning, decision-making, sensory encoding, and sensorimotor integration. Despite the advantages, curating touch events is labor intensive and often requires >3 hours per million video frames, even after leveraging automated tools like the Janelia Whisker Tracker. We address this limitation by introducing Whisker Automatic Contact Classifier (WhACC), a python package designed to identify touch periods from high-speed videos of head-fixed behaving rodents with human-level performance. WhACC leverages ResNet50V2 for feature extraction, combined with LightGBM for Classification. Performance is assessed against three expert human curators on over one million frames. Pairwise touch classification agreement on 99.5% of video frames, equal to between-human agreement. Finally, we offer a custom retraining interface to allow model customization on a small subset of data, which was validated on four million frames across 16 single-unit electrophysiology recordings. Including this retraining step, we reduce human hours required to curate a 100 million frame dataset from ~333 hours to ~6 hours.
Authors: Wen-Dong Jiang, Chih-Yung Chang, Diptendu Sinha Roy
Abstract: Recently, violence detection systems developed using unified multimodal models have achieved significant success and attracted widespread attention. However, most of these systems face two critical challenges: the lack of interpretability as black-box models and limited functionality, offering only classification or retrieval capabilities. To address these challenges, this paper proposes a novel interpretable violence detection system, termed the Three-in-One (TIO) System. The TIO system integrates knowledge graphs (KG) and graph attention networks (GAT) to provide three core functionalities: detection, retrieval, and explanation. Specifically, the system processes each video frame along with text descriptions generated by a large language model (LLM) for videos containing potential violent behavior. It employs ImageBind to generate high-dimensional embeddings for constructing a knowledge graph, uses GAT for reasoning, and applies lightweight time series modules to extract video embedding features. The final step connects a classifier and retriever for multi-functional outputs. The interpretability of KG enables the system to verify the reasoning process behind each output. Additionally, the paper introduces several lightweight methods to reduce the resource consumption of the TIO system and enhance its efficiency. Extensive experiments conducted on the XD-Violence and UCF-Crime datasets validate the effectiveness of the proposed system. A case study further reveals an intriguing phenomenon: as the number of bystanders increases, the occurrence of violent behavior tends to decrease.
Authors: Yangyang Li, Zhengya Qia, Yuelin Lia, Haorui Yanga, Ronghua Shanga, Licheng Jiaoa
Abstract: Medical images are characterized by intricate and complex features, requiring interpretation by physicians with medical knowledge and experience. Classical neural networks can reduce the workload of physicians, but can only handle these complex features to a limited extent. Theoretically, quantum computing can explore a broader parameter space with fewer parameters, but it is currently limited by the constraints of quantum hardware.Considering these factors, we propose a distributed hybrid quantum convolutional neural network based on quantum circuit splitting. This model leverages the advantages of quantum computing to effectively capture the complex features of medical images, enabling efficient classification even in resource-constrained environments. Our model employs a quantum convolutional neural network (QCNN) to extract high-dimensional features from medical images, thereby enhancing the model's expressive capability.By integrating distributed techniques based on quantum circuit splitting, the 8-qubit QCNN can be reconstructed using only 5 qubits.Experimental results demonstrate that our model achieves strong performance across 3 datasets for both binary and multiclass classification tasks. Furthermore, compared to recent technologies, our model achieves superior performance with fewer parameters, and experimental results validate the effectiveness of our model.
Authors: Subin Erattakulangara, Karthika Kelat, Katie Burnham, Rachel Balbi, Sarah E. Gerard, David Meyer, Sajan Goud Lingala
Abstract: Accurate segmentation of the vocal tract from magnetic resonance imaging (MRI) data is essential for various voice and speech applications. Manual segmentation is time intensive and susceptible to errors. This study aimed to evaluate the efficacy of deep learning algorithms for automatic vocal tract segmentation from 3D MRI.
Authors: Maxwell Meyer, Jack Spruyt
Abstract: Current approaches to dichotomous image segmentation (DIS) treat image matting and object segmentation as fundamentally different tasks. As improvements in image segmentation become increasingly challenging to achieve, combining image matting and grayscale segmentation techniques offers promising new directions for architectural innovation. Inspired by the possibility of aligning these two model tasks, we propose a new architectural approach for DIS called Confidence-Guided Matting (CGM). We created the first CGM model called Background Erase Network (BEN). BEN is comprised of two components: BEN Base for initial segmentation and BEN Refiner for confidence refinement. Our approach achieves substantial improvements over current state-of-the-art methods on the DIS5K validation dataset, demonstrating that matting-based refinement can significantly enhance segmentation quality. This work opens new possibilities for cross-pollination between matting and segmentation techniques in computer vision.
Authors: Nirit Alkalay, Roy Orfaig, Ben-Zion Bobrovsky
Abstract: 4D panoptic LiDAR segmentation is essential for scene understanding in autonomous driving and robotics ,combining semantic and instance segmentation with temporal consistency.Current methods, like 4D-PLS and 4D-STOP, use a tracking-by-detection methodology, employing deep learning networks to perform semantic and instance segmentation on each frame. To maintain temporal consistency, large-size instances detected in the current frame are compared and associated with instances within a temporal window that includes the current and preceding frames. However, their reliance on short-term instance detection, lack of motion estimation, and exclusion of small-sized instances lead to frequent identity switches and reduced tracking performance. We address these issues with the NextStop1 tracker, which integrates Kalman filter-based motion estimation, data association, and lifespan management, along with a tracklet state concept to improve prioritization. Evaluated using the LiDAR Segmentation and Tracking Quality (LSTQ) metric on the SemanticKITTI validation set, NextStop demonstrated enhanced tracking performance, particularly for small-sized objects like people and bicyclists, with fewer ID switches, earlier tracking initiation, and improved reliability in complex environments. The source code is available at https://github.com/AIROTAU/NextStopTracker
Authors: Mills Staylor, Amirreza Dolatpour Fathkouhi, Md Khairul Islam, Kaleigh O'Hara, Ryan Ghiles Goudjil, Geoffrey Fox, Judy Fox
Abstract: Large-scale astronomical image data processing and prediction is essential for astronomers, providing crucial insights into celestial objects, the universe's history, and its evolution. While modern deep learning models offer high predictive accuracy, they often demand substantial computational resources, making them resource-intensive and limiting accessibility. We introduce the Cloud-based Astronomy Inference (CAI) framework to address these challenges. This scalable solution integrates pre-trained foundation models with serverless cloud infrastructure through a Function-as-a-Service (FaaS) Message Interface (FMI). CAI enables efficient and scalable inference on astronomical images without extensive hardware. Using a foundation model for redshift prediction as a case study, our extensive experiments cover user devices, HPC (High-Performance Computing) servers, and Cloud. CAI's significant scalability improvement on large data sizes provides an accessible and effective tool for the astronomy community. The code is accessible at https://github.com/UVA-MLSys/AI-for-Astronomy.
Authors: Yunlong Tang, Junjia Guo, Pinxin Liu, Zhiyuan Wang, Hang Hua, Jia-Xing Zhong, Yunzhong Xiao, Chao Huang, Luchuan Song, Susan Liang, Yizhi Song, Liu He, Jing Bi, Mingqian Feng, Xinyang Li, Zeliang Zhang, Chenliang Xu
Abstract: Traditional Celluloid (Cel) Animation production pipeline encompasses multiple essential steps, including storyboarding, layout design, keyframe animation, inbetweening, and colorization, which demand substantial manual effort, technical expertise, and significant time investment. These challenges have historically impeded the efficiency and scalability of Cel-Animation production. The rise of generative artificial intelligence (GenAI), encompassing large language models, multimodal models, and diffusion models, offers innovative solutions by automating tasks such as inbetween frame generation, colorization, and storyboard creation. This survey explores how GenAI integration is revolutionizing traditional animation workflows by lowering technical barriers, broadening accessibility for a wider range of creators through tools like AniDoc, ToonCrafter, and AniSora, and enabling artists to focus more on creative expression and artistic innovation. Despite its potential, issues such as maintaining visual consistency, ensuring stylistic coherence, and addressing ethical considerations continue to pose challenges. Furthermore, this paper discusses future directions and explores potential advancements in AI-assisted animation. For further exploration and resources, please visit our GitHub repository: https://github.com/yunlong10/Awesome-AI4Animation
Authors: Rech Leong Tian Poh, Sye Loong Keoh, Liying Li
Abstract: As complex AI systems further prove to be an integral part of our lives, a persistent and critical problem is the underlying black-box nature of such products and systems. In pursuit of productivity enhancements, one must not forget the need for various technology to boost the overall trustworthiness of such AI systems. One example, which is studied extensively in this work, is the domain of Explainable Artificial Intelligence (XAI). Research works in this scope are centred around the objective of making AI systems more transparent and interpretable, to further boost reliability and trust in using them. In this work, we discuss the various motivation for XAI and its approaches, the underlying challenges that XAI faces, and some open problems that we believe deserve further efforts to look into. We also provide a brief discussion of various XAI approaches for image processing, and finally discuss some future directions, to hopefully express and motivate the positive development of the XAI research space.
Authors: Farina Riaz, Fakhar Zaman, Hajime Suzuki, Sharif Abuadbba, David Nguyen
Abstract: Variational Autoencoders (VAEs) are essential tools in generative modeling and image reconstruction, with their performance heavily influenced by the encoder-decoder architecture. This study aims to improve the quality of reconstructed images by enhancing their resolution and preserving finer details, particularly when working with low-resolution inputs (16x16 pixels), where traditional VAEs often yield blurred or in-accurate results. To address this, we propose a hybrid model that combines quantum computing techniques in the VAE encoder with convolutional neural networks (CNNs) in the decoder. By upscaling the resolution from 16x16 to 32x32 during the encoding process, our approach evaluates how the model reconstructs images with enhanced resolution while maintaining key features and structures. This method tests the model's robustness in handling image reconstruction and its ability to preserve essential details despite training on lower-resolution data. We evaluate our proposed down sampling filter for Quantum VAE (Q-VAE) on the MNIST and USPS datasets and compare it with classical VAEs and a variant called Classical Direct Passing VAE (CDP-VAE), which uses windowing pooling filters in the encoding process. Performance is assessed using metrics such as the Fr\'echet Inception Distance (FID) and Mean Squared Error (MSE), which measure the fidelity of reconstructed images. Our results demonstrate that the Q-VAE consistently outperforms both the Classical VAE and CDP-VAE, achieving significantly lower FID and MSE scores. Additionally, CDP-VAE yields better performance than C-VAE. These findings highlight the potential of quantum-enhanced VAEs to improve image reconstruction quality by enhancing resolution and preserving essential features, offering a promising direction for future applications in computer vision and synthetic data generation.
Authors: Huaiguang Cai
Abstract: Class Activation Mapping (CAM) methods are widely used to visualize neural network decisions, yet their underlying mechanisms remain incompletely understood. To enhance the understanding of CAM methods and improve their explainability, we introduce the Content Reserved Game-theoretic (CRG) Explainer. This theoretical framework clarifies the theoretical foundations of GradCAM and HiResCAM by modeling the neural network prediction process as a cooperative game. Within this framework, we develop ShapleyCAM, a new method that leverages gradients and the Hessian matrix to provide more precise and theoretically grounded visual explanations. Due to the computational infeasibility of exact Shapley value calculation, ShapleyCAM employs a second-order Taylor expansion of the cooperative game's utility function to derive a closed-form expression. Additionally, we propose the Residual Softmax Target-Class (ReST) utility function to address the limitations of pre-softmax and post-softmax scores. Extensive experiments across 12 popular networks on the ImageNet validation set demonstrate the effectiveness of ShapleyCAM and its variants. Our findings not only advance CAM explainability but also bridge the gap between heuristic-driven CAM methods and compute-intensive Shapley value-based methods. The code is available at \url{https://github.com/caihuaiguang/pytorch-shapley-cam}.
Authors: Mohammad Amin Mirzaee, Hung-Jui Huang, Wenzhen Yuan
Abstract: Scanning large-scale surfaces is widely demanded in surface reconstruction applications and detecting defects in industries' quality control and maintenance stages. Traditional vision-based tactile sensors have shown promising performance in high-resolution shape reconstruction while suffering limitations such as small sensing areas or susceptibility to damage when slid across surfaces, making them unsuitable for continuous sensing on large surfaces. To address these shortcomings, we introduce a novel vision-based tactile sensor designed for continuous surface sensing applications. Our design uses an elastomeric belt and two wheels to continuously scan the target surface. The proposed sensor showed promising results in both shape reconstruction and surface fusion, indicating its applicability. The dot product of the estimated and reference surface normal map is reported over the sensing area and for different scanning speeds. Results indicate that the proposed sensor can rapidly scan large-scale surfaces with high accuracy at speeds up to 45 mm/s.
Authors: Omer Aydin, Enis Karaarslan
Abstract: OpenAI released version GPT-4 on March 14, 2023, following the success of ChatGPT, which was announced in November 2022. In addition to the existing GPT-3 features, GPT-4 has the ability to interpret images. To achieve this, the processing power and model have been significantly improved. The ability to process and interpret images goes far beyond the applications and effectiveness of artificial intelligence. In this study, we will first explore the interpretation of radiological images in healthcare using artificial intelligence (AI). Then, we will experiment with the image interpretation capability of the GPT-4. In this way, we will address the question of whether artificial intelligence (AI) can replace a healthcare professional (e.g., a medical doctor) or whether it can be used as a decision support tool that makes decisions easier and more reliable.
Authors: Mahsa Geshvadi
Abstract: During tumor resection surgery, surgeons rely on neuronavigation to locate tumors and other critical structures in the brain. Most neuronavigation is based on preoperative images, such as MRI and ultrasound, to navigate through the brain. Neuronavigation acts like GPS for the brain, guiding neurosurgeons during the procedure. However, brain shift, a dynamic deformation caused by factors such as osmotic concentration, fluid levels, and tissue resection, can invalidate the preoperative images and introduce registration uncertainty. Considering and effectively visualizing this uncertainty has the potential to help surgeons trust the navigation again. Uncertainty has been studied in various domains since the 19th century. Considering uncertainty requires two essential components: 1) quantifying uncertainty; and 2) conveying the quantified values to the observer. There has been growing interest in both of these research areas during the past few decades.
Authors: Juan E. Tapia, L\'azaro Janier Gonz\'alez-Soler, Christoph Busch
Abstract: Foundation models are becoming increasingly popular due to their strong generalization capabilities resulting from being trained on huge datasets. These generalization capabilities are attractive in areas such as NIR Iris Presentation Attack Detection (PAD), in which databases are limited in the number of subjects and diversity of attack instruments, and there is no correspondence between the bona fide and attack images because, most of the time, they do not belong to the same subjects. This work explores an iris PAD approach based on two foundation models, DinoV2 and VisualOpenClip. The results show that fine-tuning prediction with a small neural network as head overpasses the state-of-the-art performance based on deep learning approaches. However, systems trained from scratch have still reached better results if bona fide and attack images are available.
Authors: Mohammad Asim, Christopher Wewer, Thomas Wimmer, Bernt Schiele, Jan Eric Lenssen
Abstract: We introduce MEt3R, a metric for multi-view consistency in generated images. Large-scale generative models for multi-view image generation are rapidly advancing the field of 3D inference from sparse observations. However, due to the nature of generative modeling, traditional reconstruction metrics are not suitable to measure the quality of generated outputs and metrics that are independent of the sampling procedure are desperately needed. In this work, we specifically address the aspect of consistency between generated multi-view images, which can be evaluated independently of the specific scene. Our approach uses DUSt3R to obtain dense 3D reconstructions from image pairs in a feed-forward manner, which are used to warp image contents from one view into the other. Then, feature maps of these images are compared to obtain a similarity score that is invariant to view-dependent effects. Using MEt3R, we evaluate the consistency of a large set of previous methods for novel view and video generation, including our open, multi-view latent diffusion model.
Authors: Navin Ranjan, Andreas Savakis
Abstract: In this paper, we propose Mix-QViT, an explainability-driven MPQ framework that systematically allocates bit-widths to each layer based on two criteria: layer importance, assessed via Layer-wise Relevance Propagation (LRP), which identifies how much each layer contributes to the final classification, and quantization sensitivity, determined by evaluating the performance impact of quantizing each layer at various precision levels while keeping others layers at a baseline. Additionally, for post-training quantization (PTQ), we introduce a clipped channel-wise quantization method designed to reduce the effects of extreme outliers in post-LayerNorm activations by removing severe inter-channel variations. We validate our approach by applying Mix-QViT to ViT, DeiT, and Swin Transformer models across multiple datasets. Our experimental results for PTQ demonstrate that both fixed-bit and mixed-bit methods outperform existing techniques, particularly at 3-bit, 4-bit, and 6-bit precision. Furthermore, in quantization-aware training, Mix-QViT achieves superior performance with 2-bit mixed-precision.
Authors: Matyas Bohacek, Hany Farid
Abstract: From a simple text prompt, generative-AI image models can create stunningly realistic and creative images bounded, it seems, by only our imagination. These models have achieved this remarkable feat thanks, in part, to the ingestion of billions of images collected from nearly every corner of the internet. Many creators have understandably expressed concern over how their intellectual property has been ingested without their permission or a mechanism to opt out of training. As a result, questions of fair use and copyright infringement have quickly emerged. We describe a method that allows us to determine if a model was trained on a specific image or set of images. This method is computationally efficient and assumes no explicit knowledge of the model architecture or weights (so-called black-box membership inference). We anticipate that this method will be crucial for auditing existing models and, looking ahead, ensuring the fairer development and deployment of generative AI models.
Authors: Youbing Hu, Yun Cheng, Olga Saukh, Firat Ozdemir, Anqi Lu, Zhiqiang Cao, Zhijun Li
Abstract: Dataset distillation has emerged as a strategy to compress real-world datasets for efficient training. However, it struggles with large-scale and high-resolution datasets, limiting its practicality. This paper introduces a novel resolution-independent dataset distillation method Focus ed Dataset Distillation (FocusDD), which achieves diversity and realism in distilled data by identifying key information patches, thereby ensuring the generalization capability of the distilled dataset across different network architectures. Specifically, FocusDD leverages a pre-trained Vision Transformer (ViT) to extract key image patches, which are then synthesized into a single distilled image. These distilled images, which capture multiple targets, are suitable not only for classification tasks but also for dense tasks such as object detection. To further improve the generalization of the distilled dataset, each synthesized image is augmented with a downsampled view of the original image. Experimental results on the ImageNet-1K dataset demonstrate that, with 100 images per class (IPC), ResNet50 and MobileNet-v2 achieve validation accuracies of 71.0% and 62.6%, respectively, outperforming state-of-the-art methods by 2.8% and 4.7%. Notably, FocusDD is the first method to use distilled datasets for object detection tasks. On the COCO2017 dataset, with an IPC of 50, YOLOv11n and YOLOv11s achieve 24.4% and 32.1% mAP, respectively, further validating the effectiveness of our approach.
Authors: Shan Zhang, Aotian Chen, Yanpeng Sun, Jindong Gu, Yi-Yu Zheng, Piotr Koniusz, Kai Zou, Anton van den Hengel, Yuan Xue
Abstract: Current multimodal large language models (MLLMs) often underperform on mathematical problem-solving tasks that require fine-grained visual understanding. The limitation is largely attributable to inadequate perception of geometric primitives during image-level contrastive pre-training (e.g., CLIP). While recent efforts to improve math MLLMs have focused on scaling up mathematical visual instruction datasets and employing stronger LLM backbones, they often overlook persistent errors in visual recognition. In this paper, we systematically evaluate the visual grounding capabilities of state-of-the-art MLLMs and reveal a significant negative correlation between visual grounding accuracy and problem-solving performance, underscoring the critical role of fine-grained visual understanding. Notably, advanced models like GPT-4o exhibit a 70% error rate when identifying geometric entities, highlighting that this remains a key bottleneck in visual mathematical reasoning. To address this, we propose a novel approach, SVE-Math (Selective Vision-Enhanced Mathematical MLLM), featuring a geometric-grounded vision encoder and a feature router that dynamically adjusts the contribution of hierarchical visual feature maps. Our model recognizes accurate visual primitives and generates precise visual prompts tailored to the language model's reasoning needs. In experiments, SVE-Math-Qwen2.5-7B outperforms other 7B models by 15% on MathVerse and is compatible with GPT-4V on MathVista. Despite being trained on smaller datasets, SVE-Math-7B achieves competitive performance on GeoQA, rivaling models trained on significantly larger datasets. Our findings emphasize the importance of incorporating fine-grained visual understanding into MLLMs and provide a promising direction for future research.
Authors: Aditya Rauniyar, Omar Alama, Silong Yong, Katia Sycara, Sebastian Scherer
Abstract: Recent photorealistic Novel View Synthesis (NVS) advances have increasingly gained attention. However, these approaches remain constrained to small indoor scenes. While optimization-based NVS models have attempted to address this, generalizable feed-forward methods, offering significant advantages, remain underexplored. In this work, we train PixelNeRF, a feed-forward NVS model, on the large-scale UrbanScene3D dataset. We propose four training strategies to cluster and train on this dataset, highlighting that performance is hindered by limited view overlap. To address this, we introduce Aug3D, an augmentation technique that leverages reconstructed scenes using traditional Structure-from-Motion (SfM). Aug3D generates well-conditioned novel views through grid and semantic sampling to enhance feed-forward NVS model learning. Our experiments reveal that reducing the number of views per cluster from 20 to 10 improves PSNR by 10%, but the performance remains suboptimal. Aug3D further addresses this by combining the newly generated novel views with the original dataset, demonstrating its effectiveness in improving the model's ability to predict novel views.
Authors: Maomao Li, Lijian Lin, Yunfei Liu, Ye Zhu, Yu Li
Abstract: This paper presents Qffusion, a dual-frame-guided framework for portrait video editing. Specifically, we consider a design principle of ``animation for editing'', and train Qffusion as a general animation framework from two still reference images while we can use it for portrait video editing easily by applying modified start and end frames as references during inference. Leveraging the powerful generative power of Stable Diffusion, we propose a Quadrant-grid Arrangement (QGA) scheme for latent re-arrangement, which arranges the latent codes of two reference images and that of four facial conditions into a four-grid fashion, separately. Then, we fuse features of these two modalities and use self-attention for both appearance and temporal learning, where representations at different times are jointly modeled under QGA. Our Qffusion can achieve stable video editing without additional networks or complex training stages, where only the input format of Stable Diffusion is modified. Further, we propose a Quadrant-grid Propagation (QGP) inference strategy, which enjoys a unique advantage on stable arbitrary-length video generation by processing reference and condition frames recursively. Through extensive experiments, Qffusion consistently outperforms state-of-the-art techniques on portrait video editing.
Authors: Yijie Li, Hewei Wang, Shaofan Wang, Yee Hui Lee, Muhammad Salman Pathan, Soumyabrata Dev
Abstract: Recent advancements in meteorology involve the use of ground-based sky cameras for cloud observation. Analyzing images from these cameras helps in calculating cloud coverage and understanding atmospheric phenomena. Traditionally, cloud image segmentation relied on conventional computer vision techniques. However, with the advent of deep learning, convolutional neural networks (CNNs) are increasingly applied for this purpose. Despite their effectiveness, CNNs often require many epochs to converge, posing challenges for real-time processing in sky camera systems. In this paper, we introduce a residual U-Net with deep supervision for cloud segmentation which provides better accuracy than previous approaches, and with less training consumption. By utilizing residual connection in encoders of UCloudNet, the feature extraction ability is further improved.
Authors: Yijie Li, Hewei Wang, Aggelos Katsaggelos
Abstract: Most of the current salient object detection approaches use deeper networks with large backbones to produce more accurate predictions, which results in a significant increase in computational complexity. A great number of network designs follow the pure UNet and Feature Pyramid Network (FPN) architecture which has limited feature extraction and aggregation ability which motivated us to design a lightweight post-decoder refinement module, the crossed post-decoder refinement (CPDR) to enhance the feature representation of a standard FPN or U-Net framework. Specifically, we introduce the Attention Down Sample Fusion (ADF), which employs channel attention mechanisms with attention maps generated by high-level representation to refine the low-level features, and Attention Up Sample Fusion (AUF), leveraging the low-level information to guide the high-level features through spatial attention. Additionally, we proposed the Dual Attention Cross Fusion (DACF) upon ADFs and AUFs, which reduces the number of parameters while maintaining the performance. Experiments on five benchmark datasets demonstrate that our method outperforms previous state-of-the-art approaches.
Authors: Ziteng Cui, Lin Gu, Tatsuya Harada
Abstract: Curve & Lookup Table (LUT) based methods directly map a pixel to the target output, making them highly efficient tools for real-time photography processing. However, due to extreme memory complexity to learn full RGB space mapping, existing methods either sample a discretized 3D lattice to build a 3D LUT or decompose into three separate curves (1D LUTs) on the RGB channels. Here, we propose a novel algorithm, IAC, to learn an image-adaptive Cartesian coordinate system in the RGB color space before performing curve operations. This end-to-end trainable approach enables us to efficiently adjust images with a jointly learned image-adaptive coordinate system and curves. Experimental results demonstrate that this simple strategy achieves state-of-the-art (SOTA) performance in various photography processing tasks, including photo retouching, exposure correction, and white-balance editing, while also maintaining a lightweight design and fast inference speed.
Authors: Zhen Hong, Bowen Wang, Haoran Duan, Yawen Huang, Xiong Li, Zhenyu Wen, Xiang Wu, Wei Xiang, Yefeng Zheng
Abstract: Neural implicit representations have recently shown promising progress in dense Simultaneous Localization And Mapping (SLAM). However, existing works have shortcomings in terms of reconstruction quality and real-time performance, mainly due to inflexible scene representation strategy without leveraging any prior information. In this paper, we introduce SP-SLAM, a novel neural RGB-D SLAM system that performs tracking and mapping in real-time. SP-SLAM computes depth images and establishes sparse voxel-encoded scene priors near the surfaces to achieve rapid convergence of the model. Subsequently, the encoding voxels computed from single-frame depth image are fused into a global volume, which facilitates high-fidelity surface reconstruction. Simultaneously, we employ tri-planes to store scene appearance information, striking a balance between achieving high-quality geometric texture mapping and minimizing memory consumption. Furthermore, in SP-SLAM, we introduce an effective optimization strategy for mapping, allowing the system to continuously optimize the poses of all historical input frames during runtime without increasing computational overhead. We conduct extensive evaluations on five benchmark datasets (Replica, ScanNet, TUM RGB-D, Synthetic RGB-D, 7-Scenes). The results demonstrate that, compared to existing methods, we achieve superior tracking accuracy and reconstruction quality, while running at a significantly faster speed.
Authors: Yuan Lai, Zhiwei Shi, Chengxi Zhu
Abstract: The 3D trajectory of a shuttlecock required for a badminton rally robot for human-robot competition demands real-time performance with high accuracy. However, the fast flight speed of the shuttlecock, along with various visual effects, and its tendency to blend with environmental elements, such as court lines and lighting, present challenges for rapid and accurate 2D detection. In this paper, we first propose the YO-CSA detection network, which optimizes and reconfigures the YOLOv8s model's backbone, neck, and head by incorporating contextual and spatial attention mechanisms to enhance model's ability in extracting and integrating both global and local features. Next, we integrate three major subtasks, detection, prediction, and compensation, into a real-time 3D shuttlecock trajectory detection system. Specifically, our system maps the 2D coordinate sequence extracted by YO-CSA into 3D space using stereo vision, then predicts the future 3D coordinates based on historical information, and re-projects them onto the left and right views to update the position constraints for 2D detection. Additionally, our system includes a compensation module to fill in missing intermediate frames, ensuring a more complete trajectory. We conduct extensive experiments on our own dataset to evaluate both YO-CSA's performance and system effectiveness. Experimental results show that YO-CSA achieves a high accuracy of 90.43% mAP@0.75, surpassing both YOLOv8s and YOLO11s. Our system performs excellently, maintaining a speed of over 130 fps across 12 test sequences.
Authors: Mehrshad Saadatinia, Minoo Ahmadi, Armin Abdollahi
Abstract: Understanding emotions in videos is a challenging task. However, videos contain several modalities which make them a rich source of data for machine learning and deep learning tasks. In this work, we aim to improve video sentiment classification by focusing on two key aspects: the video itself, the accompanying text, and the acoustic features. To address the limitations of relying on large labeled datasets, we are developing a method that utilizes clustering-based semi-supervised pre-training to extract meaningful representations from the data. This pre-training step identifies patterns in the video and text data, allowing the model to learn underlying structures and relationships without requiring extensive labeled information at the outset. Once these patterns are established, we fine-tune the system in a supervised manner to classify the sentiment expressed in videos. We believe that this multi-modal approach, combining clustering with supervised fine-tuning, will lead to more accurate and insightful sentiment classification, especially in cases where labeled data is limited.
Authors: Zhendong Zhang
Abstract: To address the high resolution of image pixels, the Swin Transformer introduces window attention. This mechanism divides an image into non-overlapping windows and restricts attention computation to within each window, significantly enhancing computational efficiency. To further optimize this process, one might consider replacing standard attention with flash attention, which has proven to be more efficient in language models. However, a direct substitution is ineffective. Flash attention is designed for long sequences, whereas window attention deals with shorter sequences but must handle numerous of them in parallel. In this report, we present an optimized solution called Flash Window Attention, tailored specifically for window attention. Flash Window Attention improves attention computation efficiency by up to 300% and enhances end-to-end runtime efficiency by up to 30%. Our code is available online.
Authors: Xiaoying Xing, Avinab Saha, Junfeng He, Susan Hao, Paul Vicol, Moonkyung Ryu, Gang Li, Sahil Singla, Sarah Young, Yinxiao Li, Feng Yang, Deepak Ramachandran
Abstract: Text-to-image (T2I) generation has made significant advances in recent years, but challenges still remain in the generation of perceptual artifacts, misalignment with complex prompts, and safety. The prevailing approach to address these issues involves collecting human feedback on generated images, training reward models to estimate human feedback, and then fine-tuning T2I models based on the reward models to align them with human preferences. However, while existing reward fine-tuning methods can produce images with higher rewards, they may change model behavior in unexpected ways. For example, fine-tuning for one quality aspect (e.g., safety) may degrade other aspects (e.g., prompt alignment), or may lead to reward hacking (e.g., finding a way to increase rewards without having the intended effect). In this paper, we propose Focus-N-Fix, a region-aware fine-tuning method that trains models to correct only previously problematic image regions. The resulting fine-tuned model generates images with the same high-level structure as the original model but shows significant improvements in regions where the original model was deficient in safety (over-sexualization and violence), plausibility, or other criteria. Our experiments demonstrate that Focus-N-Fix improves these localized quality aspects with little or no degradation to others and typically imperceptible changes in the rest of the image. Disclaimer: This paper contains images that may be overly sexual, violent, offensive, or harmful.
Authors: Qiang Qu, Yiran Shen, Xiaoming Chen, Yuk Ying Chung, Weidong Cai, Tongliang Liu
Abstract: Neural View Synthesis (NVS), such as NeRF and 3D Gaussian Splatting, effectively creates photorealistic scenes from sparse viewpoints, typically evaluated by quality assessment methods like PSNR, SSIM, and LPIPS. However, these full-reference methods, which compare synthesized views to reference views, may not fully capture the perceptual quality of neurally synthesized scenes (NSS), particularly due to the limited availability of dense reference views. Furthermore, the challenges in acquiring human perceptual labels hinder the creation of extensive labeled datasets, risking model overfitting and reduced generalizability. To address these issues, we propose NVS-SQA, a NSS quality assessment method to learn no-reference quality representations through self-supervision without reliance on human labels. Traditional self-supervised learning predominantly relies on the "same instance, similar representation" assumption and extensive datasets. However, given that these conditions do not apply in NSS quality assessment, we employ heuristic cues and quality scores as learning objectives, along with a specialized contrastive pair preparation process to improve the effectiveness and efficiency of learning. The results show that NVS-SQA outperforms 17 no-reference methods by a large margin (i.e., on average 109.5% in SRCC, 98.6% in PLCC, and 91.5% in KRCC over the second best) and even exceeds 16 full-reference methods across all evaluation metrics (i.e., 22.9% in SRCC, 19.1% in PLCC, and 18.6% in KRCC over the second best).
Authors: Wulin Xie, Lian Zhao, Jiang Long, Xiaohuan Lu, Bingyan Nie
Abstract: Multi-view multi-label classification (MvMLC) has recently garnered significant research attention due to its wide range of real-world applications. However, incompleteness in views and labels is a common challenge, often resulting from data collection oversights and uncertainties in manual annotation. Furthermore, the task of learning robust multi-view representations that are both view-consistent and view-specific from diverse views still a challenge problem in MvMLC. To address these issues, we propose a novel framework for incomplete multi-view multi-label classification (iMvMLC). Our method factorizes multi-view representations into two independent sets of factors: view-consistent and view-specific, and we correspondingly design a graph disentangling loss to fully reduce redundancy between these representations. Additionally, our framework innovatively decomposes consistent representation learning into three key sub-objectives: (i) how to extract view-shared information across different views, (ii) how to eliminate intra-view redundancy in consistent representations, and (iii) how to preserve task-relevant information. To this end, we design a robust task-relevant consistency learning module that collaboratively learns high-quality consistent representations, leveraging a masked cross-view prediction (MCP) strategy and information theory. Notably, all modules in our framework are developed to function effectively under conditions of incomplete views and labels, making our method adaptable to various multi-view and multi-label datasets. Extensive experiments on five datasets demonstrate that our method outperforms other leading approaches.
Authors: Wenshu Fan, Minxing Zhang, Hongwei Li, Wenbo Jiang, Hanxiao Chen, Xiangyu Yue, Michael Backes, Xiao Zhang
Abstract: The widespread adoption of facial recognition (FR) models raises serious concerns about their potential misuse, motivating the development of anti-facial recognition (AFR) to protect user facial privacy. In this paper, we argue that the static FR strategy, predominantly adopted in prior literature for evaluating AFR efficacy, cannot faithfully characterize the actual capabilities of determined trackers who aim to track a specific target identity. In particular, we introduce \emph{\ourAttack}, a dynamic FR strategy where the model's gallery database is iteratively updated with newly recognized target identity images. Surprisingly, such a simple approach renders all the existing AFR protections ineffective. To mitigate the privacy threats posed by DynTracker, we advocate for explicitly promoting diversity in the AFR-protected images. We hypothesize that the lack of diversity is the primary cause of the failure of existing AFR methods. Specifically, we develop \emph{DivTrackee}, a novel method for crafting diverse AFR protections that builds upon a text-guided image generation framework and diversity-promoting adversarial losses. Through comprehensive experiments on various facial image benchmarks and feature extractors, we demonstrate DynTracker's strength in breaking existing AFR methods and the superiority of DivTrackee in preventing user facial images from being identified by dynamic FR strategies. We believe our work can act as an important initial step towards developing more effective AFR methods for protecting user facial privacy against determined trackers.
Authors: Chong Zhong, Yang Li, Jinfeng Xu, Xiang Fu, Yunhao Liu, Qiuyi Huang, Danjuan Yang, Meiyan Li, Aiyi Liu, Alan H. Welsh, Xingtao Zhou, Bo Fu, Catherine C. Liu
Abstract: We aim to assist image-based myopia screening by resolving two longstanding problems, "how to integrate the information of ocular images of a pair of eyes" and "how to incorporate the inherent dependence among high-myopia status and axial length for both eyes." The classification-regression task is modeled as a novel 4-dimensional muti-response regression, where discrete responses are allowed, that relates to two dependent 3rd-order tensors (3D ultrawide-field fundus images). We present a Vision Transformer-based bi-channel architecture, named CeViT, where the common features of a pair of eyes are extracted via a shared Transformer encoder, and the interocular asymmetries are modeled through separated multilayer perceptron heads. Statistically, we model the conditional dependence among mixture of discrete-continuous responses given the image covariates by a so-called copula loss. We establish a new theoretical framework regarding fine-tuning on CeViT based on latent representations, allowing the black-box fine-tuning procedure interpretable and guaranteeing higher relative efficiency of fine-tuning weight estimation in the asymptotic setting. We apply CeViT to an annotated ultrawide-field fundus image dataset collected by Shanghai Eye \& ENT Hospital, demonstrating that CeViT enhances the baseline model in both accuracy of classifying high-myopia and prediction of AL on both eyes.
Authors: Jiahui Tang, Kaihua Zhou, Zhijian Luo, Yueen Hou
Abstract: With the development of deep learning, numerous methods for low-light image enhancement (LLIE) have demonstrated remarkable performance. Mainstream LLIE methods typically learn an end-to-end mapping based on pairs of low-light and normal-light images. However, normal-light images under varying illumination conditions serve as reference images, making it difficult to define a ``perfect'' reference image This leads to the challenge of reconciling metric-oriented and visual-friendly results. Recently, many cross-modal studies have found that side information from other related modalities can guide visual representation learning. Based on this, we introduce a Natural Language Supervision (NLS) strategy, which learns feature maps from text corresponding to images, offering a general and flexible interface for describing an image under different illumination. However, image distributions conditioned on textual descriptions are highly multimodal, which makes training difficult. To address this issue, we design a Textual Guidance Conditioning Mechanism (TCM) that incorporates the connections between image regions and sentence words, enhancing the ability to capture fine-grained cross-modal cues for images and text. This strategy not only utilizes a wider range of supervised sources, but also provides a new paradigm for LLIE based on visual and textual feature alignment. In order to effectively identify and merge features from various levels of image and textual information, we design an Information Fusion Attention (IFA) module to enhance different regions at different levels. We integrate the proposed TCM and IFA into a Natural Language Supervision network for LLIE, named NaLSuper. Finally, extensive experiments demonstrate the robustness and superior effectiveness of our proposed NaLSuper.
Authors: Yiheng Li, Yang Yang, Zhen Lei
Abstract: Fusing multi-modality inputs from different sensors is an effective way to improve the performance of 3D object detection. However, current methods overlook two important conflicts: point-pixel misalignment and sub-task suppression. The former means a pixel feature from the opaque object is projected to multiple point features of the same ray in the world space, and the latter means the classification prediction and bounding box regression may cause mutual suppression. In this paper, we propose a novel method named Conflict Resolution Network (CoreNet) to address the aforementioned issues. Specifically, we first propose a dual-stream transformation module to tackle point-pixel misalignment. It consists of ray-based and point-based 2D-to-BEV transformations. Both of them achieve approximately unique mapping from the image space to the world space. Moreover, we introduce a task-specific predictor to tackle sub-task suppression. It uses the dual-branch structure which adopts class-specific query and Bbox-specific query to corresponding sub-tasks. Each task-specific query is constructed of task-specific feature and general feature, which allows the heads to adaptively select information of interest based on different sub-tasks. Experiments on the large-scale nuScenes dataset demonstrate the superiority of our proposed CoreNet, by achieving 75.6\% NDS and 73.3\% mAP on the nuScenes test set without test-time augmentation and model ensemble techniques. The ample ablation study also demonstrates the effectiveness of each component. The code is released on https://github.com/liyih/CoreNet.
Authors: Xianwei Zhuang, Zhihong Zhu, Yuxin Xie, Liming Liang, Yuexian Zou
Abstract: Large Vision-Language Models (LVLMs) may produce outputs that are unfaithful to reality, also known as visual hallucinations (VH), which significantly impedes their real-world usage. To alleviate VH, various decoding strategies have been proposed to enhance visual information. However, many of these methods may require secondary decoding and rollback, which significantly reduces inference speed. In this work, we propose an efficient plug-and-play decoding algorithm via Visual-Aware Sparsification (VASparse) from the perspective of token sparsity for mitigating VH. VASparse is inspired by empirical observations: (1) the sparse activation of attention in LVLMs, and (2) visual-agnostic tokens sparsification exacerbates VH. Based on these insights, we propose a novel token sparsification strategy that balances efficiency and trustworthiness. Specifically, VASparse implements a visual-aware token selection strategy during decoding to reduce redundant tokens while preserving visual context effectively. Additionally, we innovatively introduce a sparse-based visual contrastive decoding method to recalibrate the distribution of hallucinated outputs without the time overhead associated with secondary decoding. Subsequently, VASparse recalibrates attention scores to penalize attention sinking of LVLMs towards text tokens. Extensive experiments across four popular benchmarks confirm the effectiveness of VASparse in mitigating VH across different LVLM families without requiring additional training or post-processing. Impressively, VASparse achieves state-of-the-art performance for mitigating VH while maintaining competitive decoding speed. Code is available at https://github.com/mengchuang123/VASparse-github.
Authors: Narges Rashvand, Ghazal Alinezhad Noghre, Armin Danesh Pazho, Shanle Yao, Hamed Tabkhi
Abstract: Shoplifting poses a significant challenge for retailers, resulting in billions of dollars in annual losses. Traditional security measures often fall short, highlighting the need for intelligent solutions capable of detecting shoplifting behaviors in real time. This paper frames shoplifting detection as an anomaly detection problem, focusing on the identification of deviations from typical shopping patterns. We introduce PoseLift, a privacy-preserving dataset specifically designed for shoplifting detection, addressing challenges such as data scarcity, privacy concerns, and model biases. PoseLift is built in collaboration with a retail store and contains anonymized human pose data from real-world scenarios. By preserving essential behavioral information while anonymizing identities, PoseLift balances privacy and utility. We benchmark state-of-the-art pose-based anomaly detection models on this dataset, evaluating performance using a comprehensive set of metrics. Our results demonstrate that pose-based approaches achieve high detection accuracy while effectively addressing privacy and bias concerns inherent in traditional methods. As one of the first datasets capturing real-world shoplifting behaviors, PoseLift offers researchers a valuable tool to advance computer vision ethically and will be publicly available to foster innovation and collaboration. The dataset is available at https://github.com/TeCSAR-UNCC/PoseLift.
Authors: Antara Firoz Parsa, S. M. Abdullah, Anika Hasan Talukder, Md. Asif Shahidullah Kabbya, Shakib Al Hasan, Md. Farhadul Islam, Jannatun Noor
Abstract: The study involves a comprehensive performance analysis of popular classification and segmentation models, applied over a Bangladeshi pothole dataset, being developed by the authors of this research. This custom dataset of 824 samples, collected from the streets of Dhaka and Bogura performs competitively against the existing industrial and custom datasets utilized in the present literature. The dataset was further augmented four-fold for segmentation and ten-fold for classification evaluation. We tested nine classification models (CCT, CNN, INN, Swin Transformer, ConvMixer, VGG16, ResNet50, DenseNet201, and Xception) and four segmentation models (U-Net, ResU-Net, U-Net++, and Attention-Unet) over both the datasets. Among the classification models, lightweight models namely CCT, CNN, INN, Swin Transformer, and ConvMixer were emphasized due to their low computational requirements and faster prediction times. The lightweight models performed respectfully, oftentimes equating to the performance of heavyweight models. In addition, augmentation was found to enhance the performance of all the tested models. The experimental results exhibit that, our dataset performs on par or outperforms the similar classification models utilized in the existing literature, reaching accuracy and f1-scores over 99%. The dataset also performed on par with the existing datasets for segmentation, achieving model Dice Similarity Coefficient up to 67.54% and IoU scores up to 59.39%.
Authors: Crespo-Orti Luis, Moreno-Cuadrado Isabel, Olivares-Mart\'inez Pablo, Sanz-Tornero Ximo
Abstract: This paper addresses the challenge of parking space detection in urban areas, focusing on the city of Granada. Utilizing aerial imagery, we develop and apply semantic segmentation techniques to accurately identify parked cars, moving cars and roads. A significant aspect of our research is the creation of a proprietary dataset specific to Granada, which is instrumental in training our neural network model. We employ Fully Convolutional Networks, Pyramid Networks and Dilated Convolutions, demonstrating their effectiveness in urban semantic segmentation. Our approach involves comparative analysis and optimization of various models, including Dynamic U-Net, PSPNet and DeepLabV3+, tailored for the segmentation of aerial images. The study includes a thorough experimentation phase, using datasets such as UDD5 and UAVid, alongside our custom Granada dataset. We evaluate our models using metrics like Foreground Accuracy, Dice Coefficient and Jaccard Index. Our results indicate that DeepLabV3+ offers the most promising performance. We conclude with future directions, emphasizing the need for a dedicated neural network for parked car detection and the potential for application in other urban environments. This work contributes to the fields of urban planning and traffic management, providing insights into efficient utilization of parking spaces through advanced image processing techniques.
Authors: Hengyuan Zhang, David Paz, Yuliang Guo, Xinyu Huang, Henrik I. Christensen, Liu Ren
Abstract: Online mapping reduces the reliance of autonomous vehicles on high-definition (HD) maps, significantly enhancing scalability. However, recent advancements often overlook cross-sensor configuration generalization, leading to performance degradation when models are deployed on vehicles with different camera intrinsics and extrinsics. With the rapid evolution of novel view synthesis methods, we investigate the extent to which these techniques can be leveraged to address the sensor configuration generalization challenge. We propose a novel framework leveraging Gaussian splatting to reconstruct scenes and render camera images in target sensor configurations. The target config sensor data, along with labels mapped to the target config, are used to train online mapping models. Our proposed framework on the nuScenes and Argoverse 2 datasets demonstrates a performance improvement of 18% through effective dataset augmentation, achieves faster convergence and efficient training, and exceeds state-of-the-art performance when using only 25% of the original training data. This enables data reuse and reduces the need for laborious data labeling. Project page at https://henryzhangzhy.github.io/mapgs.
Authors: Erjian Guo, Zicheng Wang, Zhen Zhao, Luping Zhou
Abstract: Accurate medical image segmentation is often hindered by noisy labels in training data, due to the challenges of annotating medical images. Prior research works addressing noisy labels tend to make class-dependent assumptions, overlooking the pixel-dependent nature of most noisy labels. Furthermore, existing methods typically apply fixed thresholds to filter out noisy labels, risking the removal of minority classes and consequently degrading segmentation performance. To bridge these gaps, our proposed framework, Collaborative Learning with Curriculum Selection (CLCS), addresses pixel-dependent noisy labels with class imbalance. CLCS advances the existing works by i) treating noisy labels as pixel-dependent and addressing them through a collaborative learning framework, and ii) employing a curriculum dynamic thresholding approach adapting to model learning progress to select clean data samples to mitigate the class imbalance issue, and iii) applying a noise balance loss to noisy data samples to improve data utilization instead of discarding them outright. Specifically, our CLCS contains two modules: Curriculum Noisy Label Sample Selection (CNS) and Noise Balance Loss (NBL). In the CNS module, we designed a two-branch network with discrepancy loss for collaborative learning so that different feature representations of the same instance could be extracted from distinct views and used to vote the class probabilities of pixels. Besides, a curriculum dynamic threshold is adopted to select clean-label samples through probability voting. In the NBL module, instead of directly dropping the suspiciously noisy labels, we further adopt a robust loss to leverage such instances to boost the performance.
Authors: Haoxiang Gao, Yu Zhao
Abstract: Autonomous driving (AD) has experienced significant improvements in recent years and achieved promising 3D detection, classification, and localization results. However, many challenges remain, e.g. semantic understanding of pedestrians' behaviors, and downstream handling for pedestrian interactions. Recent studies in applications of Large Language Models (LLM) and Vision-Language Models (VLM) have achieved promising results in scene understanding and high-level maneuver planning in diverse traffic scenarios. However, deploying the billion-parameter LLMs to vehicles requires significant computation and memory resources. In this paper, we analyzed effective knowledge distillation of semantic labels to smaller Vision networks, which can be used for the semantic representation of complex scenes for downstream decision-making for planning and control.
Authors: Zhonghao Yan, Zijin Yin, Tianyu Lin, Xiangzhu Zeng, Kongming Liang, Zhanyu Ma
Abstract: The Segment Anything Model (SAM) has demonstrated strong and versatile segmentation capabilities, along with intuitive prompt-based interactions. However, customizing SAM for medical image segmentation requires massive amounts of pixel-level annotations and precise point- or box-based prompt designs. To address these challenges, we introduce PGP-SAM, a novel prototype-based few-shot tuning approach that uses limited samples to replace tedious manual prompts. Our key idea is to leverage inter- and intra-class prototypes to capture class-specific knowledge and relationships. We propose two main components: (1) a plug-and-play contextual modulation module that integrates multi-scale information, and (2) a class-guided cross-attention mechanism that fuses prototypes and features for automatic prompt generation. Experiments on a public multi-organ dataset and a private ventricle dataset demonstrate that PGP-SAM achieves superior mean Dice scores compared with existing prompt-free SAM variants, while using only 10\% of the 2D slices.
Authors: Ziyang Xie, Zhizheng Liu, Zhenghao Peng, Wayne Wu, Bolei Zhou
Abstract: Sim-to-real gap has long posed a significant challenge for robot learning in simulation, preventing the deployment of learned models in the real world. Previous work has primarily focused on domain randomization and system identification to mitigate this gap. However, these methods are often limited by the inherent constraints of the simulation and graphics engines. In this work, we propose Vid2Sim, a novel framework that effectively bridges the sim2real gap through a scalable and cost-efficient real2sim pipeline for neural 3D scene reconstruction and simulation. Given a monocular video as input, Vid2Sim can generate photorealistic and physically interactable 3D simulation environments to enable the reinforcement learning of visual navigation agents in complex urban environments. Extensive experiments demonstrate that Vid2Sim significantly improves the performance of urban navigation in the digital twins and real world by 31.2% and 68.3% in success rate compared with agents trained with prior simulation methods.
Authors: Peng Liu, Sen Lei, Heng-Chao Li
Abstract: Multicategory remote object counting is a fundamental task in computer vision, aimed at accurately estimating the number of objects of various categories in remote images. Existing methods rely on CNNs and Transformers, but CNNs struggle to capture global dependencies, and Transformers are computationally expensive, which limits their effectiveness in remote applications. Recently, Mamba has emerged as a promising solution in the field of computer vision, offering a linear complexity for modeling global dependencies. To this end, we propose Mamba-MOC, a mamba-based network designed for multi-category remote object counting, which represents the first application of Mamba to remote sensing object counting. Specifically, we propose a cross-scale interaction module to facilitate the deep integration of hierarchical features. Then we design a context state space model to capture both global and local contextual information and provide local neighborhood information during the scan process. Experimental results in large-scale realistic scenarios demonstrate that our proposed method achieves state-of-the-art performance compared with some mainstream counting algorithms.
Authors: Ming Dai, Jian Li, Jiedong Zhuang, Xian Zhang, Wankou Yang
Abstract: Multi-task visual grounding involves the simultaneous execution of localization and segmentation in images based on textual expressions. The majority of advanced methods predominantly focus on transformer-based multimodal fusion, aiming to extract robust multimodal representations. However, ambiguity between referring expression comprehension (REC) and referring image segmentation (RIS) is error-prone, leading to inconsistencies between multi-task predictions. Besides, insufficient multimodal understanding directly contributes to biased target perception. To overcome these challenges, we propose a Coarse-to-fine Consistency Constraints Visual Grounding architecture ($\text{C}^3\text{VG}$), which integrates implicit and explicit modeling approaches within a two-stage framework. Initially, query and pixel decoders are employed to generate preliminary detection and segmentation outputs, a process referred to as the Rough Semantic Perception (RSP) stage. These coarse predictions are subsequently refined through the proposed Mask-guided Interaction Module (MIM) and a novel explicit bidirectional consistency constraint loss to ensure consistent representations across tasks, which we term the Refined Consistency Interaction (RCI) stage. Furthermore, to address the challenge of insufficient multimodal understanding, we leverage pre-trained models based on visual-linguistic fusion representations. Empirical evaluations on the RefCOCO, RefCOCO+, and RefCOCOg datasets demonstrate the efficacy and soundness of $\text{C}^3\text{VG}$, which significantly outperforms state-of-the-art REC and RIS methods by a substantial margin. Code and model will be available at \url{https://github.com/Dmmm1997/C3VG}.
Authors: Yuxin Wang, Qianyi Wu, Dan Xu
Abstract: This paper tackles the problem of generalizable 3D-aware generation from monocular datasets, e.g., ImageNet. The key challenge of this task is learning a robust 3D-aware representation without multi-view or dynamic data, while ensuring consistent texture and geometry across different viewpoints. Although some baseline methods are capable of 3D-aware generation, the quality of the generated images still lags behind state-of-the-art 2D generation approaches, which excel in producing high-quality, detailed images. To address this severe limitation, we propose a novel feed-forward pipeline based on pixel-aligned Gaussian Splatting, coined as F3D-Gaus, which can produce more realistic and reliable 3D renderings from monocular inputs. In addition, we introduce a self-supervised cycle-consistent constraint to enforce cross-view consistency in the learned 3D representation. This training strategy naturally allows aggregation of multiple aligned Gaussian primitives and significantly alleviates the interpolation limitations inherent in single-view pixel-aligned Gaussian Splatting. Furthermore, we incorporate video model priors to perform geometry-aware refinement, enhancing the generation of fine details in wide-viewpoint scenarios and improving the model's capability to capture intricate 3D textures. Extensive experiments demonstrate that our approach not only achieves high-quality, multi-view consistent 3D-aware generation from monocular datasets, but also significantly improves training and inference efficiency.
Authors: Ashitha Mudraje, Brian B. Moser, Stanislav Frolov, Andreas Dengel
Abstract: Satellite imagery is a cornerstone for numerous Remote Sensing (RS) applications; however, limited spatial resolution frequently hinders the precision of such systems, especially in multi-label scene classification tasks as it requires a higher level of detail and feature differentiation. In this study, we explore the efficacy of image Super-Resolution (SR) as a pre-processing step to enhance the quality of satellite images and thus improve downstream classification performance. We investigate four SR models - SRResNet, HAT, SeeSR, and RealESRGAN - and evaluate their impact on multi-label scene classification across various CNN architectures, including ResNet-50, ResNet-101, ResNet-152, and Inception-v4. Our results show that applying SR significantly improves downstream classification performance across various metrics, demonstrating its ability to preserve spatial details critical for multi-label tasks. Overall, this work offers valuable insights into the selection of SR techniques for multi-label prediction in remote sensing and presents an easy-to-integrate framework to improve existing RS systems.
Authors: Samia Mehnaz, Md. Touhidul Islam
Abstract: In nations such as Bangladesh, agriculture plays a vital role in providing livelihoods for a significant portion of the population. Identifying and classifying plant diseases early is critical to prevent their spread and minimize their impact on crop yield and quality. Various computer vision techniques can be used for such detection and classification. While CNNs have been dominant on such image classification tasks, vision transformers has become equally good in recent time also. In this paper we study the various computer vision techniques for Bangladeshi rice leaf disease detection. We use the Dhan-Shomadhan -- a Bangladeshi rice leaf disease dataset, to experiment with various CNN and ViT models. We also compared the performance of such deep neural network architecture with traditional machine learning architecture like Support Vector Machine(SVM). We leveraged transfer learning for better generalization with lower amount of training data. Among the models tested, ResNet50 exhibited the best performance over other CNN and transformer-based models making it the optimal choice for this task.
Authors: Junlong Ren, Gangjian Zhang, Haifeng Sun, Hao Wang
Abstract: Temporal sentence grounding in videos (TSGV) faces challenges due to public TSGV datasets containing significant temporal biases, which are attributed to the uneven temporal distributions of target moments. Existing methods generate augmented videos, where target moments are forced to have varying temporal locations. However, since the video lengths of the given datasets have small variations, only changing the temporal locations results in poor generalization ability in videos with varying lengths. In this paper, we propose a novel training framework complemented by diversified data augmentation and a domain discriminator. The data augmentation generates videos with various lengths and target moment locations to diversify temporal distributions. However, augmented videos inevitably exhibit distinct feature distributions which may introduce noise. To address this, we design a domain adaptation auxiliary task to diminish feature discrepancies between original and augmented videos. We also encourage the model to produce distinct predictions for videos with the same text queries but different moment locations to promote debiased training. Experiments on Charades-CD and ActivityNet-CD datasets demonstrate the effectiveness and generalization abilities of our method in multiple grounding structures, achieving state-of-the-art results.
Authors: Zhenyang Feng, Zihe Wang, Saul Ibaven Bueno, Tomasz Frelek, Advikaa Ramesh, Jingyan Bai, Lemeng Wang, Zanming Huang, Jianyang Gu, Jinsu Yoo, Tai-Yu Pan, Arpita Chowdhury, Michelle Ramirez, Elizabeth G. Campolongo, Matthew J. Thompson, Christopher G. Lawrence, Sydne Record, Neil Rosser, Anuj Karpatne, Daniel Rubenstein, Hilmar Lapp, Charles V. Stewart, Tanya Berger-Wolf, Yu Su, Wei-Lun Chao
Abstract: We study image segmentation in the biological domain, particularly trait and part segmentation from specimen images (e.g., butterfly wing stripes or beetle body parts). This is a crucial, fine-grained task that aids in understanding the biology of organisms. The conventional approach involves hand-labeling masks, often for hundreds of images per species, and training a segmentation model to generalize these labels to other images, which can be exceedingly laborious. We present a label-efficient method named Static Segmentation by Tracking (SST). SST is built upon the insight: while specimens of the same species have inherent variations, the traits and parts we aim to segment show up consistently. This motivates us to concatenate specimen images into a ``pseudo-video'' and reframe trait and part segmentation as a tracking problem. Concretely, SST generates masks for unlabeled images by propagating annotated or predicted masks from the ``pseudo-preceding'' images. Powered by Segment Anything Model 2 (SAM~2) initially developed for video segmentation, we show that SST can achieve high-quality trait and part segmentation with merely one labeled image per species -- a breakthrough for analyzing specimen images. We further develop a cycle-consistent loss to fine-tune the model, again using one labeled image. Additionally, we highlight the broader potential of SST, including one-shot instance segmentation on images taken in the wild and trait-based image retrieval.
Authors: Ji Soo Lee, Jongha Kim, Jeehye Na, Jinyoung Park, Hyunwoo J. Kim
Abstract: Despite the advancements of Video Large Language Models (VideoLLMs) in various tasks, they struggle with fine-grained temporal understanding, such as Dense Video Captioning (DVC). DVC is a complicated task of describing all events within a video while also temporally localizing them, which integrates multiple fine-grained tasks, including video segmentation, video captioning, and temporal video grounding. Previous VideoLLMs attempt to solve DVC in a single step, failing to utilize their reasoning capability. Moreover, previous training objectives for VideoLLMs do not fully reflect the evaluation metrics, therefore not providing supervision directly aligned to target tasks. To address such a problem, we propose a novel framework named VidChain comprised of Chain-of-Tasks (CoTasks) and Metric-based Direct Preference Optimization (M-DPO). CoTasks decompose a complex task into a sequence of sub-tasks, allowing VideoLLMs to leverage their reasoning capabilities more effectively. M-DPO aligns a VideoLLM with evaluation metrics, providing fine-grained supervision to each task that is well-aligned with metrics. Applied to two different VideoLLMs, VidChain consistently improves their fine-grained video understanding, thereby outperforming previous VideoLLMs on two different DVC benchmarks and also on the temporal video grounding task. Code is available at \url{https://github.com/mlvlab/VidChain}.
Authors: Seohyun Lee, Jintae Park, Sanghyeok Park
Abstract: Virtual Try-On (VTON) technology allows users to visualize how clothes would look on them without physically trying them on, gaining traction with the rise of digitalization and online shopping. Traditional VTON methods, often using Generative Adversarial Networks (GANs) and Diffusion models, face challenges in achieving high realism and handling dynamic poses. This paper introduces Outfitting Diffusion with Pose Guided Condition (ODPG), a novel approach that leverages a latent diffusion model with multiple conditioning inputs during the denoising process. By transforming garment, pose, and appearance images into latent features and integrating these features in a UNet-based denoising model, ODPG achieves non-explicit synthesis of garments on dynamically posed human images. Our experiments on the FashionTryOn and a subset of the DeepFashion dataset demonstrate that ODPG generates realistic VTON images with fine-grained texture details across various poses, utilizing an end-to-end architecture without the need for explicit garment warping processes. Future work will focus on generating VTON outputs in video format and on applying our attention mechanism, as detailed in the Method section, to other domains with limited data.
Authors: Peng Zheng, Linzhi Huang, Yizhou Yu, Yi Chang, Yilin Wang, Rui Ma
Abstract: Neural volume rendering techniques, such as NeRF, have revolutionized 3D-aware image synthesis by enabling the generation of images of a single scene or object from various camera poses. However, the high computational cost of NeRF presents challenges for synthesizing high-resolution (HR) images. Most existing methods address this issue by leveraging 2D super-resolution, which compromise 3D-consistency. Other methods propose radiance manifolds or two-stage generation to achieve 3D-consistent HR synthesis, yet they are limited to specific synthesis tasks, reducing their universality. To tackle these challenges, we propose SuperNeRF-GAN, a universal framework for 3D-consistent super-resolution. A key highlight of SuperNeRF-GAN is its seamless integration with NeRF-based 3D-aware image synthesis methods and it can simultaneously enhance the resolution of generated images while preserving 3D-consistency and reducing computational cost. Specifically, given a pre-trained generator capable of producing a NeRF representation such as tri-plane, we first perform volume rendering to obtain a low-resolution image with corresponding depth and normal map. Then, we employ a NeRF Super-Resolution module which learns a network to obtain a high-resolution NeRF. Next, we propose a novel Depth-Guided Rendering process which contains three simple yet effective steps, including the construction of a boundary-correct multi-depth map through depth aggregation, a normal-guided depth super-resolution and a depth-guided NeRF rendering. Experimental results demonstrate the superior efficiency, 3D-consistency, and quality of our approach. Additionally, ablation studies confirm the effectiveness of our proposed components.
Authors: Mahmoud Ahmed, Xiang Li, Arpit Prajapati, Mohamed Elhoseiny
Abstract: Understanding objects in 3D at the part level is essential for humans and robots to navigate and interact with the environment. Current datasets for part-level 3D object understanding encompass a limited range of categories. For instance, the ShapeNet-Part and PartNet datasets only include 16, and 24 object categories respectively. The 3DCoMPaT dataset, specifically designed for compositional understanding of parts and materials, contains only 42 object categories. To foster richer and fine-grained part-level 3D understanding, we introduce 3DCoMPaT200, a large-scale dataset tailored for compositional understanding of object parts and materials, with 200 object categories with $\approx$5 times larger object vocabulary compared to 3DCoMPaT and $\approx$ 4 times larger part categories. Concretely, 3DCoMPaT200 significantly expands upon 3DCoMPaT, featuring 1,031 fine-grained part categories and 293 distinct material classes for compositional application to 3D object parts. Additionally, to address the complexities of compositional 3D modeling, we propose a novel task of Compositional Part Shape Retrieval using ULIP to provide a strong 3D foundational model for 3D Compositional Understanding. This method evaluates the model shape retrieval performance given one, three, or six parts described in text format. These results show that the model's performance improves with an increasing number of style compositions, highlighting the critical role of the compositional dataset. Such results underscore the dataset's effectiveness in enhancing models' capability to understand complex 3D shapes from a compositional perspective. Code and Data can be found at http://github.com/3DCoMPaT200/3DCoMPaT200
Authors: Zihao Mei, Jianhao Li, Bolin Zhang, Chong Wang, Lijun Guo, Guoqi Li, Jiangbo Qian
Abstract: With the rapid growth of dynamic vision sensor (DVS) data, constructing a low-energy, efficient data retrieval system has become an urgent task. Hash learning is one of the most important retrieval technologies which can keep the distance between hash codes consistent with the distance between DVS data. As spiking neural networks (SNNs) can encode information through spikes, they demonstrate great potential in promoting energy efficiency. Based on the binary characteristics of SNNs, we first propose a novel supervised hashing method named Spikinghash with a hierarchical lightweight structure. Spiking WaveMixer (SWM) is deployed in shallow layers, utilizing a multilevel 3D discrete wavelet transform (3D-DWT) to decouple spatiotemporal features into various low-frequency and high frequency components, and then employing efficient spectral feature fusion. SWM can effectively capture the temporal dependencies and local spatial features. Spiking Self-Attention (SSA) is deployed in deeper layers to further extract global spatiotemporal information. We also design a hash layer utilizing binary characteristic of SNNs, which integrates information over multiple time steps to generate final hash codes. Furthermore, we propose a new dynamic soft similarity loss for SNNs, which utilizes membrane potentials to construct a learnable similarity matrix as soft labels to fully capture the similarity differences between classes and compensate information loss in SNNs, thereby improving retrieval performance. Experiments on multiple datasets demonstrate that Spikinghash can achieve state-of-the-art results with low energy consumption and fewer parameters.
Authors: Aafaf Ridouan, Amine Bohi, Youssef Mourchid
Abstract: Pain management and severity detection are crucial for effective treatment, yet traditional self-reporting methods are subjective and may be unsuitable for non-verbal individuals (people with limited speaking skills). To address this limitation, we explore automated pain detection using facial expressions. Our study leverages deep learning techniques to improve pain assessment by analyzing facial images from the Pain Emotion Faces Database (PEMF). We propose two novel approaches1: (1) a hybrid ConvNeXt model combined with Long Short-Term Memory (LSTM) blocks to analyze video frames and predict pain presence, and (2) a Spatio-Temporal Graph Convolution Network (STGCN) integrated with LSTM to process landmarks from facial images for pain detection. Our work represents the first use of the PEMF dataset for binary pain classification and demonstrates the effectiveness of these models through extensive experimentation. The results highlight the potential of combining spatial and temporal features for enhanced pain detection, offering a promising advancement in objective pain assessment methodologies.
Authors: Yongshuo Zhu, Lu Li, Keyan Chen, Chenyang Liu, Fugen Zhou, Zhenwei Shi
Abstract: Remote sensing image semantic change detection is a method used to analyze remote sensing images, aiming to identify areas of change as well as categorize these changes within images of the same location taken at different times. Traditional change detection methods often face challenges in generalizing across semantic categories in practical scenarios. To address this issue, we introduce a novel approach called Semantic-CD, specifically designed for semantic change detection in remote sensing images. This method incorporates the open vocabulary semantics from the vision-language foundation model, CLIP. By utilizing CLIP's extensive vocabulary knowledge, our model enhances its ability to generalize across categories and improves segmentation through fully decoupled multi-task learning, which includes both binary change detection and semantic change detection tasks. Semantic-CD consists of four main components: a bi-temporal CLIP visual encoder for extracting features from bi-temporal images, an open semantic prompter for creating semantic cost volume maps with open vocabulary, a binary change detection decoder for generating binary change detection masks, and a semantic change detection decoder for producing semantic labels. Experimental results on the SECOND dataset demonstrate that Semantic-CD achieves more accurate masks and reduces semantic classification errors, illustrating its effectiveness in applying semantic priors from vision-language foundation models to SCD tasks.
Authors: Keyan Chen, Jiafan Zhang, Chenyang Liu, Zhengxia Zou, Zhenwei Shi
Abstract: Referring remote sensing image segmentation is crucial for achieving fine-grained visual understanding through free-format textual input, enabling enhanced scene and object extraction in remote sensing applications. Current research primarily utilizes pre-trained language models to encode textual descriptions and align them with visual modalities, thereby facilitating the expression of relevant visual features. However, these approaches often struggle to establish robust alignments between fine-grained semantic concepts, leading to inconsistent representations across textual and visual information. To address these limitations, we introduce a referring remote sensing image segmentation foundational model, RSRefSeg. RSRefSeg leverages CLIP for visual and textual encoding, employing both global and local textual semantics as filters to generate referring-related visual activation features in the latent space. These activated features then serve as input prompts for SAM, which refines the segmentation masks through its robust visual generalization capabilities. Experimental results on the RRSIS-D dataset demonstrate that RSRefSeg outperforms existing methods, underscoring the effectiveness of foundational models in enhancing multimodal task comprehension. The code is available at \url{https://github.com/KyanChen/RSRefSeg}.
Authors: Minglong Xue, Shuaibin Fan, Shivakumara Palaiahnakote, Mingliang Zhou
Abstract: Image dehazing techniques aim to enhance contrast and restore details, which are essential for preserving visual information and improving image processing accuracy. Existing methods rely on a single manual prior, which cannot effectively reveal image details. To overcome this limitation, we propose an unpaired image dehazing network, called the Simple Image Dehaze Enhancer via Unpaired Rich Physical Prior (UR2P-Dehaze). First, to accurately estimate the illumination, reflectance, and color information of the hazy image, we design a shared prior estimator (SPE) that is iteratively trained to ensure the consistency of illumination and reflectance, generating clear, high-quality images. Additionally, a self-monitoring mechanism is introduced to eliminate undesirable features, providing reliable priors for image reconstruction. Next, we propose Dynamic Wavelet Separable Convolution (DWSC), which effectively integrates key features across both low and high frequencies, significantly enhancing the preservation of image details and ensuring global consistency. Finally, to effectively restore the color information of the image, we propose an Adaptive Color Corrector that addresses the problem of unclear colors. The PSNR, SSIM, LPIPS, FID and CIEDE2000 metrics on the benchmark dataset show that our method achieves state-of-the-art performance. It also contributes to the performance improvement of downstream tasks. The project code will be available at https://github.com/Fan-pixel/UR2P-Dehaze. \end{abstract}
Authors: Ruizhe Ou, Yuan Hu, Fan Zhang, Jiaxin Chen, Yu Liu
Abstract: Multi-modal large language models (MLLMs) have achieved remarkable success in image- and region-level remote sensing (RS) image understanding tasks, such as image captioning, visual question answering, and visual grounding. However, existing RS MLLMs lack the pixel-level dialogue capability, which involves responding to user instructions with segmentation masks for specific instances. In this paper, we propose GeoPix, a RS MLLM that extends image understanding capabilities to the pixel level. This is achieved by equipping the MLLM with a mask predictor, which transforms visual features from the vision encoder into masks conditioned on the LLM's segmentation token embeddings. To facilitate the segmentation of multi-scale objects in RS imagery, a class-wise learnable memory module is integrated into the mask predictor to capture and store class-wise geo-context at the instance level across the entire dataset. In addition, to address the absence of large-scale datasets for training pixel-level RS MLLMs, we construct the GeoPixInstruct dataset, comprising 65,463 images and 140,412 instances, with each instance annotated with text descriptions, bounding boxes, and masks. Furthermore, we develop a two-stage training strategy to balance the distinct requirements of text generation and masks prediction in multi-modal multi-task optimization. Extensive experiments verify the effectiveness and superiority of GeoPix in pixel-level segmentation tasks, while also maintaining competitive performance in image- and region-level benchmarks.
Authors: Syed Ali Tariq, Tehseen Zia, Mubeen Ghafoor
Abstract: Explainability of deep convolutional neural networks (DCNNs) is an important research topic that tries to uncover the reasons behind a DCNN model's decisions and improve their understanding and reliability in high-risk environments. In this regard, we propose a novel method for generating interpretable counterfactual and contrastive explanations for DCNN models. The proposed method is model intrusive that probes the internal workings of a DCNN instead of altering the input image to generate explanations. Given an input image, we provide contrastive explanations by identifying the most important filters in the DCNN representing features and concepts that separate the model's decision between classifying the image to the original inferred class or some other specified alter class. On the other hand, we provide counterfactual explanations by specifying the minimal changes necessary in such filters so that a contrastive output is obtained. Using these identified filters and concepts, our method can provide contrastive and counterfactual reasons behind a model's decisions and makes the model more transparent. One of the interesting applications of this method is misclassification analysis, where we compare the identified concepts from a particular input image and compare them with class-specific concepts to establish the validity of the model's decisions. The proposed method is compared with state-of-the-art and evaluated on the Caltech-UCSD Birds (CUB) 2011 dataset to show the usefulness of the explanations provided.
Authors: Wenqi Zhou, Kai Cao, Hao Zheng, Xinyi Zheng, Miao Liu, Per Ola Kristensson, Walterio Mayol-Cuevas, Fan Zhang, Weizhe Lin, Junxiao Shen
Abstract: Long-form egocentric video understanding provides rich contextual information and unique insights into long-term human behaviors, holding significant potential for applications in embodied intelligence, long-term activity analysis, and personalized assistive technologies. However, existing benchmark datasets primarily focus on single, short-duration videos or moderately long videos up to dozens of minutes, leaving a substantial gap in evaluating extensive, ultra-long egocentric video recordings. To address this, we introduce X-LeBench, a novel benchmark dataset specifically crafted for evaluating tasks on extremely long egocentric video recordings. Leveraging the advanced text processing capabilities of large language models (LLMs), X-LeBench develops a life-logging simulation pipeline that produces realistic, coherent daily plans aligned with real-world video data. This approach enables the flexible integration of synthetic daily plans with real-world footage from Ego4D-a massive-scale egocentric video dataset covers a wide range of daily life scenarios-resulting in 432 simulated video life logs that mirror realistic daily activities in contextually rich scenarios. The video life-log durations span from 23 minutes to 16.4 hours. The evaluation of several baseline systems and multimodal large language models (MLLMs) reveals their poor performance across the board, highlighting the inherent challenges of long-form egocentric video understanding and underscoring the need for more advanced models.
Authors: Javier Gamazo Tejero, Moritz Schmid, Pablo M\'arquez Neila, Martin S. Zinkernagel, Sebastian Wolf, Raphael Sznitman
Abstract: This paper addresses the domain adaptation challenge for semantic segmentation in medical imaging. Despite the impressive performance of recent foundational segmentation models like SAM on natural images, they struggle with medical domain images. Beyond this, recent approaches that perform end-to-end fine-tuning of models are simply not computationally tractable. To address this, we propose a novel SAM adapter approach that minimizes the number of trainable parameters while achieving comparable performances to full fine-tuning. The proposed SAM adapter is strategically placed in the mask decoder, offering excellent and broad generalization capabilities and improved segmentation across both fully supervised and test-time domain adaptation tasks. Extensive validation on four datasets showcases the adapter's efficacy, outperforming existing methods while training less than 1% of SAM's total parameters.
Authors: Bismillah Khan, Syed Ali Tariq, Tehseen Zia, Muhammad Ahsan, David Windridge
Abstract: Deep learning models in computer vision have made remarkable progress, but their lack of transparency and interpretability remains a challenge. The development of explainable AI can enhance the understanding and performance of these models. However, existing techniques often struggle to provide convincing explanations that non-experts easily understand, and they cannot accurately identify models' intrinsic decision-making processes. To address these challenges, we propose to develop a counterfactual explanation (CE) model that balances plausibility and faithfulness. This model generates easy-to-understand visual explanations by making minimum changes necessary in images without altering the pixel data. Instead, the proposed method identifies internal concepts and filters learned by models and leverages them to produce plausible counterfactual explanations. The provided explanations reflect the internal decision-making process of the model, thus ensuring faithfulness to the model.
Authors: Haojun Yu, Di Dai, Ziwei Zhao, Di He, Han Hu, Liwei Wang
Abstract: Scaling up the vocabulary of semantic segmentation models is extremely challenging because annotating large-scale mask labels is labour-intensive and time-consuming. Recently, language-guided segmentation models have been proposed to address this challenge. However, their performance drops significantly when applied to out-of-distribution categories. In this paper, we propose a new large vocabulary semantic segmentation framework, called LarvSeg. Different from previous works, LarvSeg leverages image classification data to scale the vocabulary of semantic segmentation models as large-vocabulary classification datasets usually contain balanced categories and are much easier to obtain. However, for classification tasks, the category is image-level, while for segmentation we need to predict the label at pixel level. To address this issue, we first propose a general baseline framework to incorporate image-level supervision into the training process of a pixel-level segmentation model, making the trained network perform semantic segmentation on newly introduced categories in the classification data. We then observe that a model trained on segmentation data can group pixel features of categories beyond the training vocabulary. Inspired by this finding, we design a category-wise attentive classifier to apply supervision to the precise regions of corresponding categories to improve the model performance. Extensive experiments demonstrate that LarvSeg significantly improves the large vocabulary semantic segmentation performance, especially in the categories without mask labels. For the first time, we provide a 21K-category semantic segmentation model with the help of ImageNet21K. The code is available at https://github.com/HaojunYu1998/large_voc_seg.
Authors: Mathieu Cocheteux, Julien Moreau, Franck Davoine
Abstract: Accurate sensor calibration is crucial for autonomous systems, yet its uncertainty quantification remains underexplored. We present the first approach to integrate uncertainty awareness into online extrinsic calibration, combining Monte Carlo Dropout with Conformal Prediction to generate prediction intervals with a guaranteed level of coverage. Our method proposes a framework to enhance existing calibration models with uncertainty quantification, compatible with various network architectures. Validated on KITTI (RGB Camera-LiDAR) and DSEC (Event Camera-LiDAR) datasets, we demonstrate effectiveness across different visual sensor types, measuring performance with adapted metrics to evaluate the efficiency and reliability of the intervals. By providing calibration parameters with quantifiable confidence measures, we offer insights into the reliability of calibration estimates, which can greatly improve the robustness of sensor fusion in dynamic environments and usefully serve the Computer Vision community.
Authors: Hanwen Zhong, Jiaxin Chen, Yutong Zhang, Di Huang, Yunhong Wang
Abstract: Multi-Task Learning (MTL) for Vision Transformer aims at enhancing the model capability by tackling multiple tasks simultaneously. Most recent works have predominantly focused on designing Mixture-of-Experts (MoE) structures and in tegrating Low-Rank Adaptation (LoRA) to efficiently perform multi-task learning. However, their rigid combination hampers both the optimization of MoE and the ef fectiveness of reparameterization of LoRA, leading to sub-optimal performance and low inference speed. In this work, we propose a novel approach dubbed Efficient Multi-Task Learning (EMTAL) by transforming a pre-trained Vision Transformer into an efficient multi-task learner during training, and reparameterizing the learned structure for efficient inference. Specifically, we firstly develop the MoEfied LoRA structure, which decomposes the pre-trained Transformer into a low-rank MoE structure and employ LoRA to fine-tune the parameters. Subsequently, we take into account the intrinsic asynchronous nature of multi-task learning and devise a learning Quality Retaining (QR) optimization mechanism, by leveraging the historical high-quality class logits to prevent a well-trained task from performance degradation. Finally, we design a router fading strategy to integrate the learned parameters into the original Transformer, archiving efficient inference. Extensive experiments on public benchmarks demonstrate the superiority of our method, compared to the state-of-the-art multi-task learning approaches.
Authors: Sadia Kamal, Tim Oates
Abstract: As deep learning models gain attraction in medical data, ensuring transparent and trustworthy decision-making is essential. In skin cancer diagnosis, while advancements in lesion detection and classification have improved accuracy, the black-box nature of these methods poses challenges in understanding their decision processes, leading to trust issues among physicians. This study leverages the CLIP (Contrastive Language-Image Pretraining) model, trained on different skin lesion datasets, to capture meaningful relationships between visual features and diagnostic criteria terms. To further enhance transparency, we propose a method called MedGrad E-CLIP, which builds on gradient-based E-CLIP by incorporating a weighted entropy mechanism designed for complex medical imaging like skin lesions. This approach highlights critical image regions linked to specific diagnostic descriptions. The developed integrated pipeline not only classifies skin lesions by matching corresponding descriptions but also adds an essential layer of explainability developed especially for medical data. By visually explaining how different features in an image relates to diagnostic criteria, this approach demonstrates the potential of advanced vision-language models in medical image analysis, ultimately improving transparency, robustness, and trust in AI-driven diagnostic systems.
Authors: Liyan Chen, Huangying Zhan, Kevin Chen, Xiangyu Xu, Qingan Yan, Changjiang Cai, Yi Xu
Abstract: We introduce ActiveGAMER, an active mapping system that utilizes 3D Gaussian Splatting (3DGS) to achieve high-quality, real-time scene mapping and exploration. Unlike traditional NeRF-based methods, which are computationally demanding and restrict active mapping performance, our approach leverages the efficient rendering capabilities of 3DGS, allowing effective and efficient exploration in complex environments. The core of our system is a rendering-based information gain module that dynamically identifies the most informative viewpoints for next-best-view planning, enhancing both geometric and photometric reconstruction accuracy. ActiveGAMER also integrates a carefully balanced framework, combining coarse-to-fine exploration, post-refinement, and a global-local keyframe selection strategy to maximize reconstruction completeness and fidelity. Our system autonomously explores and reconstructs environments with state-of-the-art geometric and photometric accuracy and completeness, significantly surpassing existing approaches in both aspects. Extensive evaluations on benchmark datasets such as Replica and MP3D highlight ActiveGAMER's effectiveness in active mapping tasks.
Authors: Wojciech Zielonka, Stephan J. Garbin, Alexandros Lattas, George Kopanas, Paulo Gotardo, Thabo Beeler, Justus Thies, Timo Bolkart
Abstract: We present SynShot, a novel method for the few-shot inversion of a drivable head avatar based on a synthetic prior. We tackle two major challenges. First, training a controllable 3D generative network requires a large number of diverse sequences, for which pairs of images and high-quality tracked meshes are not always available. Second, state-of-the-art monocular avatar models struggle to generalize to new views and expressions, lacking a strong prior and often overfitting to a specific viewpoint distribution. Inspired by machine learning models trained solely on synthetic data, we propose a method that learns a prior model from a large dataset of synthetic heads with diverse identities, expressions, and viewpoints. With few input images, SynShot fine-tunes the pretrained synthetic prior to bridge the domain gap, modeling a photorealistic head avatar that generalizes to novel expressions and viewpoints. We model the head avatar using 3D Gaussian splatting and a convolutional encoder-decoder that outputs Gaussian parameters in UV texture space. To account for the different modeling complexities over parts of the head (e.g., skin vs hair), we embed the prior with explicit control for upsampling the number of per-part primitives. Compared to state-of-the-art monocular methods that require thousands of real training images, SynShot significantly improves novel view and expression synthesis.
Authors: Aisha Zulfiqar, Ebroul Izquiedro
Abstract: Plant species exhibit significant intra-class variation and minimal inter-class variation. To enhance classification accuracy, it is essential to reduce intra-class variation while maximizing inter-class variation. This paper addresses plant species classification using a limited number of labelled samples and introduces a novel Local Foreground Selection(LFS) attention mechanism. LFS is a straightforward module designed to generate discriminative support and query feature maps. It operates by integrating two types of attention: local attention, which captures local spatial details to enhance feature discrimination and increase inter-class differentiation, and foreground selection attention, which emphasizes the foreground plant object while mitigating background interference. By focusing on the foreground, the query and support features selectively highlight relevant feature sequences and disregard less significant background sequences, thereby reducing intra-class differences. Experimental results from three plant species datasets demonstrate the effectiveness of the proposed LFS attention mechanism and its complementary advantages over previous feature reconstruction methods.
Authors: Woubishet Zewdu Taffese, Ritesh Sharma, Mohammad Hossein Afsharmovahed, Gunasekaran Manogaran, Genda Chen
Abstract: Ensuring the structural integrity and safety of bridges is crucial for the reliability of transportation networks and public safety. Traditional crack detection methods are increasingly being supplemented or replaced by advanced artificial intelligence (AI) techniques. However, most of the models rely on two-stage target detection algorithms, which pose concerns for real-time applications due to their lower speed. While models such as YOLO (You Only Look Once) have emerged as transformative tools due to their remarkable speed and accuracy. However, the potential of the latest YOLOv8 framework in this domain remains underexplored. This study bridges that gap by rigorously evaluating YOLOv8's performance across five model scales (nano, small, medium, large, and extra-large) using a high-quality Roboflow dataset. A comprehensive hyperparameter optimization was performed, testing six state-of-the-art optimizers-Stochastic Gradient Descent, Adaptive Moment Estimation, Adam with Decoupled Weight Decay, Root Mean Square Propagation, Rectified Adam, and Nesterov-accelerated Adam. Results revealed that YOLOv8, optimized with Stochastic Gradient Descent, delivered exceptional accuracy and speed, setting a new benchmark for real-time crack detection. Beyond its immediate application, this research positions YOLOv8 as a foundational approach for integrating advanced computer vision techniques into infrastructure monitoring. By enabling more reliable and proactive maintenance of aging bridge networks, this work paves the way for safer, more efficient transportation systems worldwide.
Authors: Xinyi Zheng, Steve Zhang, Weizhe Lin, Aaron Zhang, Walterio W. Mayol-Cuevas, Junxiao Shen
Abstract: In this paper, we present a large-scale fine-grained dataset using high-resolution images captured from locations worldwide. Compared to existing datasets, our dataset offers a significantly larger size and includes a higher level of detail, making it uniquely suited for fine-grained 3D applications. Notably, our dataset is built using drone-captured aerial imagery, which provides a more accurate perspective for capturing real-world site layouts and architectural structures. By reconstructing environments with these detailed images, our dataset supports applications such as the COLMAP format for Gaussian Splatting and the Structure-from-Motion (SfM) method. It is compatible with widely-used techniques including SLAM, Multi-View Stereo, and Neural Radiance Fields (NeRF), enabling accurate 3D reconstructions and point clouds. This makes it a benchmark for reconstruction and segmentation tasks. The dataset enables seamless integration with multi-modal data, supporting a range of 3D applications, from architectural reconstruction to virtual tourism. Its flexibility promotes innovation, facilitating breakthroughs in 3D modeling and analysis.
Authors: Yuli Wang, Kritika Iyer, Sep Farhand, Yoshihisa Shinagawa
Abstract: The automatic identification of Magnetic Resonance Imaging (MRI) sequences can streamline clinical workflows by reducing the time radiologists spend manually sorting and identifying sequences, thereby enabling faster diagnosis and treatment planning for patients. However, the lack of standardization in the parameters of MRI scans poses challenges for automated systems and complicates the generation and utilization of datasets for machine learning research. To address this issue, we propose a system for MRI sequence identification using an unsupervised contrastive deep learning framework. By training a convolutional neural network based on the ResNet-18 architecture, our system classifies nine common MRI sequence types as a 9-class classification problem. The network was trained using an in-house internal dataset and validated on several public datasets, including BraTS, ADNI, Fused Radiology-Pathology Prostate Dataset, the Breast Cancer Dataset (ACRIN), among others, encompassing diverse acquisition protocols and requiring only 2D slices for training. Our system achieves a classification accuracy of over 0.95 across the nine most common MRI sequence types.
Authors: Mozhgan Nasr Azadani, James Riddell, Sean Sedwards, Krzysztof Czarnecki
Abstract: Enhanced visual understanding serves as a cornerstone for multimodal large language models (MLLMs). Recent hybrid MLLMs incorporate a mixture of vision experts to address the limitations of using a single vision encoder and excessively long visual tokens. Despite the progress of these MLLMs, a research gap remains in effectively integrating diverse vision encoders. This work explores fusion strategies of visual tokens for hybrid MLLMs, leading to the design of LEO, a novel MLLM with a dual-branch vision encoder framework that incorporates a post-adaptation fusion strategy and adaptive tiling: for each segmented tile of the input images, LEO sequentially interleaves the visual tokens from its two vision encoders. Extensive evaluation across 13 vision-language benchmarks reveals that LEO outperforms state-of-the-art open-source MLLMs and hybrid MLLMs on the majority of tasks. Furthermore, we show that LEO can be adapted to the specialized domain of autonomous driving without altering the model architecture or training recipe, achieving competitive performance compared to existing baselines. The code and model will be publicly available.
Authors: Yue Hu, Rong Liu, Meida Chen, Andrew Feng, Peter Beerel
Abstract: Achieving high-fidelity 3D reconstruction from monocular video remains challenging due to the inherent limitations of traditional methods like Structure-from-Motion (SfM) and monocular SLAM in accurately capturing scene details. While differentiable rendering techniques such as Neural Radiance Fields (NeRF) address some of these challenges, their high computational costs make them unsuitable for real-time applications. Additionally, existing 3D Gaussian Splatting (3DGS) methods often focus on photometric consistency, neglecting geometric accuracy and failing to exploit SLAM's dynamic depth and pose updates for scene refinement. We propose a framework integrating dense SLAM with 3DGS for real-time, high-fidelity dense reconstruction. Our approach introduces SLAM-Informed Adaptive Densification, which dynamically updates and densifies the Gaussian model by leveraging dense point clouds from SLAM. Additionally, we incorporate Geometry-Guided Optimization, which combines edge-aware geometric constraints and photometric consistency to jointly optimize the appearance and geometry of the 3DGS scene representation, enabling detailed and accurate SLAM mapping reconstruction. Experiments on the Replica and TUM-RGBD datasets demonstrate the effectiveness of our approach, achieving state-of-the-art results among monocular systems. Specifically, our method achieves a PSNR of 36.864, SSIM of 0.985, and LPIPS of 0.040 on Replica, representing improvements of 10.7%, 6.4%, and 49.4%, respectively, over the previous SOTA. On TUM-RGBD, our method outperforms the closest baseline by 10.2%, 6.6%, and 34.7% in the same metrics. These results highlight the potential of our framework in bridging the gap between photometric and geometric dense 3D scene representations, paving the way for practical and efficient monocular dense reconstruction.
Authors: Xuhui Guo, Tanmoy Dam, Rohan Dhamdhere, Gourav Modanwal, Anant Madabhushi
Abstract: 3D medical image segmentation has progressed considerably due to Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), yet these methods struggle to balance long-range dependency acquisition with computational efficiency. To address this challenge, we propose UNETVL (U-Net Vision-LSTM), a novel architecture that leverages recent advancements in temporal information processing. UNETVL incorporates Vision-LSTM (ViL) for improved scalability and memory functions, alongside an efficient Chebyshev Kolmogorov-Arnold Networks (KAN) to handle complex and long-range dependency patterns more effectively. We validated our method on the ACDC and AMOS2022 (post challenge Task 2) benchmark datasets, showing a significant improvement in mean Dice score compared to recent state-of-the-art approaches, especially over its predecessor, UNETR, with increases of 7.3% on ACDC and 15.6% on AMOS, respectively. Extensive ablation studies were conducted to demonstrate the impact of each component in UNETVL, providing a comprehensive understanding of its architecture. Our code is available at https://github.com/tgrex6/UNETVL, facilitating further research and applications in this domain.
Authors: Subrata Kumer Paul, Abu Saleh Musa Miah, Rakhi Rani Paul, Md. Ekramul Hamid, Jungpil Shin, Md Abdur Rahim
Abstract: The Internet of Things (IoT) and mobile technology have significantly transformed healthcare by enabling real-time monitoring and diagnosis of patients. Recognizing medical-related human activities (MRHA) is pivotal for healthcare systems, particularly for identifying actions that are critical to patient well-being. However, challenges such as high computational demands, low accuracy, and limited adaptability persist in Human Motion Recognition (HMR). While some studies have integrated HMR with IoT for real-time healthcare applications, limited research has focused on recognizing MRHA as essential for effective patient monitoring. This study proposes a novel HMR method for MRHA detection, leveraging multi-stage deep learning techniques integrated with IoT. The approach employs EfficientNet to extract optimized spatial features from skeleton frame sequences using seven Mobile Inverted Bottleneck Convolutions (MBConv) blocks, followed by ConvLSTM to capture spatio-temporal patterns. A classification module with global average pooling, a fully connected layer, and a dropout layer generates the final predictions. The model is evaluated on the NTU RGB+D 120 and HMDB51 datasets, focusing on MRHA, such as sneezing, falling, walking, sitting, etc. It achieves 94.85% accuracy for cross-subject evaluations and 96.45% for cross-view evaluations on NTU RGB+D 120, along with 89.00% accuracy on HMDB51. Additionally, the system integrates IoT capabilities using a Raspberry Pi and GSM module, delivering real-time alerts via Twilios SMS service to caregivers and patients. This scalable and efficient solution bridges the gap between HMR and IoT, advancing patient monitoring, improving healthcare outcomes, and reducing costs.
Authors: Jinjing Zhu, Songze Li, Lin Wang
Abstract: Conventional knowledge distillation (KD) approaches are designed for the student model to predict similar output as the teacher model for each sample. Unfortunately, the relationship across samples with same class is often neglected. In this paper, we explore to redefine the knowledge in distillation, capturing the relationship between each sample and its corresponding in-context samples (a group of similar samples with the same or different classes), and perform KD from an in-context sample retrieval perspective. As KD is a type of learned label smoothing regularization (LSR), we first conduct a theoretical analysis showing that the teacher's knowledge from the in-context samples is a crucial contributor to regularize the student training with the corresponding samples. Buttressed by the analysis, we propose a novel in-context knowledge distillation (IC-KD) framework that shows its superiority across diverse KD paradigms (offline, online, and teacher-free KD). Firstly, we construct a feature memory bank from the teacher model and retrieve in-context samples for each corresponding sample through retrieval-based learning. We then introduce Positive In-Context Distillation (PICD) to reduce the discrepancy between a sample from the student and the aggregated in-context samples with the same class from the teacher in the logit space. Moreover, Negative In-Context Distillation (NICD) is introduced to separate a sample from the student and the in-context samples with different classes from the teacher in the logit space. Extensive experiments demonstrate that IC-KD is effective across various types of KD, and consistently achieves state-of-the-art performance on CIFAR-100 and ImageNet datasets.
Authors: Jialin Wu, Kaikai Pan, Yanjiao Chen, Jiangyi Deng, Shengyuan Pang, Wenyuan Xu
Abstract: Transformer models have excelled in natural language tasks, prompting the vision community to explore their implementation in computer vision problems. However, these models are still influenced by adversarial examples. In this paper, we investigate the attack capabilities of six common adversarial attacks on three pretrained ViT models to reveal the vulnerability of ViT models. To understand and analyse the bias in neural network decisions when the input is adversarial, we use two visualisation techniques that are attention rollout and grad attention rollout. To prevent ViT models from adversarial attack, we propose Protego, a detection framework that leverages the transformer intrinsic capabilities to detection adversarial examples of ViT models. Nonetheless, this is challenging due to a diversity of attack strategies that may be adopted by adversaries. Inspired by the attention mechanism, we know that the token of prediction contains all the information from the input sample. Additionally, the attention region for adversarial examples differs from that of normal examples. Given these points, we can train a detector that achieves superior performance than existing detection methods to identify adversarial examples. Our experiments have demonstrated the high effectiveness of our detection method. For these six adversarial attack methods, our detector's AUC scores all exceed 0.95. Protego may advance investigations in metaverse security.
Authors: Yee-Fan Tan, Jun Lin Liow, Pei-Sze Tan, Fuad Noman, Raphael C. -W. Phan, Hernando Ombao, Chee-Ming Ting
Abstract: Modern brain imaging technologies have enabled the detailed reconstruction of human brain connectomes, capturing structural connectivity (SC) from diffusion MRI and functional connectivity (FC) from functional MRI. Understanding the intricate relationships between SC and FC is vital for gaining deeper insights into the brain's functional and organizational mechanisms. However, obtaining both SC and FC modalities simultaneously remains challenging, hindering comprehensive analyses. Existing deep generative models typically focus on synthesizing a single modality or unidirectional translation between FC and SC, thereby missing the potential benefits of bi-directional translation, especially in scenarios where only one connectome is available. Therefore, we propose Structural-Functional Connectivity GAN (SFC-GAN), a novel framework for bidirectional translation between SC and FC. This approach leverages the CycleGAN architecture, incorporating convolutional layers to effectively capture the spatial structures of brain connectomes. To preserve the topological integrity of these connectomes, we employ a structure-preserving loss that guides the model in capturing both global and local connectome patterns while maintaining symmetry. Our framework demonstrates superior performance in translating between SC and FC, outperforming baseline models in similarity and graph property evaluations compared to ground truth data, each translated modality can be effectively utilized for downstream classification.
Authors: Minhui Xie, Hao Peng, Pu Li, Guangjie Zeng, Shuhai Wang, Jia Wu, Peng Li, Philip S. Yu
Abstract: Superpixel segmentation is a foundation for many higher-level computer vision tasks, such as image segmentation, object recognition, and scene understanding. Existing graph-based superpixel segmentation methods typically concentrate on the relationships between a given pixel and its directly adjacent pixels while overlooking the influence of non-adjacent pixels. These approaches do not fully leverage the global information in the graph, leading to suboptimal segmentation quality. To address this limitation, we present SIT-HSS, a hierarchical superpixel segmentation method based on structural information theory. Specifically, we first design a novel graph construction strategy that incrementally explores the pixel neighborhood to add edges based on 1-dimensional structural entropy (1D SE). This strategy maximizes the retention of graph information while avoiding an overly complex graph structure. Then, we design a new 2D SE-guided hierarchical graph partitioning method, which iteratively merges pixel clusters layer by layer to reduce the graph's 2D SE until a predefined segmentation scale is achieved. Experimental results on three benchmark datasets demonstrate that the SIT-HSS performs better than state-of-the-art unsupervised superpixel segmentation algorithms. The source code is available at \url{https://github.com/SELGroup/SIT-HSS}.
Authors: Zhen Xiong, Yuqi Li, Chuanguang Yang, Tiao Tan, Zhihong Zhu, Siyuan Li, Yue Ma
Abstract: The diffusion transformer (DiT) architecture has attracted significant attention in image generation, achieving better fidelity, performance, and diversity. However, most existing DiT - based image generation methods focus on global - aware synthesis, and regional prompt control has been less explored. In this paper, we propose a coarse - to - fine generation pipeline for regional prompt - following generation. Specifically, we first utilize the powerful large language model (LLM) to generate both high - level descriptions of the image (such as content, topic, and objects) and low - level descriptions (such as details and style). Then, we explore the influence of cross - attention layers at different depths. We find that deeper layers are always responsible for high - level content control, while shallow layers handle low - level content control. Various prompts are injected into the proposed regional cross - attention control for coarse - to - fine generation. By using the proposed pipeline, we enhance the controllability of DiT - based image generation. Extensive quantitative and qualitative results show that our pipeline can improve the performance of the generated images.
Authors: Shivangi Rai, Rini Smita Thakur, Kunal Jangid, Vinod K Kurmi
Abstract: Source-free domain adaptation (SFDA) utilizes a pre-trained source model with unlabeled target data. Self-supervised SFDA techniques generate pseudolabels from the pre-trained source model, but these pseudolabels often contain noise due to domain discrepancies between the source and target domains. Traditional self-supervised SFDA techniques rely on deterministic model predictions using the softmax function, leading to unreliable pseudolabels. In this work, we propose to introduce predictive uncertainty and softmax calibration for pseudolabel refinement using evidential deep learning. The Dirichlet prior is placed over the output of the target network to capture uncertainty using evidence with a single forward pass. Furthermore, softmax calibration solves the translation invariance problem to assist in learning with noisy labels. We incorporate a combination of evidential deep learning loss and information maximization loss with calibrated softmax in both prior and non-prior target knowledge SFDA settings. Extensive experimental analysis shows that our method outperforms other state-of-the-art methods on benchmark datasets.
Authors: Tongxu Zhang, Bei Wang
Abstract: In recent years, point cloud upsampling has been widely applied in fields such as 3D reconstruction. Our study investigates the factors influencing point cloud upsampling on both global and local levels through representation learning. Specifically, the paper inputs global and local information of the same point cloud model object into two encoders to extract these features, fuses them, and then feeds the combined features into an upsampling decoder. The goal is to address issues of sparsity and noise in point clouds by leveraging prior knowledge from both global and local inputs. And the proposed framework can be applied to any state-of-the-art point cloud upsampling neural network. Experiments were conducted on a series of autoencoder-based models utilizing deep learning, yielding interpretability for both global and local inputs, and it has been proven in the results that our proposed framework can further improve the upsampling effect in previous SOTA works. At the same time, the Saliency Map reflects the differences between global and local feature inputs, as well as the effectiveness of training with both inputs in parallel.
Authors: Jiebin Yan, Lei Wu, Yuming Fang, Xuelin Liu, Xue Xia, Weide Liu
Abstract: With the rapid development of multimedia processing and deep learning technologies, especially in the field of video understanding, video quality assessment (VQA) has achieved significant progress. Although researchers have moved from designing efficient video quality mapping models to various research directions, in-depth exploration of the effectiveness-efficiency trade-offs of spatio-temporal modeling in VQA models is still less sufficient. Considering the fact that videos have highly redundant information, this paper investigates this problem from the perspective of joint spatial and temporal sampling, aiming to seek the answer to how little information we should keep at least when feeding videos into the VQA models while with acceptable performance sacrifice. To this end, we drastically sample the video's information from both spatial and temporal dimensions, and the heavily squeezed video is then fed into a stable VQA model. Comprehensive experiments regarding joint spatial and temporal sampling are conducted on six public video quality databases, and the results demonstrate the acceptable performance of the VQA model when throwing away most of the video information. Furthermore, with the proposed joint spatial and temporal sampling strategy, we make an initial attempt to design an online VQA model, which is instantiated by as simple as possible a spatial feature extractor, a temporal feature fusion module, and a global quality regression module. Through quantitative and qualitative experiments, we verify the feasibility of online VQA model by simplifying itself and reducing input.
Authors: Tze Ho Elden Tse, Runyang Feng, Linfang Zheng, Jiho Park, Yixing Gao, Jihie Kim, Ales Leonardis, Hyung Jin Chang
Abstract: With the availability of egocentric 3D hand-object interaction datasets, there is increasing interest in developing unified models for hand-object pose estimation and action recognition. However, existing methods still struggle to recognise seen actions on unseen objects due to the limitations in representing object shape and movement using 3D bounding boxes. Additionally, the reliance on object templates at test time limits their generalisability to unseen objects. To address these challenges, we propose to leverage superquadrics as an alternative 3D object representation to bounding boxes and demonstrate their effectiveness on both template-free object reconstruction and action recognition tasks. Moreover, as we find that pure appearance-based methods can outperform the unified methods, the potential benefits from 3D geometric information remain unclear. Therefore, we study the compositionality of actions by considering a more challenging task where the training combinations of verbs and nouns do not overlap with the testing split. We extend H2O and FPHA datasets with compositional splits and design a novel collaborative learning framework that can explicitly reason about the geometric relations between hands and the manipulated object. Through extensive quantitative and qualitative evaluations, we demonstrate significant improvements over the state-of-the-arts in (compositional) action recognition.
Authors: ZhouRui Zhang, Jun Li, JiaYan Li, ZhiJian Wu, JianHua Xu
Abstract: Recent feature masking knowledge distillation methods make use of attention mechanisms to identify either important spatial regions or channel clues for discriminative feature reconstruction. However, most of existing strategies perform global attention-guided feature masking distillation without delving into fine-grained visual clues in feature maps. In particular, uncovering locality-aware clues across different scales are conducive to reconstructing region-aware features, thereby significantly benefiting distillation performance. In this study, we propose a fine-grained adaptive feature masking distillation framework for accurate object detection. Different from previous methods in which global masking is performed on single-scale feature maps, we explore the scale-aware feature masking by performing feature distillation across various scales, such that the object-aware locality is encoded for improved feature reconstruction. In addition, our fine-grained feature distillation strategy is combined with a masking logits distillation scheme in which logits difference between teacher and student networks is utilized to guide the distillation process. Thus, it can help the student model to better learn from the teacher counterpart with improved knowledge transfer. Extensive experiments for detection task demonstrate the superiority of our method. For example, when RetinaNet, RepPoints and Cascade Mask RCNN are used as teacher detectors, the student network achieves mAP scores of 41.5\%, 42.9\%, and 42.6\%, respectively, outperforming state-of-the-art methods such as DMKD and FreeKD.
Authors: Sen Peng, Weixing Xie, Zilong Wang, Xiaohu Guo, Zhonggui Chen, Baorong Yang, Xiao Dong
Abstract: We introduce RMAvatar, a novel human avatar representation with Gaussian splatting embedded on mesh to learn clothed avatar from a monocular video. We utilize the explicit mesh geometry to represent motion and shape of a virtual human and implicit appearance rendering with Gaussian Splatting. Our method consists of two main modules: Gaussian initialization module and Gaussian rectification module. We embed Gaussians into triangular faces and control their motion through the mesh, which ensures low-frequency motion and surface deformation of the avatar. Due to the limitations of LBS formula, the human skeleton is hard to control complex non-rigid transformations. We then design a pose-related Gaussian rectification module to learn fine-detailed non-rigid deformations, further improving the realism and expressiveness of the avatar. We conduct extensive experiments on public datasets, RMAvatar shows state-of-the-art performance on both rendering quality and quantitative evaluations. Please see our project page at https://rm-avatar.github.io.
Authors: Anupam Pandey, Deepjyoti Bodo, Arpan Phukan, Asif Ekbal
Abstract: Visual Question Answering (VQA) is an interdisciplinary field that bridges the gap between computer vision (CV) and natural language processing(NLP), enabling Artificial Intelligence(AI) systems to answer questions about images. Since its inception in 2015, VQA has rapidly evolved, driven by advances in deep learning, attention mechanisms, and transformer-based models. This survey traces the journey of VQA from its early days, through major breakthroughs, such as attention mechanisms, compositional reasoning, and the rise of vision-language pre-training methods. We highlight key models, datasets, and techniques that shaped the development of VQA systems, emphasizing the pivotal role of transformer architectures and multimodal pre-training in driving recent progress. Additionally, we explore specialized applications of VQA in domains like healthcare and discuss ongoing challenges, such as dataset bias, model interpretability, and the need for common-sense reasoning. Lastly, we discuss the emerging trends in large multimodal language models and the integration of external knowledge, offering insights into the future directions of VQA. This paper aims to provide a comprehensive overview of the evolution of VQA, highlighting both its current state and potential advancements.
Authors: Han Liu, Yinwei Wei, Fan Liu, Wenjie Wang, Liqiang Nie, Tat-Seng Chua
Abstract: Multimodal information (e.g., visual, acoustic, and textual) has been widely used to enhance representation learning for micro-video recommendation. For integrating multimodal information into a joint representation of micro-video, multimodal fusion plays a vital role in the existing micro-video recommendation approaches. However, the static multimodal fusion used in previous studies is insufficient to model the various relationships among multimodal information of different micro-videos. In this paper, we develop a novel meta-learning-based multimodal fusion framework called Meta Multimodal Fusion (MetaMMF), which dynamically assigns parameters to the multimodal fusion function for each micro-video during its representation learning. Specifically, MetaMMF regards the multimodal fusion of each micro-video as an independent task. Based on the meta information extracted from the multimodal features of the input task, MetaMMF parameterizes a neural network as the item-specific fusion function via a meta learner. We perform extensive experiments on three benchmark datasets, demonstrating the significant improvements over several state-of-the-art multimodal recommendation models, like MMGCN, LATTICE, and InvRL. Furthermore, we lighten our model by adopting canonical polyadic decomposition to improve the training efficiency, and validate its effectiveness through experimental results. Codes are available at https://github.com/hanliu95/MetaMMF.
Authors: Zhuohang Yu, Kai Wang, Juyong Zhang
Abstract: We present a novel approach for depth estimation from images captured by structured light systems. Unlike many previous methods that rely on image matching process, our approach uses a density voxel grid to represent scene geometry, which is trained via self-supervised differentiable volume rendering. Our method leverages color fields derived from projected patterns in structured light systems during the rendering process, enabling the isolated optimization of the geometry field. This contributes to faster convergence and high-quality output. Additionally, we incorporate normalized device coordinates (NDC), a distortion loss, and a novel surface-based color loss to enhance geometric fidelity. Experimental results demonstrate that our method outperforms existing matching-based techniques in geometric performance for few-shot scenarios, achieving approximately a 60% reduction in average estimated depth errors on synthetic scenes and about 30% on real-world captured scenes. Furthermore, our approach delivers fast training, with a speed roughly three times faster than previous matching-free methods that employ implicit representations.
Authors: Zhong Peng, Yishi Xu, Gerong Wang, Wenchao Chen, Bo Chen, Jing Zhang
Abstract: Compositional Zero-Shot Learning (CZSL) aims to enable models to recognize novel compositions of visual states and objects that were absent during training. Existing methods predominantly focus on learning semantic representations of seen compositions but often fail to disentangle the independent features of states and objects in images, thereby limiting their ability to generalize to unseen compositions. To address this challenge, we propose Duplex, a novel dual-prototype learning method that integrates semantic and visual prototypes through a carefully designed dual-branch architecture, enabling effective representation learning for compositional tasks. Duplex utilizes a Graph Neural Network (GNN) to adaptively update visual prototypes, capturing complex interactions between states and objects. Additionally, it leverages the strong visual-semantic alignment of pre-trained Vision-Language Models (VLMs) and employs a multi-path architecture combined with prompt engineering to align image and text representations, ensuring robust generalization. Extensive experiments on three benchmark datasets demonstrate that Duplex outperforms state-of-the-art methods in both closed-world and open-world settings.
Authors: Xiantong Zhao, Xiuping Liu, Shengjing Tian, Yinan Han
Abstract: 3D single object tracking (3DSOT) in LiDAR point clouds is a critical task for outdoor perception, enabling real-time perception of object location, orientation, and motion. Despite the impressive performance of current 3DSOT methods, evaluating them on clean datasets inadequately reflects their comprehensive performance, as the adverse weather conditions in real-world surroundings has not been considered. One of the main obstacles is the lack of adverse weather benchmarks for the evaluation of 3DSOT. To this end, this work proposes a challenging benchmark for LiDAR-based 3DSOT in adverse weather, which comprises two synthetic datasets (KITTI-A and nuScenes-A) and one real-world dataset (CADC-SOT) spanning three weather types: rain, fog, and snow. Based on this benchmark, five representative 3D trackers from different tracking frameworks conducted robustness evaluation, resulting in significant performance degradations. This prompts the question: What are the factors that cause current advanced methods to fail on such adverse weather samples? Consequently, we explore the impacts of adverse weather and answer the above question from three perspectives: 1) target distance; 2) template shape corruption; and 3) target shape corruption. Finally, based on domain randomization and contrastive learning, we designed a dual-branch tracking framework for adverse weather, named DRCT, achieving excellent performance in benchmarks.
Authors: Wassim Kabbani, Kiran Raja, Raghavendra Ramachandra, Christoph Busch
Abstract: Fair operational systems are crucial in gaining and maintaining society's trust in face recognition systems (FRS). FRS start with capturing an image and assessing its quality before using it further for enrollment or verification. Fair Face Image Quality Assessment (FIQA) schemes therefore become equally important in the context of fair FRS. This work examines the sclera as a quality assessment region for obtaining a fair FIQA. The sclera region is agnostic to demographic variations and skin colour for assessing the quality of a face image. We analyze three skin tone related ISO/IEC face image quality assessment measures and assess the sclera region as an alternative area for assessing FIQ. Our analysis of the face dataset of individuals from different demographic groups representing different skin tones indicates sclera as an alternative to measure dynamic range, over- and under-exposure of face using sclera region alone. The sclera region being agnostic to skin tone, i.e., demographic factors, provides equal utility as a fair FIQA as shown by our Error-vs-Discard Characteristic (EDC) curve analysis.
Authors: Weizhi Li
Abstract: Unlike image classification and annotation, for which deep network models have achieved dominating superior performances compared to traditional computer vision algorithms, deep learning for automatic image segmentation still faces critical challenges. One of such hurdles is to obtain ground-truth segmentations as the training labels for deep network training. Especially when we study biomedical images, such as histopathological images (histo-images), it is unrealistic to ask for manual segmentation labels as the ground truth for training due to the fine image resolution as well as the large image size and complexity. In this paper, instead of relying on clean segmentation labels, we study whether and how integrating imperfect or noisy segmentation results from off-the-shelf segmentation algorithms may help achieve better segmentation results through a new Adaptive Noise-Tolerant Network (ANTN) model. We extend the noisy label deep learning to image segmentation with two novel aspects: (1) multiple noisy labels can be integrated into one deep learning model; (2) noisy segmentation modeling, including probabilistic parameters, is adaptive, depending on the given testing image appearance. Implementation of the new ANTN model on both the synthetic data and real-world histo-images demonstrates its effectiveness and superiority over off-the-shelf and other existing deep-learning-based image segmentation algorithms.
Authors: Alejandro Lozano, Min Woo Sun, James Burgess, Liangyu Chen, Jeffrey J Nirschl, Jeffrey Gu, Ivan Lopez, Josiah Aklilu, Austin Wolfgang Katzer, Collin Chiu, Anita Rau, Xiaohan Wang, Yuhui Zhang, Alfred Seunghoon Song, Robert Tibshirani, Serena Yeung-Levy
Abstract: The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset.Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally.On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.
Authors: Wassim Kabbani, Tristan Le Pessot, Kiran Raja, Raghavendra Ramachandra, Christoph Busch
Abstract: Acquiring face images of sufficiently high quality is important for online ID and travel document issuance applications using face recognition systems (FRS). Low-quality, manipulated (intentionally or unintentionally), or distorted images degrade the FRS performance and facilitate documents' misuse. Securing quality for enrolment images, especially in the unsupervised self-enrolment scenario via a smartphone, becomes important to assure FRS performance. In this work, we focus on the less studied area of radial distortion (a.k.a., the fish-eye effect) in face images and its impact on FRS performance. We introduce an effective radial distortion detection model that can detect and flag radial distortion in the enrolment scenario. We formalize the detection model as a face image quality assessment (FIQA) algorithm and provide a careful inspection of the effect of radial distortion on FRS performance. Evaluation results show excellent detection results for the proposed models, and the study on the impact on FRS uncovers valuable insights into how to best use these models in operational systems.
Authors: Paul Melki (IMS), Lionel Bombrun (IMS), Boubacar Diallo (IMS), J\'er\^ome Dias (IMS), Jean-Pierre da Costa (IMS)
Abstract: Precision agriculture in general, and precision weeding in particular, have greatly benefited from the major advancements in deep learning and computer vision. A large variety of commercial robotic solutions are already available and deployed. However, the adoption by farmers of such solutions is still low for many reasons, an important one being the lack of trust in these systems. This is in great part due to the opaqueness and complexity of deep neural networks and the manufacturers' inability to provide valid guarantees on their performance. Conformal prediction, a well-established methodology in the machine learning community, is an efficient and reliable strategy for providing trustworthy guarantees on the predictions of any black-box model under very minimal constraints. Bridging the gap between the safe machine learning and precision agriculture communities, this article showcases conformal prediction in action on the task of precision weeding through deep learning-based image classification. After a detailed presentation of the conformal prediction methodology and the development of a precision spraying pipeline based on a ''conformalized'' neural network and well-defined spraying decision rules, the article evaluates this pipeline on two real-world scenarios: one under in-distribution conditions, the other reflecting a near out-of-distribution setting. The results show that we are able to provide formal, i.e. certifiable, guarantees on spraying at least 90% of the weeds.
Authors: Zhongyang Li, Xin Yuan, Wei Liu, Xin Xu
Abstract: Cross-view object geo-localization (CVOGL) aims to locate an object of interest in a captured ground- or drone-view image within the satellite image. However, existing works treat ground-view and drone-view query images equivalently, overlooking their inherent viewpoint discrepancies and the spatial correlation between the query image and the satellite-view reference image. To this end, this paper proposes a novel View-specific Attention Geo-localization method (VAGeo) for accurate CVOGL. Specifically, VAGeo contains two key modules: view-specific positional encoding (VSPE) module and channel-spatial hybrid attention (CSHA) module. In object-level, according to the characteristics of different viewpoints of ground and drone query images, viewpoint-specific positional codings are designed to more accurately identify the click-point object of the query image in the VSPE module. In feature-level, a hybrid attention in the CSHA module is introduced by combining channel attention and spatial attention mechanisms simultaneously for learning discriminative features. Extensive experimental results demonstrate that the proposed VAGeo gains a significant performance improvement, i.e., improving acc@0.25/acc@0.5 on the CVOGL dataset from 45.43%/42.24% to 48.21%/45.22% for ground-view, and from 61.97%/57.66% to 66.19%/61.87% for drone-view.
Authors: Wassim Kabbani, Kiran Raja, Raghavendra Ramachandra, Christoph Busch
Abstract: A face image is a mandatory part of ID and travel documents. Obtaining high-quality face images when issuing such documents is crucial for both human examiners and automated face recognition systems. In several international standards, face image quality requirements are intricate and defined in detail. Identifying and understanding non-compliance or defects in the submitted face images is crucial for both issuing authorities and applicants. In this work, we introduce FaceOracle, an LLM-powered AI assistant that helps its users analyze a face image in a natural conversational manner using standard compliant algorithms. Leveraging the power of LLMs, users can get explanations of various face image quality concepts as well as interpret the outcome of face image quality assessment (FIQA) algorithms. We implement a proof-of-concept that demonstrates how experts at an issuing authority could integrate FaceOracle into their workflow to analyze, understand, and communicate their decisions more efficiently, resulting in enhanced productivity.
Authors: Sirnam Swetha, Hilde Kuehne, Mubarak Shah
Abstract: Temporal logical understanding, a core facet of human cognition, plays a pivotal role in capturing complex sequential events and their temporal relationships within videos. This capability is particularly crucial in tasks like Video Question Answering (VideoQA), where the goal is to process visual data over time together with textual data to provide coherent answers. However, current VideoQA benchmarks devote little focus to evaluating this critical skill due to the challenge of annotating temporal logic. Despite the advancement of vision-language models, assessing their temporal logical reasoning powers remains a challenge, primarily due to the lack QA pairs that demand formal, complex temporal reasoning. To bridge this gap, we introduce the TimeLogic QA (TLQA) framework to automatically generate the QA pairs, specifically designed to evaluate the temporal logical understanding. To this end, TLQA leverages temporal annotations from existing video datasets together with temporal operators derived from logic theory to construct questions that test understanding of event sequences and their temporal relationships. TLQA framework is generic and scalable, capable of leveraging both, existing video action datasets with temporal action segmentation annotations, or video datasets with temporal scene graph annotations, to automatically generate temporal logical questions. We leverage 4 datasets, STAR, Breakfast, AGQA, and CrossTask, and generate two VideoQA dataset variants - small (TLQA-S) and large (TLQA-L) - containing 2k and 10k QA pairs for each category, resulting in 32k and 160k total pairs per dataset. We undertake a comprehensive evaluation of leading-edge VideoQA models, employing the TLQA to benchmark their temporal logical understanding capabilities. We assess the VideoQA model's temporal reasoning performance on 16 categories of temporal logic with varying temporal complexity.
Authors: Andrzej D. Dobrzycki, Ana M. Bernardos, Luca Bergesio, Andrzej Pomirski, Daniel S\'aez-Trigueros
Abstract: Accurate human posture classification in images and videos is crucial for automated applications across various fields, including work safety, physical rehabilitation, sports training, or daily assisted living. Recently, multimodal learning methods, such as Contrastive Language-Image Pretraining (CLIP), have advanced significantly in jointly understanding images and text. This study aims to assess the effectiveness of CLIP in classifying human postures, focusing on its application in yoga. Despite the initial limitations of the zero-shot approach, applying transfer learning on 15,301 images (real and synthetic) with 82 classes has shown promising results. The article describes the full procedure for fine-tuning, including the choice for image description syntax, models and hyperparameters adjustment. The fine-tuned CLIP model, tested on 3826 images, achieves an accuracy of over 85%, surpassing the current state-of-the-art of previous works on the same dataset by approximately 6%, its training time being 3.5 times lower than what is needed to fine-tune a YOLOv8-based model. For more application-oriented scenarios, with smaller datasets of six postures each, containing 1301 and 401 training images, the fine-tuned models attain an accuracy of 98.8% and 99.1%, respectively. Furthermore, our experiments indicate that training with as few as 20 images per pose can yield around 90% accuracy in a six-class dataset. This study demonstrates that this multimodal technique can be effectively used for yoga pose classification, and possibly for human posture classification, in general. Additionally, CLIP inference time (around 7 ms) supports that the model can be integrated into automated systems for posture evaluation, e.g., for developing a real-time personal yoga assistant for performance assessment.
Authors: Tieyuan Chen, Huabin Liu, Yi Wang, Yihang Chen, Tianyao He, Chaofan Gan, Huanyu He, Weiyao Lin
Abstract: Video causal reasoning aims to achieve a high-level understanding of videos from a causal perspective. However, it exhibits limitations in its scope, primarily executed in a question-answering paradigm and focusing on brief video segments containing isolated events and basic causal relations, lacking comprehensive and structured causality analysis for videos with multiple interconnected events. To fill this gap, we introduce a new task and dataset, Multi-Event Causal Discovery (MECD). It aims to uncover the causal relations between events distributed chronologically across long videos. Given visual segments and textual descriptions of events, MECD identifies the causal associations between these events to derive a comprehensive and structured event-level video causal graph explaining why and how the result event occurred. To address the challenges of MECD, we devise a novel framework inspired by the Granger Causality method, incorporating an efficient mask-based event prediction model to perform an Event Granger Test. It estimates causality by comparing the predicted result event when premise events are masked versus unmasked. Furthermore, we integrate causal inference techniques such as front-door adjustment and counterfactual inference to mitigate challenges in MECD like causality confounding and illusory causality. Additionally, context chain reasoning is introduced to conduct more robust and generalized reasoning. Experiments validate the effectiveness of our framework in reasoning complete causal relations, outperforming GPT-4o and VideoChat2 by 5.77% and 2.70%, respectively. Further experiments demonstrate that causal relation graphs can also contribute to downstream video understanding tasks such as video question answering and video event prediction.
Authors: Tieyuan Chen, Huabin Liu, Chern Hong Lim, John See, Xing Gao, Junhui Hou, Weiyao Lin
Abstract: Continual learning aims to acquire new knowledge while retaining past information. Class-incremental learning (CIL) presents a challenging scenario where classes are introduced sequentially. For video data, the task becomes more complex than image data because it requires learning and preserving both spatial appearance and temporal action involvement. To address this challenge, we propose a novel exemplar-free framework that equips separate spatiotemporal adapters to learn new class patterns, accommodating the incremental information representation requirements unique to each class. While separate adapters are proven to mitigate forgetting and fit unique requirements, naively applying them hinders the intrinsic connection between spatial and temporal information increments, affecting the efficiency of representing newly learned class information. Motivated by this, we introduce two key innovations from a causal perspective. First, a causal distillation module is devised to maintain the relation between spatial-temporal knowledge for a more efficient representation. Second, a causal compensation mechanism is proposed to reduce the conflicts during increment and memorization between different types of information. Extensive experiments conducted on benchmark datasets demonstrate that our framework can achieve new state-of-the-art results, surpassing current example-based methods by 4.2% in accuracy on average.
Authors: Oikantik Nath, Hanani Bathina, Mohammed Safi Ur Rahman Khan, Mitesh M. Khapra
Abstract: Recent advancements in Vision-Language Models (VLMs) have opened new possibilities in automatic grading of handwritten student responses, particularly in mathematics. However, a comprehensive study to test the ability of VLMs to evaluate and reason over handwritten content remains absent. To address this gap, we introduce FERMAT, a benchmark designed to assess the ability of VLMs to detect, localize and correct errors in handwritten mathematical content. FERMAT spans four key error dimensions - computational, conceptual, notational, and presentation - and comprises over 2,200 handwritten math solutions derived from 609 manually curated problems from grades 7-12 with intentionally introduced perturbations. Using FERMAT we benchmark nine VLMs across three tasks: error detection, localization, and correction. Our results reveal significant shortcomings in current VLMs in reasoning over handwritten text, with Gemini-1.5-Pro achieving the highest error correction rate (77%). We also observed that some models struggle with processing handwritten content, as their accuracy improves when handwritten inputs are replaced with printed text or images. These findings highlight the limitations of current VLMs and reveal new avenues for improvement. We release FERMAT and all the associated resources in the open-source to drive further research.
Authors: Oleg Perezyabov, Mikhail Gavrilenkov, Ilya Afanasyev
Abstract: This paper is devoted to the detection of objects on a road, performed with a combination of two methods based on both the use of depth information and video analysis of data from a stereo camera. Since neither the time of the appearance of an object on the road, nor its size and shape is known in advance, ML/DL-based approaches are not applicable. The task becomes more complicated due to variations in artificial illumination, inhomogeneous road surface texture, and unknown character and features of the object. To solve this problem we developed the depth and image fusion method that complements a search of small contrast objects by RGB-based method, and obstacle detection by stereo image-based approach with SLIC superpixel segmentation. We conducted experiments with static and low speed obstacles in an underground parking lot and demonstrated the successful work of the developed technique for detecting and even tracking small objects, which can be parking infrastructure objects, things left on the road, wheels, dropped boxes, etc.
Authors: Chong Zhou, Chenchen Zhu, Yunyang Xiong, Saksham Suri, Fanyi Xiao, Lemeng Wu, Raghuraman Krishnamoorthi, Bo Dai, Chen Change Loy, Vikas Chandra, Bilge Soran
Abstract: On top of Segment Anything Model (SAM), SAM 2 further extends its capability from image to video inputs through a memory bank mechanism and obtains a remarkable performance compared with previous methods, making it a foundation model for video segmentation task. In this paper, we aim at making SAM 2 much more efficient so that it even runs on mobile devices while maintaining a comparable performance. Despite several works optimizing SAM for better efficiency, we find they are not sufficient for SAM 2 because they all focus on compressing the image encoder, while our benchmark shows that the newly introduced memory attention blocks are also the latency bottleneck. Given this observation, we propose EdgeTAM, which leverages a novel 2D Spatial Perceiver to reduce the computational cost. In particular, the proposed 2D Spatial Perceiver encodes the densely stored frame-level memories with a lightweight Transformer that contains a fixed set of learnable queries. Given that video segmentation is a dense prediction task, we find preserving the spatial structure of the memories is essential so that the queries are split into global-level and patch-level groups. We also propose a distillation pipeline that further improves the performance without inference overhead. As a result, EdgeTAM achieves 87.7, 70.0, 72.3, and 71.7 J&F on DAVIS 2017, MOSE, SA-V val, and SA-V test, while running at 16 FPS on iPhone 15 Pro Max.
Authors: Li Liang, Naveed Akhtar, Jordan Vice, Xiangrui Kong, Ajmal Saeed Mian
Abstract: 3D semantic scene completion is critical for multiple downstream tasks in autonomous systems. It estimates missing geometric and semantic information in the acquired scene data. Due to the challenging real-world conditions, this task usually demands complex models that process multi-modal data to achieve acceptable performance. We propose a unique neural model, leveraging advances from the state space and diffusion generative modeling to achieve remarkable 3D semantic scene completion performance with monocular image input. Our technique processes the data in the conditioned latent space of a variational autoencoder where diffusion modeling is carried out with an innovative state space technique. A key component of our neural network is the proposed Skimba (Skip Mamba) denoiser, which is adept at efficiently processing long-sequence data. The Skimba diffusion model is integral to our 3D scene completion network, incorporating a triple Mamba structure, dimensional decomposition residuals and varying dilations along three directions. We also adopt a variant of this network for the subsequent semantic segmentation stage of our method. Extensive evaluation on the standard SemanticKITTI and SSCBench-KITTI360 datasets show that our approach not only outperforms other monocular techniques by a large margin, it also achieves competitive performance against stereo methods. The code is available at https://github.com/xrkong/skimba
Authors: Renkai Li, Xin Yuan, Wei Liu, Xin Xu
Abstract: Video-based person re-identification (ReID) has become increasingly important due to its applications in video surveillance applications. By employing events in video-based person ReID, more motion information can be provided between continuous frames to improve recognition accuracy. Previous approaches have assisted by introducing event data into the video person ReID task, but they still cannot avoid the privacy leakage problem caused by RGB images. In order to avoid privacy attacks and to take advantage of the benefits of event data, we consider using only event data. To make full use of the information in the event stream, we propose a Cross-Modality and Temporal Collaboration (CMTC) network for event-based video person ReID. First, we design an event transform network to obtain corresponding auxiliary information from the input of raw events. Additionally, we propose a differential modality collaboration module to balance the roles of events and auxiliaries to achieve complementary effects. Furthermore, we introduce a temporal collaboration module to exploit motion information and appearance cues. Experimental results demonstrate that our method outperforms others in the task of event-based video person ReID.
Authors: Zhimeng Xin, Tianxu Wu, Shiming Chen, Shuo Ye, Zijing Xie, Yixiong Zou, Xinge You, Yufei Guo
Abstract: Camouflaged object detection (COD) primarily relies on semantic or instance segmentation methods. While these methods have made significant advancements in identifying the contours of camouflaged objects, they may be inefficient or cost-effective for tasks that only require the specific location of the object. Object detection algorithms offer an optimized solution for Realistic Camouflaged Object Detection (RCOD) in such cases. However, detecting camouflaged objects remains a formidable challenge due to the high degree of similarity between the features of the objects and their backgrounds. Unlike segmentation methods that perform pixel-wise comparisons to differentiate between foreground and background, object detectors omit this analysis, further aggravating the challenge. To solve this problem, we propose a camouflage-aware feature refinement (CAFR) strategy. Since camouflaged objects are not rare categories, CAFR fully utilizes a clear perception of the current object within the prior knowledge of large models to assist detectors in deeply understanding the distinctions between background and foreground. Specifically, in CAFR, we introduce the Adaptive Gradient Propagation (AGP) module that fine-tunes all feature extractor layers in large detection models to fully refine class-specific features from camouflaged contexts. We then design the Sparse Feature Refinement (SFR) module that optimizes the transformer-based feature extractor to focus primarily on capturing class-specific features in camouflaged scenarios. To facilitate the assessment of RCOD tasks, we manually annotate the labels required for detection on three existing segmentation COD datasets, creating a new benchmark for RCOD tasks. Code and datasets are available at: https://github.com/zhimengXin/RCOD.
Authors: Kankana Roy, Lars Kr\"amer, Sebastian Domaschke, Malik Haris, Roland Aydin, Fabian Isensee, Martin Held
Abstract: Learning from tabular data is of paramount importance, as it complements the conventional analysis of image and video data by providing a rich source of structured information that is often critical for comprehensive understanding and decision-making processes. We present Multi-task Contrastive Masked Tabular Modeling (MT-CMTM), a novel method aiming to enhance tabular models by leveraging the correlation between tabular data and corresponding images. MT-CMTM employs a dual strategy combining contrastive learning with masked tabular modeling, optimizing the synergy between these data modalities. Central to our approach is a 1D Convolutional Neural Network with residual connections and an attention mechanism (1D-ResNet-CBAM), designed to efficiently process tabular data without relying on images. This enables MT-CMTM to handle purely tabular data for downstream tasks, eliminating the need for potentially costly image acquisition and processing. We evaluated MT-CMTM on the DVM car dataset, which is uniquely suited for this particular scenario, and the newly developed HIPMP dataset, which connects membrane fabrication parameters with image data. Our MT-CMTM model outperforms the proposed tabular 1D-ResNet-CBAM, which is trained from scratch, achieving a relative 1.48% improvement in relative MSE on HIPMP and a 2.38% increase in absolute accuracy on DVM. These results demonstrate MT-CMTM's robustness and its potential to advance the field of multi-modal learning.
Authors: Xinyang Zhou, Fanyue Wei, Lixin Duan, Wen Li
Abstract: Given a textual query along with a corresponding video, the objective of moment retrieval aims to localize the moments relevant to the query within the video. While commendable results have been demonstrated by existing transformer-based approaches, predicting the accurate temporal span of the target moment is currently still a major challenge. In this paper, we reveal that a crucial reason stems from the spurious correlation between the text queries and the moment context. Namely, the model may associate the textual query with the background frames rather than the target moment. To address this issue, we propose a temporal dynamic learning approach for moment retrieval, where two strategies are designed to mitigate the spurious correlation. First, we introduce a novel video synthesis approach to construct a dynamic context for the relevant moment. With separate yet similar videos mixed up, the synthesis approach empowers our model to attend to the target moment of the corresponding query under various dynamic contexts. Second, we enhance the representation by learning temporal dynamics. Besides the visual representation, text queries are aligned with temporal dynamic representations, which enables our model to establish a non-spurious correlation between the query-related moment and context. With the aforementioned proposed method, the spurious correlation issue in moment retrieval can be largely alleviated. Our method establishes a new state-of-the-art performance on two popular benchmarks of moment retrieval, \ie, QVHighlights and Charades-STA. In addition, the detailed ablation analyses demonstrate the effectiveness of the proposed strategies. Our code will be publicly available.
Authors: Sujia Wang, Xiangwei Shen, Yansong Tang, Xin Dong, Wenjia Geng, Lei Chen
Abstract: Repetitive action counting (RAC) aims to estimate the number of class-agnostic action occurrences in a video without exemplars. Most current RAC methods rely on a raw frame-to-frame similarity representation for period prediction. However, this approach can be significantly disrupted by common noise such as action interruptions and inconsistencies, leading to sub-optimal counting performance in realistic scenarios. In this paper, we introduce a foreground localization optimization objective into similarity representation learning to obtain more robust and efficient video features. We propose a Localization-Aware Multi-Scale Representation Learning (LMRL) framework. Specifically, we apply a Multi-Scale Period-Aware Representation (MPR) with a scale-specific design to accommodate various action frequencies and learn more flexible temporal correlations. Furthermore, we introduce the Repetition Foreground Localization (RFL) method, which enhances the representation by coarsely identifying periodic actions and incorporating global semantic information. These two modules can be jointly optimized, resulting in a more discerning periodic action representation. Our approach significantly reduces the impact of noise, thereby improving counting accuracy. Additionally, the framework is designed to be scalable and adaptable to different types of video content. Experimental results on the RepCountA and UCFRep datasets demonstrate that our proposed method effectively handles repetitive action counting.
Authors: Zuzana Berger Haladova, Michal Zrubec, Zuzana Cernekova
Abstract: Roadside billboards and other forms of outdoor advertising play a crucial role in marketing initiatives; however, they can also distract drivers, potentially contributing to accidents. This study delves into the significance of roadside advertising in images captured from a driver's perspective. Firstly, it evaluates the effectiveness of neural networks in detecting advertising along roads, focusing on the YOLOv5 and Faster R-CNN models. Secondly, the study addresses the determination of billboard significance using methods for saliency extraction. The UniSal and SpectralResidual methods were employed to create saliency maps for each image. The study establishes a database of eye tracking sessions captured during city highway driving to assess the saliency models.
Authors: Daniel Steininger, Julia Simon, Andreas Trondl, Markus Murschitz
Abstract: Timber represents an increasingly valuable and versatile resource. However, forestry operations such as harvesting, handling and measuring logs still require substantial human labor in remote environments posing significant safety risks. Progressively automating these tasks has the potential of increasing their efficiency as well as safety, but requires an accurate detection of individual logs as well as live trees and their context. Although initial approaches have been proposed for this challenging application domain, specialized data and algorithms are still too scarce to develop robust solutions. To mitigate this gap, we introduce the TimberVision dataset, consisting of more than 2k annotated RGB images containing a total of 51k trunk components including cut and lateral surfaces, thereby surpassing any existing dataset in this domain in terms of both quantity and detail by a large margin. Based on this data, we conduct a series of ablation experiments for oriented object detection and instance segmentation and evaluate the influence of multiple scene parameters on model performance. We introduce a generic framework to fuse the components detected by our models for both tasks into unified trunk representations. Furthermore, we automatically derive geometric properties and apply multi-object tracking to further enhance robustness. Our detection and tracking approach provides highly descriptive and accurate trunk representations solely from RGB image data, even under challenging environmental conditions. Our solution is suitable for a wide range of application scenarios and can be readily combined with other sensor modalities.
Authors: Zhipeng Deng, Zhe Xu, Tsuyoshi Isshiki, Yefeng Zheng
Abstract: Medical image segmentation is challenging due to the diversity of medical images and the lack of labeled data, which motivates recent developments in federated semi-supervised learning (FSSL) to leverage a large amount of unlabeled data from multiple centers for model training without sharing raw data. However, what remains under-explored in FSSL is the domain shift problem which may cause suboptimal model aggregation and low effectivity of the utilization of unlabeled data, eventually leading to unsatisfactory performance in unseen domains. In this paper, we explore this previously ignored scenario, namely domain generalized federated semi-supervised learning (FedSemiDG), which aims to learn a model in a distributed manner from multiple domains with limited labeled data and abundant unlabeled data such that the model can generalize well to unseen domains. We present a novel framework, Federated Generalization-Aware SemiSupervised Learning (FGASL), to address the challenges in FedSemiDG by effectively tackling critical issues at both global and local levels. Globally, we introduce Generalization-Aware Aggregation (GAA), assigning adaptive weights to local models based on their generalization performance. Locally, we use a Dual-Teacher Adaptive Pseudo Label Refinement (DR) strategy to combine global and domain-specific knowledge, generating more reliable pseudo labels. Additionally, Perturbation-Invariant Alignment (PIA) enforces feature consistency under perturbations, promoting domain-invariant learning. Extensive experiments on three medical segmentation tasks (cardiac MRI, spine MRI and bladder cancer MRI) demonstrate that our method significantly outperforms state-of-the-art FSSL and domain generalization approaches, achieving robust generalization on unseen domains.
Authors: Xianping Ma, Ziyao Wang, Yin Hu, Xiaokang Zhang, Man-On Pun
Abstract: Semantic segmentation plays a crucial role in remote sensing applications, where the accurate extraction and representation of features are essential for high-quality results. Despite the widespread use of encoder-decoder architectures, existing methods often struggle with fully utilizing the high-dimensional features extracted by the encoder and efficiently recovering detailed information during decoding. To address these problems, we propose a novel semantic segmentation network, namely DeepKANSeg, including two key innovations based on the emerging Kolmogorov Arnold Network (KAN). Notably, the advantage of KAN lies in its ability to decompose high-dimensional complex functions into univariate transformations, enabling efficient and flexible representation of intricate relationships in data. First, we introduce a KAN-based deep feature refinement module, namely DeepKAN to effectively capture complex spatial and rich semantic relationships from high-dimensional features. Second, we replace the traditional multi-layer perceptron (MLP) layers in the global-local combined decoder with KAN-based linear layers, namely GLKAN. This module enhances the decoder's ability to capture fine-grained details during decoding. To evaluate the effectiveness of the proposed method, experiments are conducted on two well-known fine-resolution remote sensing benchmark datasets, namely ISPRS Vaihingen and ISPRS Potsdam. The results demonstrate that the KAN-enhanced segmentation model achieves superior performance in terms of accuracy compared to state-of-the-art methods. They highlight the potential of KANs as a powerful alternative to traditional architectures in semantic segmentation tasks. Moreover, the explicit univariate decomposition provides improved interpretability, which is particularly beneficial for applications requiring explainable learning in remote sensing.
Authors: Yasiru Ranasinghe, Vibashan VS, James Uplinger, Celso De Melo, Vishal M. Patel
Abstract: Automatic target recognition (ATR) plays a critical role in tasks such as navigation and surveillance, where safety and accuracy are paramount. In extreme use cases, such as military applications, these factors are often challenged due to the presence of unknown terrains, environmental conditions, and novel object categories. Current object detectors, including open-world detectors, lack the ability to confidently recognize novel objects or operate in unknown environments, as they have not been exposed to these new conditions. However, Large Vision-Language Models (LVLMs) exhibit emergent properties that enable them to recognize objects in varying conditions in a zero-shot manner. Despite this, LVLMs struggle to localize objects effectively within a scene. To address these limitations, we propose a novel pipeline that combines the detection capabilities of open-world detectors with the recognition confidence of LVLMs, creating a robust system for zero-shot ATR of novel classes and unknown domains. In this study, we compare the performance of various LVLMs for recognizing military vehicles, which are often underrepresented in training datasets. Additionally, we examine the impact of factors such as distance range, modality, and prompting methods on the recognition performance, providing insights into the development of more reliable ATR systems for novel conditions and classes.
Authors: Shuo Zhang, Runpu Wei, Kongming Liang
Abstract: The rapid advancements in generative models, particularly diffusion-based techniques, have revolutionized image inpainting tasks by enabling the generation of high-fidelity and diverse content. However, object removal remains under-explored as a specific subset of inpainting, facing challenges such as inadequate semantic understanding and the unintended generation of artifacts. Existing datasets for object removal often rely on synthetic data, which fails to align with real-world scenarios, limiting model performance. Although some real-world datasets address these issues partially, they suffer from scalability, annotation inefficiencies, and limited realism in physical phenomena such as lighting and shadows. To address these limitations, this paper introduces a novel approach to object removal by constructing a high-resolution real-world dataset through long-duration video capture with fixed camera settings. Leveraging advanced tools such as Grounding-DINO, Segment-Anything-Model, and MASA for automated annotation, we provides image, background, and mask pairs while significantly reducing annotation time and labor. With our efficient annotation pipeline, we release the first fully open, high-resolution real-world dataset for object removal, and improved performance in object removal tasks through fine-tuning of pre-trained diffusion models.
Authors: Xiyue Zhu, Dou Hoon Kwark, Ruike Zhu, Kaiwen Hong, Yiqi Tao, Shirui Luo, Yudu Li, Zhi-Pei Liang, Volodymyr Kindratenko
Abstract: Despite success in volume-to-volume translations in medical images, most existing models struggle to effectively capture the inherent volumetric distribution using 3D representations. The current state-of-the-art approach combines multiple 2D-based networks through weighted averaging, thereby neglecting the 3D spatial structures. Directly training 3D models in medical imaging presents significant challenges due to high computational demands and the need for large-scale datasets. To address these challenges, we introduce Diff-Ensembler, a novel hybrid 2D-3D model for efficient and effective volumetric translations by ensembling perpendicularly trained 2D diffusion models with a 3D network in each diffusion step. Moreover, our model can naturally be used to ensemble diffusion models conditioned on different modalities, allowing flexible and accurate fusion of input conditions. Extensive experiments demonstrate that Diff-Ensembler attains superior accuracy and volumetric realism in 3D medical image super-resolution and modality translation. We further demonstrate the strength of our model's volumetric realism using tumor segmentation as a downstream task.
Authors: S. B. van Rooij, G. J. Burghouts
Abstract: Localizing object parts precisely is essential for tasks such as object recognition and robotic manipulation. Recent part segmentation methods require extensive training data and labor-intensive annotations. Segment-Anything Model (SAM) has demonstrated good performance on a wide range of segmentation problems, but requires (manual) positional prompts to guide it where to segment. Furthermore, since it has been trained on full objects instead of object parts, it is prone to over-segmentation of parts. To address this, we propose a novel approach that guides SAM towards the relevant object parts. Our method learns positional prompts from coarse patch annotations that are easier and cheaper to acquire. We train classifiers on image patches to identify part classes and aggregate patches into regions of interest (ROIs) with positional prompts. SAM is conditioned on these ROIs and prompts. This approach, termed `Guided SAM', enhances efficiency and reduces manual effort, allowing effective part segmentation with minimal labeled data. We demonstrate the efficacy of Guided SAM on a dataset of car parts, improving the average IoU on state of the art models from 0.37 to 0.49 with annotations that are on average five times more efficient to acquire.
Authors: Fabio Montello, Ronja G\"uldenring, Simone Scardapane, Lazaros Nalpantidis
Abstract: Model compression is essential in the deployment of large Computer Vision models on embedded devices. However, static optimization techniques (e.g. pruning, quantization, etc.) neglect the fact that different inputs have different complexities, thus requiring different amount of computations. Dynamic Neural Networks allow to condition the number of computations to the specific input. The current literature on the topic is very extensive and fragmented. We present a comprehensive survey that synthesizes and unifies existing Dynamic Neural Networks research in the context of Computer Vision. Additionally, we provide a logical taxonomy based on which component of the network is adaptive: the output, the computation graph or the input. Furthermore, we argue that Dynamic Neural Networks are particularly beneficial in the context of Sensor Fusion for better adaptivity, noise reduction and information prioritization. We present preliminary works in this direction.
Authors: Wenping Jin, Li Zhu, Jing Sun
Abstract: Weakly supervised violence detection refers to the technique of training models to identify violent segments in videos using only video-level labels. Among these approaches, multimodal violence detection, which integrates modalities such as audio and optical flow, holds great potential. Existing methods in this domain primarily focus on designing multimodal fusion models to address modality discrepancies. In contrast, we take a different approach; leveraging the inherent discrepancies across modalities in violence event representation to propose a novel multimodal semantic feature alignment method. This method sparsely maps the semantic features of local, transient, and less informative modalities ( such as audio and optical flow ) into the more informative RGB semantic feature space. Through an iterative process, the method identifies the suitable no-zero feature matching subspace and aligns the modality-specific event representations based on this subspace, enabling the full exploitation of information from all modalities during the subsequent modality fusion stage. Building on this, we design a new weakly supervised violence detection framework that consists of unimodal multiple-instance learning for extracting unimodal semantic features, multimodal alignment, multimodal fusion, and final detection. Experimental results on benchmark datasets demonstrate the effectiveness of our method, achieving an average precision (AP) of 86.07% on the XD-Violence dataset. Our code is available at https://github.com/xjpp2016/MAVD.
Authors: Yaqing Ding, Viktor Kocur, Zuzana Berger Haladov\'a, Qianliang Wu, Shen Cai, Jian Yang, Zuzana Kukelova
Abstract: In this paper, we propose a novel approach for recovering focal lengths from three-view homographies. By examining the consistency of normal vectors between two homographies, we derive new explicit constraints between the focal lengths and homographies using an elimination technique. We demonstrate that three-view homographies provide two additional constraints, enabling the recovery of one or two focal lengths. We discuss four possible cases, including three cameras having an unknown equal focal length, three cameras having two different unknown focal lengths, three cameras where one focal length is known, and the other two cameras have equal or different unknown focal lengths. All the problems can be converted into solving polynomials in one or two unknowns, which can be efficiently solved using Sturm sequence or hidden variable technique. Evaluation using both synthetic and real data shows that the proposed solvers are both faster and more accurate than methods relying on existing two-view solvers. The code and data are available on https://github.com/kocurvik/hf
Authors: Difei Gu, Yunhe Gao, Yang Zhou, Mu Zhou, Dimitris Metaxas
Abstract: Automated chest radiographs interpretation requires both accurate disease classification and detailed radiology report generation, presenting a significant challenge in the clinical workflow. Current approaches either focus on classification accuracy at the expense of interpretability or generate detailed but potentially unreliable reports through image captioning techniques. In this study, we present RadAlign, a novel framework that combines the predictive accuracy of vision-language models (VLMs) with the reasoning capabilities of large language models (LLMs). Inspired by the radiologist's workflow, RadAlign first employs a specialized VLM to align visual features with key medical concepts, achieving superior disease classification with an average AUC of 0.885 across multiple diseases. These recognized medical conditions, represented as text-based concepts in the aligned visual-language space, are then used to prompt LLM-based report generation. Enhanced by a retrieval-augmented generation mechanism that grounds outputs in similar historical cases, RadAlign delivers superior report quality with a GREEN score of 0.678, outperforming state-of-the-art methods' 0.634. Our framework maintains strong clinical interpretability while reducing hallucinations, advancing automated medical imaging and report analysis through integrated predictive and generative AI. Code is available at https://github.com/difeigu/RadAlign.
Authors: Tharun Anand, Aryan Garg, Kaushik Mitra
Abstract: Facial video editing has become increasingly important for content creators, enabling the manipulation of facial expressions and attributes. However, existing models encounter challenges such as poor editing quality, high computational costs and difficulties in preserving facial identity across diverse edits. Additionally, these models are often constrained to editing predefined facial attributes, limiting their flexibility to diverse editing prompts. To address these challenges, we propose a novel facial video editing framework that leverages the rich latent space of pre-trained text-to-image (T2I) diffusion models and fine-tune them specifically for facial video editing tasks. Our approach introduces a targeted fine-tuning scheme that enables high quality, localized, text-driven edits while ensuring identity preservation across video frames. Additionally, by using pre-trained T2I models during inference, our approach significantly reduces editing time by 80%, while maintaining temporal consistency throughout the video sequence. We evaluate the effectiveness of our approach through extensive testing across a wide range of challenging scenarios, including varying head poses, complex action sequences, and diverse facial expressions. Our method consistently outperforms existing techniques, demonstrating superior performance across a broad set of metrics and benchmarks.
Authors: Shiman Zhang, Lakshmikar Reddy Polamreddy, Youshan Zhang
Abstract: Canine cardiomegaly, marked by an enlarged heart, poses serious health risks if undetected, requiring accurate diagnostic methods. Current detection models often rely on small, poorly annotated datasets and struggle to generalize across diverse imaging conditions, limiting their real-world applicability. To address these issues, we propose a Confident Pseudo-labeled Diffusion Augmentation (CDA) model for identifying canine cardiomegaly. Our approach addresses the challenge of limited high-quality training data by employing diffusion models to generate synthetic X-ray images and annotate Vertebral Heart Score key points, thereby expanding the dataset. We also employ a pseudo-labeling strategy with Monte Carlo Dropout to select high-confidence labels, refine the synthetic dataset, and improve accuracy. Iteratively incorporating these labels enhances the model's performance, overcoming the limitations of existing approaches. Experimental results show that the CDA model outperforms traditional methods, achieving state-of-the-art accuracy in canine cardiomegaly detection. The code implementation is available at https://github.com/Shira7z/CDA.
Authors: Varun Biyyala, Bharat Chanderprakash Kathuria, Jialu Li, Youshan Zhang
Abstract: Video editing models have advanced significantly, but evaluating their performance remains challenging. Traditional metrics, such as CLIP text and image scores, often fall short: text scores are limited by inadequate training data and hierarchical dependencies, while image scores fail to assess temporal consistency. We present SST-EM (Semantic, Spatial, and Temporal Evaluation Metric), a novel evaluation framework that leverages modern Vision-Language Models (VLMs), Object Detection, and Temporal Consistency checks. SST-EM comprises four components: (1) semantic extraction from frames using a VLM, (2) primary object tracking with Object Detection, (3) focused object refinement via an LLM agent, and (4) temporal consistency assessment using a Vision Transformer (ViT). These components are integrated into a unified metric with weights derived from human evaluations and regression analysis. The name SST-EM reflects its focus on Semantic, Spatial, and Temporal aspects of video evaluation. SST-EM provides a comprehensive evaluation of semantic fidelity and temporal smoothness in video editing. The source code is available in the \textbf{\href{https://github.com/custommetrics-sst/SST_CustomEvaluationMetrics.git}{GitHub Repository}}.
URLs: https://github.com/custommetrics-sst/SST_CustomEvaluationMetrics.git
Authors: Xingyi He, Hao Yu, Sida Peng, Dongli Tan, Zehong Shen, Hujun Bao, Xiaowei Zhou
Abstract: Image matching, which aims to identify corresponding pixel locations between images, is crucial in a wide range of scientific disciplines, aiding in image registration, fusion, and analysis. In recent years, deep learning-based image matching algorithms have dramatically outperformed humans in rapidly and accurately finding large amounts of correspondences. However, when dealing with images captured under different imaging modalities that result in significant appearance changes, the performance of these algorithms often deteriorates due to the scarcity of annotated cross-modal training data. This limitation hinders applications in various fields that rely on multiple image modalities to obtain complementary information. To address this challenge, we propose a large-scale pre-training framework that utilizes synthetic cross-modal training signals, incorporating diverse data from various sources, to train models to recognize and match fundamental structures across images. This capability is transferable to real-world, unseen cross-modality image matching tasks. Our key finding is that the matching model trained with our framework achieves remarkable generalizability across more than eight unseen cross-modality registration tasks using the same network weight, substantially outperforming existing methods, whether designed for generalization or tailored for specific tasks. This advancement significantly enhances the applicability of image matching technologies across various scientific disciplines and paves the way for new applications in multi-modality human and artificial intelligence analysis and beyond.
Authors: Xinyu Zhang, Zicheng Duan, Dong Gong, Lingqiao Liu
Abstract: In this paper, we address the challenge of generating temporally consistent videos with motion guidance. While many existing methods depend on additional control modules or inference-time fine-tuning, recent studies suggest that effective motion guidance is achievable without altering the model architecture or requiring extra training. Such approaches offer promising compatibility with various video generation foundation models. However, existing training-free methods often struggle to maintain consistent temporal coherence across frames or to follow guided motion accurately. In this work, we propose a simple yet effective solution that combines an initial-noise-based approach with a novel motion consistency loss, the latter being our key innovation. Specifically, we capture the inter-frame feature correlation patterns of intermediate features from a video diffusion model to represent the motion pattern of the reference video. We then design a motion consistency loss to maintain similar feature correlation patterns in the generated video, using the gradient of this loss in the latent space to guide the generation process for precise motion control. This approach improves temporal consistency across various motion control tasks while preserving the benefits of a training-free setup. Extensive experiments show that our method sets a new standard for efficient, temporally coherent video generation.
Authors: Xingchen Liu, Piyush Tayal, Jianyuan Wang, Jesus Zarzar, Tom Monnier, Konstantinos Tertikas, Jiali Duan, Antoine Toisoul, Jason Y. Zhang, Natalia Neverova, Andrea Vedaldi, Roman Shapovalov, David Novotny
Abstract: We introduce Uncommon Objects in 3D (uCO3D), a new object-centric dataset for 3D deep learning and 3D generative AI. uCO3D is the largest publicly-available collection of high-resolution videos of objects with 3D annotations that ensures full-360$^{\circ}$ coverage. uCO3D is significantly more diverse than MVImgNet and CO3Dv2, covering more than 1,000 object categories. It is also of higher quality, due to extensive quality checks of both the collected videos and the 3D annotations. Similar to analogous datasets, uCO3D contains annotations for 3D camera poses, depth maps and sparse point clouds. In addition, each object is equipped with a caption and a 3D Gaussian Splat reconstruction. We train several large 3D models on MVImgNet, CO3Dv2, and uCO3D and obtain superior results using the latter, showing that uCO3D is better for learning applications.
Authors: Jiacheng Cui, Zhaoyi Li, Xiaochen Ma, Xinyue Bi, Yaxin Luo, Zhiqiang Shen
Abstract: Dataset distillation aims to synthesize a smaller, representative dataset that preserves the essential properties of the original data, enabling efficient model training with reduced computational resources. Prior work has primarily focused on improving the alignment or matching process between original and synthetic data, or on enhancing the efficiency of distilling large datasets. In this work, we introduce ${\bf C}$ommittee ${\bf V}$oting for ${\bf D}$ataset ${\bf D}$istillation (CV-DD), a novel and orthogonal approach that leverages the collective wisdom of multiple models or experts to create high-quality distilled datasets. We start by showing how to establish a strong baseline that already achieves state-of-the-art accuracy through leveraging recent advancements and thoughtful adjustments in model design and optimization processes. By integrating distributions and predictions from a committee of models while generating high-quality soft labels, our method captures a wider spectrum of data features, reduces model-specific biases and the adverse effects of distribution shifts, leading to significant improvements in generalization. This voting-based strategy not only promotes diversity and robustness within the distilled dataset but also significantly reduces overfitting, resulting in improved performance on post-eval tasks. Extensive experiments across various datasets and IPCs (images per class) demonstrate that Committee Voting leads to more reliable and adaptable distilled data compared to single/multi-model distillation methods, demonstrating its potential for efficient and accurate dataset distillation. Code is available at: https://github.com/Jiacheng8/CV-DD.
Authors: Gent Wu
Abstract: Vision Transformers (ViTs) have demonstrated remarkable success on large-scale datasets, but their performance on smaller datasets often falls short of convolutional neural networks (CNNs). This paper explores the design and optimization of Tiny ViTs for small datasets, using CIFAR-10 as a benchmark. We systematically evaluate the impact of data augmentation, patch token initialization, low-rank compression, and multi-class token strategies on model performance. Our experiments reveal that low-rank compression of queries in Multi-Head Latent Attention (MLA) incurs minimal performance loss, indicating redundancy in ViTs. Additionally, introducing multiple CLS tokens improves global representation capacity, boosting accuracy. These findings provide a comprehensive framework for optimizing Tiny ViTs, offering practical insights for efficient and effective designs. Code is available at https://github.com/erow/PoorViTs.
Authors: Yiqin Zhang, Meiling Chen, Zhengjie Zhang
Abstract: Whether during the early days of popularization or in the present, the window setting in Computed Tomography (CT) has always been an indispensable part of the CT analysis process. Although research has investigated the capabilities of CT multi-window fusion in enhancing neural networks, there remains a paucity of domain-invariant, intuitively interpretable methodologies for Auto Window Setting. In this work, we propose an plug-and-play module originate from Tanh activation function, which is compatible with mainstream deep learning architectures. Starting from the physical principles of CT, we adhere to the principle of interpretability to ensure the module's reliability for medical implementations. The domain-invariant design facilitates observation of the preference decisions rendered by the adaptive mechanism from a clinically intuitive perspective. This enables the proposed method to be understood not only by experts in neural networks but also garners higher trust from clinicians. We confirm the effectiveness of the proposed method in multiple open-source datasets, yielding 10%~200% Dice improvements on hard segment targets.
Authors: Kancharagunta Kishan Babu, Ashreen Tabassum, Bommakanti Navaneeth, Tenneti Jahnavi, Yenka Akshaya
Abstract: In recent years, there has been a surge of research focused on underwater image enhancement using Generative Adversarial Networks (GANs), driven by the need to overcome the challenges posed by underwater environments. Issues such as light attenuation, scattering, and color distortion severely degrade the quality of underwater images, limiting their use in critical applications. Generative Adversarial Networks (GANs) have emerged as a powerful tool for enhancing underwater photos due to their ability to learn complex transformations and generate realistic outputs. These advancements have been applied to real-world applications, including marine biology and ecosystem monitoring, coral reef health assessment, underwater archaeology, and autonomous underwater vehicle (AUV) navigation. This paper explores all major approaches to underwater image enhancement, from physical and physics-free models to Convolutional Neural Network (CNN)-based models and state-of-the-art GAN-based methods. It provides a comprehensive analysis of these methods, evaluation metrics, datasets, and loss functions, offering a holistic view of the field. Furthermore, the paper delves into the limitations and challenges faced by current methods, such as generalization issues, high computational demands, and dataset biases, while suggesting potential directions for future research.
Authors: Yu-Cheng Chou, Gary Y. Li, Li Chen, Mohsen Zahiri, Naveen Balaraju, Shubham Patil, Bryson Hicks, Nikolai Schnittke, David O. Kessler, Jeffrey Shupp, Maria Parker, Cristiana Baloescu, Christopher Moore, Cynthia Gregory, Kenton Gregory, Balasundar Raju, Jochen Kruecker, Alvin Chen
Abstract: Developing reliable healthcare AI models requires training with representative and diverse data. In imbalanced datasets, model performance tends to plateau on the more prevalent classes while remaining low on less common cases. To overcome this limitation, we propose DiffUltra, the first generative AI technique capable of synthesizing realistic Lung Ultrasound (LUS) images with extensive lesion variability. Specifically, we condition the generative AI by the introduced Lesion-anatomy Bank, which captures the lesion's structural and positional properties from real patient data to guide the image synthesis.We demonstrate that DiffUltra improves consolidation detection by 5.6% in AP compared to the models trained solely on real patient data. More importantly, DiffUltra increases data diversity and prevalence of rare cases, leading to a 25% AP improvement in detecting rare instances such as large lung consolidations, which make up only 10% of the dataset.
Authors: Kunpeng Xu, Lifei Chen, Shengrui Wang
Abstract: Kernel-based subspace clustering, which addresses the nonlinear structures in data, is an evolving area of research. Despite noteworthy progressions, prevailing methodologies predominantly grapple with limitations relating to (i) the influence of predefined kernels on model performance; (ii) the difficulty of preserving the original manifold structures in the nonlinear space; (iii) the dependency of spectral-type strategies on the ideal block diagonal structure of the affinity matrix. This paper presents DKLM, a novel paradigm for kernel-induced nonlinear subspace clustering. DKLM provides a data-driven approach that directly learns the kernel from the data's self-representation, ensuring adaptive weighting and satisfying the multiplicative triangle inequality constraint, which enhances the robustness of the learned kernel. By leveraging this learned kernel, DKLM preserves the local manifold structure of data in a nonlinear space while promoting the formation of an optimal block-diagonal affinity matrix. A thorough theoretical examination of DKLM reveals its relationship with existing clustering paradigms. Comprehensive experiments on synthetic and real-world datasets demonstrate the effectiveness of the proposed method.
Authors: Yousef Heider, Fadi Aldakheel, Wolfgang Ehlers
Abstract: This work introduces a novel application for predicting the macroscopic intrinsic permeability tensor in deformable porous media, using a limited set of micro-CT images of real microgeometries. The primary goal is to develop an efficient, machine-learning (ML)-based method that overcomes the limitations of traditional permeability estimation techniques, which often rely on time-consuming experiments or computationally expensive fluid dynamics simulations. The novelty of this work lies in leveraging Convolutional Neural Networks (CNN) to predict pore-fluid flow behavior under deformation and anisotropic flow conditions. Particularly, the described approach employs binarized CT images of porous micro-structure as inputs to predict the symmetric second-order permeability tensor, a critical parameter in continuum porous media flow modeling. The methodology comprises four key steps: (1) constructing a dataset of CT images from Bentheim sandstone at different volumetric strain levels; (2) performing pore-scale simulations of single-phase flow using the lattice Boltzmann method (LBM) to generate permeability data; (3) training the CNN model with the processed CT images as inputs and permeability tensors as outputs; and (4) exploring techniques to improve model generalization, including data augmentation and alternative CNN architectures. Examples are provided to demonstrate the CNN's capability to accurately predict the permeability tensor, a crucial parameter in various disciplines such as geotechnical engineering, hydrology, and material science. An exemplary source code is made available for interested readers.
Authors: Meihua Dang, Anikait Singh, Linqi Zhou, Stefano Ermon, Jiaming Song
Abstract: RLHF techniques like DPO can significantly improve the generation quality of text-to-image diffusion models. However, these methods optimize for a single reward that aligns model generation with population-level preferences, neglecting the nuances of individual users' beliefs or values. This lack of personalization limits the efficacy of these models. To bridge this gap, we introduce PPD, a multi-reward optimization objective that aligns diffusion models with personalized preferences. With PPD, a diffusion model learns the individual preferences of a population of users in a few-shot way, enabling generalization to unseen users. Specifically, our approach (1) leverages a vision-language model (VLM) to extract personal preference embeddings from a small set of pairwise preference examples, and then (2) incorporates the embeddings into diffusion models through cross attention. Conditioning on user embeddings, the text-to-image models are fine-tuned with the DPO objective, simultaneously optimizing for alignment with the preferences of multiple users. Empirical results demonstrate that our method effectively optimizes for multiple reward functions and can interpolate between them during inference. In real-world user scenarios, with as few as four preference examples from a new user, our approach achieves an average win rate of 76\% over Stable Cascade, generating images that more accurately reflect specific user preferences.
Authors: Yiming Lin, Mawil Hasan, Rohan Kosalge, Alvin Cheung, Aditya G. Parameswaran
Abstract: Many documents, that we call templatized documents, are programmatically generated by populating fields in a visual template. Effective data extraction from these documents is crucial to supporting downstream analytical tasks. Current data extraction tools often struggle with complex document layouts, incur high latency and/or cost on large datasets, and often require significant human effort, when extracting tables or values given user-specified fields from documents. The key insight of our tool, TWIX, is to predict the underlying template used to create such documents, modeling the visual and structural commonalities across documents. Data extraction based on this predicted template provides a more principled, accurate, and efficient solution at a low cost. Comprehensive evaluations on 34 diverse real-world datasets show that uncovering the template is crucial for data extraction from templatized documents. TWIX achieves over 90% precision and recall on average, outperforming tools from industry: Textract and Azure Document Intelligence, and vision-based LLMs like GPT-4-Vision, by over 25% in precision and recall. TWIX scales easily to large datasets and is 734X faster and 5836X cheaper than vision-based LLMs for extracting data from a large document collection with 817 pages.
Authors: Takamasa Terada, Masahiro Toyoura
Abstract: Wearable electrocardiogram (ECG) measurement using dry electrodes has a problem with high-intensity noise distortion. Hence, a robust noise reduction method is required. However, overlapping frequency bands of ECG and noise make noise reduction difficult. Hence, it is necessary to provide a mechanism that changes the characteristics of the noise based on its intensity and type. This study proposes a convolutional neural network (CNN) model with an additional wavelet transform layer that extracts the specific frequency features in a clean ECG. Testing confirms that the proposed method effectively predicts accurate ECG behavior with reduced noise by accounting for all frequency domains. In an experiment, noisy signals in the signal-to-noise ratio (SNR) range of -10-10 are evaluated, demonstrating that the efficiency of the proposed method is higher when the SNR is small.
Authors: Michael Toker, Ido Galil, Hadas Orgad, Rinon Gal, Yoad Tewel, Gal Chechik, Yonatan Belinkov
Abstract: Text-to-image (T2I) diffusion models rely on encoded prompts to guide the image generation process. Typically, these prompts are extended to a fixed length by adding padding tokens before text encoding. Despite being a default practice, the influence of padding tokens on the image generation process has not been investigated. In this work, we conduct the first in-depth analysis of the role padding tokens play in T2I models. We develop two causal techniques to analyze how information is encoded in the representation of tokens across different components of the T2I pipeline. Using these techniques, we investigate when and how padding tokens impact the image generation process. Our findings reveal three distinct scenarios: padding tokens may affect the model's output during text encoding, during the diffusion process, or be effectively ignored. Moreover, we identify key relationships between these scenarios and the model's architecture (cross or self-attention) and its training process (frozen or trained text encoder). These insights contribute to a deeper understanding of the mechanisms of padding tokens, potentially informing future model design and training practices in T2I systems.
Authors: Du Chen, Liyi Chen, Zhengqiang Zhang, Lei Zhang
Abstract: Equipped with the continuous representation capability of Multi-Layer Perceptron (MLP), Implicit Neural Representation (INR) has been successfully employed for Arbitrary-scale Super-Resolution (ASR). However, the limited receptive field of the linear layers in MLP restricts the representation capability of INR, while it is computationally expensive to query the MLP numerous times to render each pixel. Recently, Gaussian Splatting (GS) has shown its advantages over INR in both visual quality and rendering speed in 3D tasks, which motivates us to explore whether GS can be employed for the ASR task. However, directly applying GS to ASR is exceptionally challenging because the original GS is an optimization-based method through overfitting each single scene, while in ASR we aim to learn a single model that can generalize to different images and scaling factors. We overcome these challenges by developing two novel techniques. Firstly, to generalize GS for ASR, we elaborately design an architecture to predict the corresponding image-conditioned Gaussians of the input low-resolution image in a feed-forward manner. Secondly, we implement an efficient differentiable 2D GPU/CUDA-based scale-aware rasterization to render super-resolved images by sampling discrete RGB values from the predicted contiguous Gaussians. Via end-to-end training, our optimized network, namely GSASR, can perform ASR for any image and unseen scaling factors. Extensive experiments validate the effectiveness of our proposed method. The project page can be found at \url{https://mt-cly.github.io/GSASR.github.io/}.
Authors: Raghav Singhal, Zachary Horvitz, Ryan Teehan, Mengye Ren, Zhou Yu, Kathleen McKeown, Rajesh Ranganath
Abstract: Diffusion models produce impressive results in modalities ranging from images and video to protein design and text. However, generating samples with user-specified properties remains a challenge. Recent research proposes fine-tuning models to maximize rewards that capture desired properties, but these methods require expensive training and are prone to mode collapse. In this work, we propose Feynman Kac (FK) steering, an inference-time framework for steering diffusion models with reward functions. FK steering works by sampling a system of multiple interacting diffusion processes, called particles, and resampling particles at intermediate steps based on scores computed using functions called potentials. Potentials are defined using rewards for intermediate states and are selected such that a high value indicates that the particle will yield a high-reward sample. We explore various choices of potentials, intermediate rewards, and samplers. We evaluate FK steering on text-to-image and text diffusion models. For steering text-to-image models with a human preference reward, we find that FK steering a 0.8B parameter model outperforms a 2.6B parameter fine-tuned model on prompt fidelity, with faster sampling and no training. For steering text diffusion models with rewards for text quality and specific text attributes, we find that FK steering generates lower perplexity, more linguistically acceptable outputs and enables gradient-free control of attributes like toxicity. Our results demonstrate that inference-time scaling and steering of diffusion models, even with off-the-shelf rewards, can provide significant sample quality gains and controllability benefits. Code is available at https://github.com/zacharyhorvitz/Fk-Diffusion-Steering .
URLs: https://github.com/zacharyhorvitz/Fk-Diffusion-Steering
Authors: Haojun Yu, Youcheng Li, Nan Zhang, Zihan Niu, Xuantong Gong, Yanwen Luo, Haotian Ye, Siyu He, Quanlin Wu, Wangyan Qin, Mengyuan Zhou, Jie Han, Jia Tao, Ziwei Zhao, Di Dai, Di He, Dong Wang, Binghui Tang, Ling Huo, James Zou, Qingli Zhu, Yong Wang, Liwei Wang
Abstract: Foundational models have emerged as powerful tools for addressing various tasks in clinical settings. However, their potential development to breast ultrasound analysis remains untapped. In this paper, we present BUSGen, the first foundational generative model specifically designed for breast ultrasound image analysis. Pretrained on over 3.5 million breast ultrasound images, BUSGen has acquired extensive knowledge of breast structures, pathological features, and clinical variations. With few-shot adaptation, BUSGen can generate repositories of realistic and informative task-specific data, facilitating the development of models for a wide range of downstream tasks. Extensive experiments highlight BUSGen's exceptional adaptability, significantly exceeding real-data-trained foundational models in breast cancer screening, diagnosis, and prognosis. In breast cancer early diagnosis, our approach outperformed all board-certified radiologists (n=9), achieving an average sensitivity improvement of 16.5% (P-value<0.0001). Additionally, we characterized the scaling effect of using generated data which was as effective as the collected real-world data for training diagnostic models. Moreover, extensive experiments demonstrated that our approach improved the generalization ability of downstream models. Importantly, BUSGen protected patient privacy by enabling fully de-identified data sharing, making progress forward in secure medical data utilization. An online demo of BUSGen is available at https://aibus.bio.
URLs: https://aibus.bio.
Authors: Jiayi Huang, Feiyun Zhao, Lieyang Chen
Abstract: This study proposes an advanced method for surface defect detection in printed circuit boards (PCBs) using an improved YOLOv11 model enhanced with a generative adversarial network (GAN). The approach focuses on identifying six common defect types: missing hole, rat bite, open circuit, short circuit, burr, and virtual welding. By employing GAN to generate synthetic defect images, the dataset is augmented with diverse and realistic patterns, improving the model's ability to generalize, particularly for complex and infrequent defects like burrs. The enhanced YOLOv11 model is evaluated on a PCB defect dataset, demonstrating significant improvements in accuracy, recall, and robustness, especially when dealing with defects in complex environments or small targets. This research contributes to the broader field of electronic design automation (EDA), where efficient defect detection is a crucial step in ensuring high-quality PCB manufacturing. By integrating advanced deep learning techniques, this approach enhances the automation and precision of defect detection, reducing reliance on manual inspection and accelerating design-to-production workflows. The findings underscore the importance of incorporating GAN-based data augmentation and optimized detection architectures in EDA processes, providing valuable insights for improving reliability and efficiency in PCB defect detection within industrial applications.
Authors: Shan Jiang, Zhenhua Han, Haisheng Tan, Xinyang Jiang, Yifan Yang, Xiaoxi Zhang, Hongqiu Ni, Yuqing Yang, Xiang-Yang Li
Abstract: Online Cloud gaming demands real-time, high-quality video transmission across variable wide-area networks (WANs). Neural-enhanced video transmission algorithms employing super-resolution (SR) for video quality enhancement have effectively challenged WAN environments. However, these SR-based methods require intensive fine-tuning for the whole video, making it infeasible in diverse online cloud gaming. To address this, we introduce River, a cloud gaming delivery framework designed based on the observation that video segment features in cloud gaming are typically repetitive and redundant. This permits a significant opportunity to reuse fine-tuned SR models, reducing the fine-tuning latency of minutes to query latency of milliseconds. To enable the idea, we design a practical system that addresses several challenges, such as model organization, online model scheduler, and transfer strategy. River first builds a content-aware encoder that fine-tunes SR models for diverse video segments and stores them in a lookup table. When delivering cloud gaming video streams online, River checks the video features and retrieves the most relevant SR models to enhance the frame quality. Meanwhile, if no existing SR model performs well enough for some video segments, River will further fine-tune new models and update the lookup table. Finally, to avoid the overhead of streaming model weight to the clients, River designs a prefetching strategy that predicts the models with the highest possibility of being retrieved. Our evaluation based on real video game streaming demonstrates River can reduce redundant training overhead by 44% and improve the Peak-Signal-to-Noise-Ratio by 1.81dB compared to the SOTA solutions. Practical deployment shows River meets real-time requirements, achieving approximately 720p 20fps on mobile devices.
Authors: Aparna Joshi, Kojo Adugyamfi, Jennifer Merickel, Pujitha Gunaratne, Anuj Sharma
Abstract: By 2030, the senior population aged 65 and older is expected to increase by over 50%, significantly raising the number of older drivers on the road. Drivers over 70 face higher crash death rates compared to those in their forties and fifties, underscoring the importance of developing more effective safety interventions for this demographic. Although the impact of aging on driving behavior has been studied, there is limited research on how these behaviors translate into real-world driving scenarios. This study addresses this need by leveraging Naturalistic Driving Data (NDD) to analyze driving performance measures - specifically, speed limit adherence on interstates and deceleration at stop intersections, both of which may be influenced by age-related declines. Using NDD, we developed Cumulative Distribution Functions (CDFs) to establish benchmarks for key driving behaviors among senior and young drivers. Our analysis, which included anomaly detection, benchmark comparisons, and accuracy evaluations, revealed significant differences in driving patterns primarily related to speed limit adherence at 75mph. While our approach shows promising potential for enhancing Advanced Driver Assistance Systems (ADAS) by providing tailored interventions based on age-specific adherence to speed limit driving patterns, we recognize the need for additional data to refine and validate metrics for other driving behaviors. By establishing precise benchmarks for various driving performance metrics, ADAS can effectively identify anomalies, such as abrupt deceleration, which may indicate impaired driving or other safety concerns. This study lays a strong foundation for future research aimed at improving safety interventions through detailed driving behavior analysis.
Authors: Evgeny Ugolkov, Xupeng He, Hyung Kwak, Hussein Hoteit
Abstract: We develop a procedure for substantially improving the quality of segmented 3D micro-Computed Tomography (micro-CT) images of rocks with a Machine Learning (ML) Generative Model. The proposed model enhances the resolution eightfold (8x) and addresses segmentation inaccuracies due to the overlapping X-ray attenuation in micro-CT measurement for different rock minerals and phases. The proposed generative model is a 3D Deep Convolutional Wasserstein Generative Adversarial Network with Gradient Penalty (3D DC WGAN-GP). The algorithm is trained on segmented 3D low-resolution micro-CT images and segmented unpaired complementary 2D high-resolution Laser Scanning Microscope (LSM) images. The algorithm was demonstrated on multiple samples of Berea sandstones. We achieved high-quality super-resolved 3D images with a resolution of 0.4375 micro-m/voxel and accurate segmentation for constituting minerals and pore space. The described procedure can significantly expand the modern capabilities of digital rock physics.
Authors: Vouk Praun-Petrovic, Aadhvika Koundinya, Lavanya Prahallad
Abstract: Generative AI, powered by large language models (LLMs), has revolutionized applications across text, audio, images, and video. This study focuses on developing and evaluating encoder-decoder architectures for the American Sign Language (ASL) image dataset, consisting of 87,000 images across 29 hand sign classes. Three approaches were compared: Feedforward Autoencoders, Convolutional Autoencoders, and Diffusion Autoencoders. The Diffusion Autoencoder outperformed the others, achieving the lowest mean squared error (MSE) and highest Mean Opinion Score (MOS) due to its probabilistic noise modeling and iterative denoising capabilities. The Convolutional Autoencoder demonstrated effective spatial feature extraction but lacked the robustness of the diffusion process, while the Feedforward Autoencoder served as a baseline with limitations in handling complex image data. Objective and subjective evaluations confirmed the superiority of the Diffusion Autoencoder for high-fidelity image reconstruction, emphasizing its potential in multimodal AI applications such as sign language recognition and generation. This work provides critical insights into designing robust encoder-decoder systems to advance multimodal AI capabilities.
Authors: Binyu Zhang, Shichao Li, Junpeng Jian, Zhu Meng, Limei Guo, Zhicheng Zhao
Abstract: Prognostic task is of great importance as it closely related to the survival analysis of patients, the optimization of treatment plans and the allocation of resources. The existing prognostic models have shown promising results on specific datasets, but there are limitations in two aspects. On the one hand, they merely explore certain types of modal data, such as patient histopathology WSI and gene expression analysis. On the other hand, they adopt the per-cancer-per-model paradigm, which means the trained models can only predict the prognostic effect of a single type of cancer, resulting in weak generalization ability. In this paper, a deep-learning based model, named UMPSNet, is proposed. Specifically, to comprehensively understand the condition of patients, in addition to constructing encoders for histopathology images and genomic expression profiles respectively, UMPSNet further integrates four types of important meta data (demographic information, cancer type information, treatment protocols, and diagnosis results) into text templates, and then introduces a text encoder to extract textual features. In addition, the optimal transport OT-based attention mechanism is utilized to align and fuse features of different modalities. Furthermore, a guided soft mixture of experts (GMoE) mechanism is introduced to effectively address the issue of distribution differences among multiple cancer datasets. By incorporating the multi-modality of patient data and joint training, UMPSNet outperforms all SOTA approaches, and moreover, it demonstrates the effectiveness and generalization ability of the proposed learning paradigm of a single model for multiple cancer types. The code of UMPSNet is available at https://github.com/binging512/UMPSNet.
Authors: Zong Ke, Shicheng Zhou, Yining Zhou, Chia Hong Chang, Rong Zhang
Abstract: This study explores the use of Generative Adversarial Networks (GANs) to detect AI deepfakes and fraudulent activities in online payment systems. With the growing prevalence of deepfake technology, which can manipulate facial features in images and videos, the potential for fraud in online transactions has escalated. Traditional security systems struggle to identify these sophisticated forms of fraud. This research proposes a novel GAN-based model that enhances online payment security by identifying subtle manipulations in payment images. The model is trained on a dataset consisting of real-world online payment images and deepfake images generated using advanced GAN architectures, such as StyleGAN and DeepFake. The results demonstrate that the proposed model can accurately distinguish between legitimate transactions and deepfakes, achieving a high detection rate above 95%. This approach significantly improves the robustness of payment systems against AI-driven fraud. The paper contributes to the growing field of digital security, offering insights into the application of GANs for fraud detection in financial services. Keywords- Payment Security, Image Recognition, Generative Adversarial Networks, AI Deepfake, Fraudulent Activities
Authors: Xiaoxian Yang, Qi Wang, Kaiqi Zhang, Ke Wei, Jun Lyu, Lingchao Chen
Abstract: Ultrasound imaging frequently encounters challenges, such as those related to elevated noise levels, diminished spatiotemporal resolution, and the complexity of anatomical structures. These factors significantly hinder the model's ability to accurately capture and analyze structural relationships and dynamic patterns across various regions of the heart. Mamba, an emerging model, is one of the most cutting-edge approaches that is widely applied to diverse vision and language tasks. To this end, this paper introduces a U-shaped deep learning model incorporating a large-window Mamba scale (LMS) module and a hierarchical feature fusion approach for echocardiographic segmentation. First, a cascaded residual block serves as an encoder and is employed to incrementally extract multiscale detailed features. Second, a large-window multiscale mamba module is integrated into the decoder to capture global dependencies across regions and enhance the segmentation capability for complex anatomical structures. Furthermore, our model introduces auxiliary losses at each decoder layer and employs a dual attention mechanism to fuse multilayer features both spatially and across channels. This approach enhances segmentation performance and accuracy in delineating complex anatomical structures. Finally, the experimental results using the EchoNet-Dynamic and CAMUS datasets demonstrate that the model outperforms other methods in terms of both accuracy and robustness. For the segmentation of the left ventricular endocardium (${LV}_{endo}$), the model achieved optimal values of 95.01 and 93.36, respectively, while for the left ventricular epicardium (${LV}_{epi}$), values of 87.35 and 87.80, respectively, were achieved. This represents an improvement ranging between 0.54 and 1.11 compared with the best-performing model.
Authors: Duc Anh Vu, Anh Tuan Tran, Cong Tran, Cuong Pham
Abstract: Backdoor attacks have become a critical threat to deep neural networks (DNNs), drawing many research interests. However, most of the studied attacks employ a single type of trigger. Consequently, proposed backdoor defenders often rely on the assumption that triggers would appear in a unified way. In this paper, we show that this naive assumption can create a loophole, allowing more sophisticated backdoor attacks to bypass. We design a novel backdoor attack mechanism that incorporates multiple types of backdoor triggers, focusing on stealthiness and effectiveness. Our journey begins with the intriguing observation that the performance of a backdoor attack in deep learning models, as well as its detectability and removability, are all proportional to the magnitude of the trigger. Based on this correlation, we propose reducing the magnitude of each trigger type and combining them to achieve a strong backdoor relying on the combined trigger while still staying safely under the radar of defenders. Extensive experiments on three standard datasets demonstrate that our method can achieve high attack success rates (ASRs) while consistently bypassing state-of-the-art defenses.
Authors: Aryan Chaudhari, Ankush Singh, Sanchi Gajbhiye, Pratham Agrawal
Abstract: In this paper we discuss lung cancer detection using hybrid model of Convolutional-Neural-Networks (CNNs) and Support-Vector-Machines-(SVMs) in order to gain early detection of tumors, benign or malignant. The work uses this hybrid model by training upon the Computed Tomography scans (CT scans) as dataset. Using deep learning for detecting lung cancer early is a cutting-edge method.
Authors: Mohamed Ala Yahyaoui, Mouaad Oujabour, Leila Ben Letaifa, Amine Bohi
Abstract: The integration of dialogue interfaces in mobile devices has become ubiquitous, providing a wide array of services. As technology progresses, humanoid robots designed with human-like features to interact effectively with people are gaining prominence, and the use of advanced human-robot dialogue interfaces is continually expanding. In this context, emotion recognition plays a crucial role in enhancing human-robot interaction by enabling robots to understand human intentions. This research proposes a facial emotion detection interface integrated into a mobile humanoid robot, capable of displaying real-time emotions from multiple individuals on a user interface. To this end, various deep neural network models for facial expression recognition were developed and evaluated under consistent computer-based conditions, yielding promising results. Afterwards, a trade-off between accuracy and memory footprint was carefully considered to effectively implement this application on a mobile humanoid robot.
Authors: Mathias Micheelsen Lowes, Jonas Jalili Pedersen, Bj{\o}rn S. Hansen, Klaus Fuglsang Kofoed, Maxime Sermesant, Rasmus R. Paulsen
Abstract: Understanding the movement of the left ventricle myocardium (LVmyo) during the cardiac cycle is essential for assessing cardiac function. One way to model this movement is through a series of deformable image registrations (DIRs) of the LVmyo. Traditional deep learning methods for DIRs, such as those based on convolutional neural networks, often require substantial memory and computational resources. In contrast, implicit neural representations (INRs) offer an efficient approach by operating on any number of continuous points. This study extends the use of INRs for DIR to cardiac computed tomography (CT), focusing on LVmyo registration. To enhance the precision of the registration around the LVmyo, we incorporate the signed distance field of the LVmyo with the Hounsfield Unit values from the CT frames. This guides the registration of the LVmyo, while keeping the tissue information from the CT frames. Our framework demonstrates high registration accuracy and provides a robust method for temporal registration that facilitates further analysis of LVmyo motion.
Authors: Ping Guo, Cheng Gong, Xi Lin, Fei Liu, Zhichao Lu, Qingfu Zhang, Zhenkun Wang
Abstract: Crafting adversarial examples is crucial for evaluating and enhancing the robustness of Deep Neural Networks (DNNs), presenting a challenge equivalent to maximizing a non-differentiable 0-1 loss function. However, existing single objective methods, namely adversarial attacks focus on a surrogate loss function, do not fully harness the benefits of engaging multiple loss functions, as a result of insufficient understanding of their synergistic and conflicting nature. To overcome these limitations, we propose the Multi-Objective Set-based Attack (MOS Attack), a novel adversarial attack framework leveraging multiple loss functions and automatically uncovering their interrelations. The MOS Attack adopts a set-based multi-objective optimization strategy, enabling the incorporation of numerous loss functions without additional parameters. It also automatically mines synergistic patterns among various losses, facilitating the generation of potent adversarial attacks with fewer objectives. Extensive experiments have shown that our MOS Attack outperforms single-objective attacks. Furthermore, by harnessing the identified synergistic patterns, MOS Attack continues to show superior results with a reduced number of loss functions.
Authors: Tita Enstad, Trond Trosterud, Marie Iversdatter R{\o}sok, Yngvil Beyer, Marie Roald
Abstract: Optical Character Recognition (OCR) is crucial to the National Library of Norway's (NLN) digitisation process as it converts scanned documents into machine-readable text. However, for the S\'ami documents in NLN's collection, the OCR accuracy is insufficient. Given that OCR quality affects downstream processes, evaluating and improving OCR for text written in S\'ami languages is necessary to make these resources accessible. To address this need, this work fine-tunes and evaluates three established OCR approaches, Transkribus, Tesseract and TrOCR, for transcribing S\'ami texts from NLN's collection. Our results show that Transkribus and TrOCR outperform Tesseract on this task, while Tesseract achieves superior performance on an out-of-domain dataset. Furthermore, we show that fine-tuning pre-trained models and supplementing manual annotations with machine annotations and synthetic text images can yield accurate OCR for S\'ami languages, even with a moderate amount of manually annotated data.
Authors: Manuel Eberhardinger, Patrick Takenaka, Daniel Grie{\ss}haber, Johannes Maucher
Abstract: The steadily increasing utilization of data-driven methods and approaches in areas that handle sensitive personal information such as in law enforcement mandates an ever increasing effort in these institutions to comply with data protection guidelines. In this work, we present a system for automatically anonymizing images of scanned documents, reducing manual effort while ensuring data protection compliance. Our method considers the viability of further forensic processing after anonymization by minimizing automatically redacted areas by combining automatic detection of sensitive regions with knowledge from a manually anonymized reference document. Using a self-supervised image model for instance retrieval of the reference document, our approach requires only one anonymized example to efficiently redact all documents of the same type, significantly reducing processing time. We show that our approach outperforms both a purely automatic redaction system and also a naive copy-paste scheme of the reference anonymization to other documents on a hand-crafted dataset of ground truth redactions.
Authors: Ting-Yu Dai, Hayato Ushijima-Mwesigwa
Abstract: A recent report from the World Meteorological Organization (WMO) highlights that water-related disasters have caused the highest human losses among natural disasters over the past 50 years, with over 91\% of deaths occurring in low-income countries. This disparity is largely due to the lack of adequate ground monitoring stations, such as weather surveillance radars (WSR), which are expensive to install. For example, while the US and Europe combined possess over 600 WSRs, Africa, despite having almost one and half times their landmass, has fewer than 40. To address this issue, satellite-based observations offer a global, near-real-time monitoring solution. However, they face several challenges like accuracy, bias, and low spatial resolution. This study leverages the power of diffusion models and residual learning to address these limitations in a unified framework. We introduce the first diffusion model for correcting the inconsistency between different precipitation products. Our method demonstrates the effectiveness in downscaling satellite precipitation estimates from 10 km to 1 km resolution. Extensive experiments conducted in the Seattle region demonstrate significant improvements in accuracy, bias reduction, and spatial detail. Importantly, our approach achieves these results using only precipitation data, showcasing the potential of a purely computer vision-based approach for enhancing satellite precipitation products and paving the way for further advancements in this domain.
Authors: Lewis A G Stuart, Michael P Pound
Abstract: 3D Gaussian Splatting (3DGS) excels at producing highly detailed 3D reconstructions, but these scenes often require specialised renderers for effective visualisation. In contrast, point clouds are a widely used 3D representation and are compatible with most popular 3D processing software, yet converting 3DGS scenes into point clouds is a complex challenge. In this work we introduce 3DGS-to-PC, a flexible and highly customisable framework that is capable of transforming 3DGS scenes into dense, high-accuracy point clouds. We sample points probabilistically from each Gaussian as a 3D density function. We additionally threshold new points using the Mahalanobis distance to the Gaussian centre, preventing extreme outliers. The result is a point cloud that closely represents the shape encoded into the 3D Gaussian scene. Individual Gaussians use spherical harmonics to adapt colours depending on view, and each point may contribute only subtle colour hints to the resulting rendered scene. To avoid spurious or incorrect colours that do not fit with the final point cloud, we recalculate Gaussian colours via a customised image rendering approach, assigning each Gaussian the colour of the pixel to which it contributes most across all views. 3DGS-to-PC also supports mesh generation through Poisson Surface Reconstruction, applied to points sampled from predicted surface Gaussians. This allows coloured meshes to be generated from 3DGS scenes without the need for re-training. This package is highly customisable and capability of simple integration into existing 3DGS pipelines. 3DGS-to-PC provides a powerful tool for converting 3DGS data into point cloud and surface-based formats.
Authors: Chengzu Li, Wenshan Wu, Huanyu Zhang, Yan Xia, Shaoguang Mao, Li Dong, Ivan Vuli\'c, Furu Wei
Abstract: Chain-of-Thought (CoT) prompting has proven highly effective for enhancing complex reasoning in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs). Yet, it struggles in complex spatial reasoning tasks. Nonetheless, human cognition extends beyond language alone, enabling the remarkable capability to think in both words and images. Inspired by this mechanism, we propose a new reasoning paradigm, Multimodal Visualization-of-Thought (MVoT). It enables visual thinking in MLLMs by generating image visualizations of their reasoning traces. To ensure high-quality visualization, we introduce token discrepancy loss into autoregressive MLLMs. This innovation significantly improves both visual coherence and fidelity. We validate this approach through several dynamic spatial reasoning tasks. Experimental results reveal that MVoT demonstrates competitive performance across tasks. Moreover, it exhibits robust and reliable improvements in the most challenging scenarios where CoT fails. Ultimately, MVoT establishes new possibilities for complex reasoning tasks where visual thinking can effectively complement verbal reasoning.
Authors: Vihaan Misra, Peter Schaldenbrand, Jean Oh
Abstract: If a picture paints a thousand words, sound may voice a million. While recent robotic painting and image synthesis methods have achieved progress in generating visuals from text inputs, the translation of sound into images is vastly unexplored. Generally, sound-based interfaces and sonic interactions have the potential to expand accessibility and control for the user and provide a means to convey complex emotions and the dynamic aspects of the real world. In this paper, we propose an approach for using sound and speech to guide a robotic painting process, known here as robot synesthesia. For general sound, we encode the simulated paintings and input sounds into the same latent space. For speech, we decouple speech into its transcribed text and the tone of the speech. Whereas we use the text to control the content, we estimate the emotions from the tone to guide the mood of the painting. Our approach has been fully integrated with FRIDA, a robotic painting framework, adding sound and speech to FRIDA's existing input modalities, such as text and style. In two surveys, participants were able to correctly guess the emotion or natural sound used to generate a given painting more than twice as likely as random chance. On our sound-guided image manipulation and music-guided paintings, we discuss the results qualitatively.
Authors: Guolei Sun, Xiaogang Cheng, Zhaochong An, Xiaokang Wang, Yun Liu, Deng-Ping Fan, Ming-Ming Cheng, Luc Van Gool
Abstract: Recently, indiscernible/camouflaged scene understanding has attracted lots of research attention in the vision community. We further advance the frontier of this field by systematically studying a new challenge named indiscernible object counting (IOC), the goal of which is to count objects that are blended with respect to their surroundings. Due to a lack of appropriate IOC datasets, we present a large-scale dataset IOCfish5K which contains a total of 5,637 high-resolution images and 659,024 annotated center points. Our dataset consists of a large number of indiscernible objects (mainly fish) in underwater scenes, making the annotation process all the more challenging. IOCfish5K is superior to existing datasets with indiscernible scenes because of its larger scale, higher image resolutions, more annotations, and denser scenes. All these aspects make it the most challenging dataset for IOC so far, supporting progress in this area. Benefiting from the recent advancements of depth estimation foundation models, we construct high-quality depth maps for IOCfish5K by generating pseudo labels using the Depth Anything V2 model. The RGB-D version of IOCfish5K is named IOCfish5K-D. For benchmarking purposes on IOCfish5K, we select 14 mainstream methods for object counting and carefully evaluate them. For multimodal IOCfish5K-D, we evaluate other 4 popular multimodal counting methods. Furthermore, we propose IOCFormer, a new strong baseline that combines density and regression branches in a unified framework and can effectively tackle object counting under concealed scenes. We also propose IOCFormer-D to enable the effective usage of depth modality in helping detect and count objects hidden in their environments. Experiments show that IOCFormer and IOCFormer-D achieve state-of-the-art scores on IOCfish5K and IOCfish5K-D, respectively.
Authors: Wisdom Oluchi Ikezogwo, Mehmet Saygin Seyfioglu, Fatemeh Ghezloo, Dylan Stefan Chan Geva, Fatwir Sheikh Mohammed, Pavan Kumar Anand, Ranjay Krishna, Linda Shapiro
Abstract: Recent accelerations in multi-modal applications have been made possible with the plethora of image and text data available online. However, the scarcity of analogous data in the medical field, specifically in histopathology, has slowed comparable progress. To enable similar representation learning for histopathology, we turn to YouTube, an untapped resource of videos, offering $1,087$ hours of valuable educational histopathology videos from expert clinicians. From YouTube, we curate QUILT: a large-scale vision-language dataset consisting of $802, 144$ image and text pairs. QUILT was automatically curated using a mixture of models, including large language models, handcrafted algorithms, human knowledge databases, and automatic speech recognition. In comparison, the most comprehensive datasets curated for histopathology amass only around $200$K samples. We combine QUILT with datasets from other sources, including Twitter, research papers, and the internet in general, to create an even larger dataset: QUILT-1M, with $1$M paired image-text samples, marking it as the largest vision-language histopathology dataset to date. We demonstrate the value of QUILT-1M by fine-tuning a pre-trained CLIP model. Our model outperforms state-of-the-art models on both zero-shot and linear probing tasks for classifying new histopathology images across $13$ diverse patch-level datasets of $8$ different sub-pathologies and cross-modal retrieval tasks.
Authors: Haodong He, Jian Ding, Bowen Xu, Gui-Song Xia
Abstract: The robustness of object detection models is a major concern when applied to real-world scenarios. The performance of most models tends to degrade when confronted with images affected by corruptions, since they are usually trained and evaluated on clean datasets. While numerous studies have explored the robustness of object detection models on natural images, there is a paucity of research focused on models applied to aerial images, which feature complex backgrounds, substantial variations in scales, and orientations of objects. This paper addresses the challenge of assessing the robustness of object detection models on aerial images, with a specific emphasis on scenarios where images are affected by clouds. In this study, we introduce two novel benchmarks based on DOTA-v1.0. The first benchmark encompasses 19 prevalent corruptions, while the second focuses on the cloud-corrupted condition-a phenomenon uncommon in natural images yet frequent in aerial photography. We systematically evaluate the robustness of mainstream object detection models and perform necessary ablation experiments. Through our investigations, we find that rotation-invariant modeling and enhanced backbone architectures can improve the robustness of models. Furthermore, increasing the capacity of Transformer-based backbones can strengthen their robustness. The benchmarks we propose and our comprehensive experimental analyses can facilitate research on robust object detection on aerial images. The codes and datasets are available at: https://github.com/hehaodong530/DOTA-C.
Authors: Mehmet Saygin Seyfioglu, Wisdom O. Ikezogwo, Fatemeh Ghezloo, Ranjay Krishna, Linda Shapiro
Abstract: Diagnosis in histopathology requires a global whole slide images (WSIs) analysis, requiring pathologists to compound evidence from different WSI patches. The gigapixel scale of WSIs poses a challenge for histopathology multi-modal models. Training multi-model models for histopathology requires instruction tuning datasets, which currently contain information for individual image patches, without a spatial grounding of the concepts within each patch and without a wider view of the WSI. Therefore, they lack sufficient diagnostic capacity for histopathology. To bridge this gap, we introduce Quilt-Instruct, a large-scale dataset of 107,131 histopathology-specific instruction question/answer pairs, grounded within diagnostically relevant image patches that make up the WSI. Our dataset is collected by leveraging educational histopathology videos from YouTube, which provides spatial localization of narrations by automatically extracting the narrators' cursor positions. Quilt-Instruct supports contextual reasoning by extracting diagnosis and supporting facts from the entire WSI. Using Quilt-Instruct, we train Quilt-LLaVA, which can reason beyond the given single image patch, enabling diagnostic reasoning across patches. To evaluate Quilt-LLaVA, we propose a comprehensive evaluation dataset created from 985 images and 1283 human-generated question-answers. We also thoroughly evaluate Quilt-LLaVA using public histopathology datasets, where Quilt-LLaVA significantly outperforms SOTA by over 10% on relative GPT-4 score and 4% and 9% on open and closed set VQA. Our code, data, and model are publicly accessible at quilt-llava.github.io.
Authors: Fernando P\'erez-Garc\'ia, Harshita Sharma, Sam Bond-Taylor, Kenza Bouzid, Valentina Salvatelli, Maximilian Ilse, Shruthi Bannur, Daniel C. Castro, Anton Schwaighofer, Matthew P. Lungren, Maria Wetscherek, Noel Codella, Stephanie L. Hyland, Javier Alvarez-Valle, Ozan Oktay
Abstract: Language-supervised pre-training has proven to be a valuable method for extracting semantically meaningful features from images, serving as a foundational element in multimodal systems within the computer vision and medical imaging domains. However, the computed features are limited by the information contained in the text, which is particularly problematic in medical imaging, where the findings described by radiologists focus on specific observations. This challenge is compounded by the scarcity of paired imaging-text data due to concerns over leakage of personal health information. In this work, we fundamentally challenge the prevailing reliance on language supervision for learning general-purpose biomedical imaging encoders. We introduce RAD-DINO, a biomedical image encoder pre-trained solely on unimodal biomedical imaging data that obtains similar or greater performance than state-of-the-art biomedical language-supervised models on a diverse range of benchmarks. Specifically, the quality of learned representations is evaluated on standard imaging tasks (classification and semantic segmentation), and a vision-language alignment task (text report generation from images). To further demonstrate the drawback of language supervision, we show that features from RAD-DINO correlate with other medical records (e.g., sex or age) better than language-supervised models, which are generally not mentioned in radiology reports. Finally, we conduct a series of ablations determining the factors in RAD-DINO's performance; notably, we observe that RAD-DINO's downstream performance scales well with the quantity and diversity of training data, demonstrating that image-only supervision is a scalable approach for training a foundational biomedical image encoder. Model weights of RAD-DINO trained on publicly available datasets are available at https://huggingface.co/microsoft/rad-dino.
Authors: Ci-Siang Lin, Chien-Yi Wang, Yu-Chiang Frank Wang, Min-Hung Chen
Abstract: Weakly-Supervised Semantic Segmentation (WSSS) aims to train segmentation models using image data with only image-level supervision. Since precise pixel-level annotations are not accessible, existing methods typically focus on producing pseudo masks for training segmentation models by refining CAM-like heatmaps. However, the produced heatmaps may capture only the discriminative image regions of object categories or the associated co-occurring backgrounds. To address the issues, we propose a Semantic Prompt Learning for WSSS (SemPLeS) framework, which learns to effectively prompt the CLIP latent space to enhance the semantic alignment between the segmented regions and the target object categories. More specifically, we propose Contrastive Prompt Learning and Prompt-guided Semantic Refinement to learn the prompts that adequately describe and suppress the co-occurring backgrounds associated with each object category. In this way, SemPLeS can perform better semantic alignment between object regions and class labels, resulting in desired pseudo masks for training segmentation models. The proposed SemPLeS framework achieves competitive performance on standard WSSS benchmarks, PASCAL VOC 2012 and MS COCO 2014, and shows compatibility with other WSSS methods. Code: https://github.com/NVlabs/SemPLeS.
Authors: Renqiu Xia, Bo Zhang, Hancheng Ye, Xiangchao Yan, Qi Liu, Hongbin Zhou, Zijun Chen, Min Dou, Botian Shi, Junchi Yan, Yu Qiao
Abstract: Recently, many versatile Multi-modal Large Language Models (MLLMs) have emerged continuously. However, their capacity to query information depicted in visual charts and engage in reasoning based on the queried contents remains under-explored. In this paper, to comprehensively and rigorously benchmark the ability of the off-the-shelf MLLMs in the chart domain, we construct ChartX, a multi-modal evaluation set covering 18 chart types, 7 chart tasks, 22 disciplinary topics, and high-quality chart data. Besides, we develop ChartVLM to offer a new perspective on handling multi-modal tasks that strongly depend on interpretable patterns, such as reasoning tasks in the field of charts or geometric images. We evaluate the chart-related ability of mainstream MLLMs and our ChartVLM on the proposed ChartX evaluation set. Extensive experiments demonstrate that ChartVLM surpasses both versatile and chart-related large models, achieving results comparable to GPT-4V. We believe that our study can pave the way for further exploration in creating a more comprehensive chart evaluation set and developing more interpretable multi-modal models. Both ChartX and ChartVLM are available at: https://github.com/Alpha-Innovator/ChartVLM
Authors: Junbiao Pang, Tianyang Cai, Baochang Zhang, Jiaqi Wu
Abstract: Although existing Quantization-Aware Training (QAT) methods intensively depend on knowledge distillation to guarantee performance, QAT still suffers from severe performance drop. The experiments have shown that vanilla quantization is sensitive to the perturbation from both the input and weights. Therefore, we assume that the generalization ability of QAT is predominantly caused by both the intrinsic instability (training time) and the limited generalization ability (testing time). In this paper, we address both issues from a new perspective by leveraging Consistency Regularization (CR) to improve the generalization ability of QAT. Empirical results and theoretical analysis verify that CR would bring a good generalization ability to different network architectures and various QAT methods. Extensive experiments demonstrate that our approach significantly outperforms current state-of-the-art QAT methods and even the FP counterparts. On CIFAR-10, the proposed method improves by 3.79% compared to the baseline method using ResNet18, and improves by 3.84% compared to the baseline method using the lightweight model MobileNet.
Authors: Daniel Schug, Tyler J. Kovach, M. A. Wolfe, Jared Benson, Sanghyeok Park, J. P. Dodson, J. Corrigan, M. A. Eriksson, Justyna P. Zwolak
Abstract: The rapid development of quantum dot (QD) devices for quantum computing has necessitated more efficient and automated methods for device characterization and tuning. This work demonstrates the feasibility and advantages of applying explainable machine learning techniques to the analysis of quantum dot measurements, paving the way for further advances in automated and transparent QD device tuning. Many of the measurements acquired during the tuning process come in the form of images that need to be properly analyzed to guide the subsequent tuning steps. By design, features present in such images capture certain behaviors or states of the measured QD devices. When considered carefully, such features can aid the control and calibration of QD devices. An important example of such images are so-called $\textit{triangle plots}$, which visually represent current flow and reveal characteristics important for QD device calibration. While image-based classification tools, such as convolutional neural networks (CNNs), can be used to verify whether a given measurement is $\textit{good}$ and thus warrants the initiation of the next phase of tuning, they do not provide any insights into how the device should be adjusted in the case of $\textit{bad}$ images. This is because CNNs sacrifice prediction and model intelligibility for high accuracy. To ameliorate this trade-off, a recent study introduced an image vectorization approach that relies on the Gabor wavelet transform (Schug $\textit{et al.}$ 2024 $\textit{Proc. XAI4Sci: Explainable Machine Learning for Sciences Workshop (AAAI 2024) (Vancouver, Canada)}$ pp 1-6). Here we propose an alternative vectorization method that involves mathematical modeling of synthetic triangles to mimic the experimental data. Using explainable boosting machines, we show that this new method offers superior explainability of model prediction without sacrificing accuracy.
Authors: Zunnan Xu, Yukang Lin, Haonan Han, Sicheng Yang, Ronghui Li, Yachao Zhang, Xiu Li
Abstract: Gesture synthesis is a vital realm of human-computer interaction, with wide-ranging applications across various fields like film, robotics, and virtual reality. Recent advancements have utilized the diffusion model and attention mechanisms to improve gesture synthesis. However, due to the high computational complexity of these techniques, generating long and diverse sequences with low latency remains a challenge. We explore the potential of state space models (SSMs) to address the challenge, implementing a two-stage modeling strategy with discrete motion priors to enhance the quality of gestures. Leveraging the foundational Mamba block, we introduce MambaTalk, enhancing gesture diversity and rhythm through multimodal integration. Extensive experiments demonstrate that our method matches or exceeds the performance of state-of-the-art models.
Authors: Yifan Li, Hangyu Guo, Kun Zhou, Wayne Xin Zhao, Ji-Rong Wen
Abstract: In this paper, we study the harmlessness alignment problem of multimodal large language models (MLLMs). We conduct a systematic empirical analysis of the harmlessness performance of representative MLLMs and reveal that the image input poses the alignment vulnerability of MLLMs. Inspired by this, we propose a novel jailbreak method named HADES, which hides and amplifies the harmfulness of the malicious intent within the text input, using meticulously crafted images. Experimental results show that HADES can effectively jailbreak existing MLLMs, which achieves an average Attack Success Rate (ASR) of 90.26% for LLaVA-1.5 and 71.60% for Gemini Pro Vision. Our code and data are available at https://github.com/RUCAIBox/HADES.
Authors: Thomas Melistas, Nikos Spyrou, Nefeli Gkouti, Pedro Sanchez, Athanasios Vlontzos, Yannis Panagakis, Giorgos Papanastasiou, Sotirios A. Tsaftaris
Abstract: Generative AI has revolutionised visual content editing, empowering users to effortlessly modify images and videos. However, not all edits are equal. To perform realistic edits in domains such as natural image or medical imaging, modifications must respect causal relationships inherent to the data generation process. Such image editing falls into the counterfactual image generation regime. Evaluating counterfactual image generation is substantially complex: not only it lacks observable ground truths, but also requires adherence to causal constraints. Although several counterfactual image generation methods and evaluation metrics exist, a comprehensive comparison within a unified setting is lacking. We present a comparison framework to thoroughly benchmark counterfactual image generation methods. We integrate all models that have been used for the task at hand and expand them to novel datasets and causal graphs, demonstrating the superiority of Hierarchical VAEs across most datasets and metrics. Our framework is implemented in a user-friendly Python package that can be extended to incorporate additional SCMs, causal methods, generative models, and datasets for the community to build on. Code: https://github.com/gulnazaki/counterfactual-benchmark.
URLs: https://github.com/gulnazaki/counterfactual-benchmark.
Authors: Yehui Shen, Mingmin Liu, Huimin Lu, Xieyuanli Chen
Abstract: Visual place recognition (VPR) plays a pivotal role in autonomous exploration and navigation of mobile robots within complex outdoor environments. While cost-effective and easily deployed, camera sensors are sensitive to lighting and weather changes, and even slight image alterations can greatly affect VPR efficiency and precision. Existing methods overcome this by exploiting powerful yet large networks, leading to significant consumption of computational resources. In this paper, we propose a high-performance teacher and lightweight student distillation framework called TSCM. It exploits our devised cross-metric knowledge distillation to narrow the performance gap between the teacher and student models, maintaining superior performance while enabling minimal computational load during deployment. We conduct comprehensive evaluations on large-scale datasets, namely Pittsburgh30k and Pittsburgh250k. Experimental results demonstrate the superiority of our method over baseline models in terms of recognition accuracy and model parameter efficiency. Moreover, our ablation studies show that the proposed knowledge distillation technique surpasses other counterparts. The code of our method has been released at https://github.com/nubot-nudt/TSCM.
Authors: Ayumu Saito, Prachi Kudeshia, Jiju Poovvancheri
Abstract: Recent advancements in self-supervised learning in the point cloud domain have demonstrated significant potential. However, these methods often suffer from drawbacks, including lengthy pre-training time, the necessity of reconstruction in the input space, or the necessity of additional modalities. In order to address these issues, we introduce Point-JEPA, a joint embedding predictive architecture designed specifically for point cloud data. To this end, we introduce a sequencer that orders point cloud patch embeddings to efficiently compute and utilize their proximity based on the indices during target and context selection. The sequencer also allows shared computations of the patch embeddings' proximity between context and target selection, further improving the efficiency. Experimentally, our method achieves competitive results with state-of-the-art methods while avoiding the reconstruction in the input space or additional modality.
Authors: Da Li, Guoqiang Zhao, Houjun Sun, Jiacheng Bao
Abstract: Multi-baseline Synthetic Aperture Radar (SAR) three-dimensional (3D) tomography is a crucial remote sensing technique that provides 3D resolution unavailable in conventional SAR imaging. However, achieving high-quality imaging typically requires multi-angle or full-aperture data, resulting in significant imaging costs. Recent advancements in sparse 3D SAR, which rely on data from limited apertures, have gained attention as a cost-effective alternative. Notably, deep learning techniques have markedly enhanced the imaging quality of sparse 3D SAR. Despite these advancements, existing methods primarily depend on high-resolution radar images for supervising the training of deep neural networks (DNNs). This exclusive dependence on single-modal data prevents the introduction of complementary information from other data sources, limiting further improvements in imaging performance. In this paper, we introduce a Cross-Modal 3D-SAR Reconstruction Network (CMAR-Net) to enhance 3D SAR imaging by integrating heterogeneous information. Leveraging cross-modal supervision from 2D optical images and error transfer guaranteed by differentiable rendering, CMAR-Net achieves efficient training and reconstructs highly sparse multi-baseline SAR data into visually structured and accurate 3D images, particularly for vehicle targets. Extensive experiments on simulated and real-world datasets demonstrate that CMAR-Net significantly outperforms SOTA sparse reconstruction algorithms based on compressed sensing (CS) and deep learning (DL). Furthermore, our method eliminates the need for time-consuming full-aperture data preprocessing and relies solely on computer-rendered optical images, significantly reducing dataset construction costs. This work highlights the potential of deep learning for multi-baseline SAR 3D imaging and introduces a novel framework for radar imaging research through cross-modal learning.
Authors: Yirui Chen, Xudong Huang, Quan Zhang, Wei Li, Mingjian Zhu, Qiangyu Yan, Simiao Li, Hanting Chen, Hailin Hu, Jie Yang, Wei Liu, Jie Hu
Abstract: The extraordinary ability of generative models emerges as a new trend in image editing and generating realistic images, posing a serious threat to the trustworthiness of multimedia data and driving the research of image manipulation detection and location (IMDL). However, the lack of a large-scale data foundation makes the IMDL task unattainable. In this paper, we build a local manipulation data generation pipeline that integrates the powerful capabilities of SAM, LLM, and generative models. Upon this basis, we propose the GIM dataset, which has the following advantages: 1) Large scale, GIM includes over one million pairs of AI-manipulated images and real images. 2) Rich image content, GIM encompasses a broad range of image classes. 3) Diverse generative manipulation, the images are manipulated images with state-of-the-art generators and various manipulation tasks. The aforementioned advantages allow for a more comprehensive evaluation of IMDL methods, extending their applicability to diverse images. We introduce the GIM benchmark with two settings to evaluate existing IMDL methods. In addition, we propose a novel IMDL framework, termed GIMFormer, which consists of a ShadowTracer, Frequency-Spatial block (FSB), and a Multi-Window Anomalous Modeling (MWAM) module. Extensive experiments on the GIM demonstrate that GIMFormer surpasses the previous state-of-the-art approach on two different benchmarks.
Authors: Zhuoyuan Li, Yubo Ai, Jiahao Lu, ChuXin Wang, Jiacheng Deng, Hanzhi Chang, Yanzhe Liang, Wenfei Yang, Shifeng Zhang, Tianzhu Zhang
Abstract: Transformers have demonstrated impressive results for 3D point cloud semantic segmentation. However, the quadratic complexity of transformer makes computation costs high, limiting the number of points that can be processed simultaneously and impeding the modeling of long-range dependencies between objects in a single scene. Drawing inspiration from the great potential of recent state space models (SSM) for long sequence modeling, we introduce Mamba, an SSM-based architecture, to the point cloud domain and propose Pamba, a novel architecture with strong global modeling capability under linear complexity. Specifically, to make the disorderness of point clouds fit in with the causal nature of Mamba, we propose a multi-path serialization strategy applicable to point clouds. Besides, we propose the ConvMamba block to compensate for the shortcomings of Mamba in modeling local geometries and in unidirectional modeling. Pamba obtains state-of-the-art results on several 3D point cloud segmentation tasks, including ScanNet v2, ScanNet200, S3DIS and nuScenes, while its effectiveness is validated by extensive experiments.
Authors: Kejia Zhang, Juanjuan Weng, Yuanzheng Cai, Zhiming Luo, Shaozi Li
Abstract: Ensuring the robustness of deep neural networks against adversarial attacks remains a fundamental challenge in computer vision. While adversarial training (AT) has emerged as a promising defense strategy, our analysis reveals a critical limitation: AT-trained models exhibit a bias toward low-frequency features while neglecting high-frequency components. This bias is particularly concerning as each frequency component carries distinct and crucial information: low-frequency features encode fundamental structural patterns, while high-frequency features capture intricate details and textures. To address this limitation, we propose High-Frequency Feature Disentanglement and Recalibration (HFDR), a novel module that strategically separates and recalibrates frequency-specific features to capture latent semantic cues. We further introduce frequency attention regularization to harmonize feature extraction across the frequency spectrum and mitigate the inherent low-frequency bias of AT. Extensive experiments demonstrate our method's superior performance against white-box attacks and transfer attacks, while exhibiting strong generalization capabilities across diverse scenarios.
Authors: Jihyung Kil, Zheda Mai, Justin Lee, Zihe Wang, Kerrie Cheng, Lemeng Wang, Ye Liu, Arpita Chowdhury, Wei-Lun Chao
Abstract: The ability to compare objects, scenes, or situations is crucial for effective decision-making and problem-solving in everyday life. For instance, comparing the freshness of apples enables better choices during grocery shopping while comparing sofa designs helps optimize the aesthetics of our living space. Despite its significance, the comparative capability is largely unexplored in artificial general intelligence (AGI). In this paper, we introduce MLLM-CompBench, a benchmark designed to evaluate the comparative reasoning capability of multimodal large language models (MLLMs). MLLM-CompBench mines and pairs images through visually oriented questions covering eight dimensions of relative comparison: visual attribute, existence, state, emotion, temporality, spatiality, quantity, and quality. We curate a collection of around 40K image pairs using metadata from diverse vision datasets and CLIP similarity scores. These image pairs span a broad array of visual domains, including animals, fashion, sports, and both outdoor and indoor scenes. The questions are carefully crafted to discern relative characteristics between two images and are labeled by human annotators for accuracy and relevance. We use MLLM-CompBench to evaluate recent MLLMs, including GPT-4V(ision), Gemini-Pro, and LLaVA-1.6. Our results reveal notable shortcomings in their comparative abilities. We believe MLLM-COMPBENCH not only sheds light on these limitations but also establishes a solid foundation for future enhancements in the comparative capability of MLLMs.
Authors: Jingjing Wu, Zhengyao Fang, Pengyuan Lyu, Chengquan Zhang, Fanglin Chen, Guangming Lu, Wenjie Pei
Abstract: Transcription-only Supervised Text Spotting aims to learn text spotters relying only on transcriptions but no text boundaries for supervision, thus eliminating expensive boundary annotation. The crux of this task lies in locating each transcription in scene text images without location annotations. In this work, we formulate this challenging problem as a Weakly Supervised Cross-modality Contrastive Learning problem, and design a simple yet effective model dubbed WeCromCL that is able to detect each transcription in a scene image in a weakly supervised manner. Unlike typical methods for cross-modality contrastive learning that focus on modeling the holistic semantic correlation between an entire image and a text description, our WeCromCL conducts atomistic contrastive learning to model the character-wise appearance consistency between a text transcription and its correlated region in a scene image to detect an anchor point for the transcription in a weakly supervised manner. The detected anchor points by WeCromCL are further used as pseudo location labels to guide the learning of text spotting. Extensive experiments on four challenging benchmarks demonstrate the superior performance of our model over other methods. Code will be released.
Authors: Delyan Boychev, Radostin Cholakov
Abstract: Recent generative models produce images with a level of authenticity that makes them nearly indistinguishable from real photos and artwork. Potential harmful use cases of these models, necessitate the creation of robust synthetic image detectors. However, current datasets in the field contain generated images with questionable quality or have examples from one predominant content type which leads to poor generalizability of the underlying detectors. We find that the curation of a balanced amount of high-resolution generated images across various content types is crucial for the generalizability of detectors, and introduce ImagiNet, a dataset of 200K examples, spanning four categories: photos, paintings, faces, and miscellaneous. Synthetic images in ImagiNet are produced with both open-source and proprietary generators, whereas real counterparts for each content type are collected from public datasets. The structure of ImagiNet allows for a two-track evaluation system: i) classification as real or synthetic and ii) identification of the generative model. To establish a strong baseline, we train a ResNet-50 model using a self-supervised contrastive objective (SelfCon) for each track which achieves evaluation AUC of up to 0.99 and balanced accuracy ranging from 86% to 95%, even under conditions that involve compression and resizing. The provided model is generalizable enough to achieve zero-shot state-of-the-art performance on previous synthetic detection benchmarks. We provide ablations to demonstrate the importance of content types and publish code and data.
Authors: Yunfeng Zhao, Huiyu Zhou, Fei Wu, Xifeng Wu
Abstract: Image classification is a fundamental computer vision task and an important baseline for deep metric learning. In decades efforts have been made on enhancing image classification accuracy by using deep learning models while less attention has been paid on the reasoning aspect of the recognition, i.e., predictions could be made because of background or other surrounding objects rather than the target object. Hierarchical knowledge about image categories depicts inter-class similarities or dissimilarities. Effective fusion of such knowledge with deep learning image classification models is promising in improving target object identification and enhancing the reasoning aspect of the recognition. In this paper, we propose a novel deep metric learning based method to effectively fuse prior knowledge about image categories with mainstream backbone image classification models and enhance the reasoning aspect of the recognition in an end-to-end manner. Existing deep metric learning incorporated image classification methods mainly focus on whether sampled images are from the same class. A new triplet loss function term that aligns distances in the model latent space with those in knowledge space is presented and incorporated in the proposed method to facilitate the dual-modality fusion. Extensive experiments on the CIFAR-10, CIFAR-100, Mini-ImageNet, and ImageNet-1K datasets evaluated the proposed method, and results indicate that the proposed method is effective in enhancing the reasoning aspect of image recognition in terms of weakly-supervised object localization performance.
Authors: Xiaozheng Zheng, Chao Wen, Zhaohu Li, Weiyi Zhang, Zhuo Su, Xu Chang, Yang Zhao, Zheng Lv, Xiaoyuan Zhang, Yongjie Zhang, Guidong Wang, Lan Xu
Abstract: In this paper, we present a novel 3D head avatar creation approach capable of generalizing from few-shot in-the-wild data with high-fidelity and animatable robustness. Given the underconstrained nature of this problem, incorporating prior knowledge is essential. Therefore, we propose a framework comprising prior learning and avatar creation phases. The prior learning phase leverages 3D head priors derived from a large-scale multi-view dynamic dataset, and the avatar creation phase applies these priors for few-shot personalization. Our approach effectively captures these priors by utilizing a Gaussian Splatting-based auto-decoder network with part-based dynamic modeling. Our method employs identity-shared encoding with personalized latent codes for individual identities to learn the attributes of Gaussian primitives. During the avatar creation phase, we achieve fast head avatar personalization by leveraging inversion and fine-tuning strategies. Extensive experiments demonstrate that our model effectively exploits head priors and successfully generalizes them to few-shot personalization, achieving photo-realistic rendering quality, multi-view consistency, and stable animation.
Authors: Erjin Bao, Ching-Chun Chang, Hanrui Wang, Isao Echizen
Abstract: With the proliferation of AI agents in various domains, protecting the ownership of AI models has become crucial due to the significant investment in their development. Unauthorized use and illegal distribution of these models pose serious threats to intellectual property, necessitating effective copyright protection measures. Model watermarking has emerged as a key technique to address this issue, embedding ownership information within models to assert rightful ownership during copyright disputes. This paper presents several contributions to model watermarking: a self-authenticating black-box watermarking protocol using hash techniques, a study on evidence forgery attacks using adversarial perturbations, a proposed defense involving a purification step to counter adversarial attacks, and a purification-agnostic curriculum proxy learning method to enhance watermark robustness and model performance. Experimental results demonstrate the effectiveness of these approaches in improving the security, reliability, and performance of watermarked models.
Authors: Muye Huang, Han Lai, Xinyu Zhang, Wenjun Wu, Jie Ma, Lingling Zhang, Jun Liu
Abstract: Chart understanding enables automated data analysis for humans, which requires models to achieve highly accurate visual comprehension. While existing Visual Language Models (VLMs) have shown progress in chart understanding, the lack of high-quality training data and comprehensive evaluation benchmarks hinders VLM chart comprehension. In this paper, we introduce EvoChart, a novel self-training method for generating synthetic chart data to enhance VLMs' capabilities in real-world chart comprehension. We also propose EvoChart-QA, a noval benchmark for measuring models' chart comprehension abilities in real-world scenarios. Specifically, EvoChart is a unique self-training data synthesis approach that simultaneously produces high-quality training corpus and a high-performance chart understanding model. EvoChart-QA consists of 650 distinct real-world charts collected from 140 different websites and 1,250 expert-curated questions that focus on chart understanding. Experimental results on various open-source and proprietary VLMs tested on EvoChart-QA demonstrate that even the best proprietary model, GPT-4o, achieves only 49.8% accuracy. Moreover, the EvoChart method significantly boosts the performance of open-source VLMs on real-world chart understanding tasks, achieving 54.2% accuracy on EvoChart-QA.
Authors: Yan Chen, Di Huang, Zhichao Liao, Xi Cheng, Xinghui Li, Long Zeng
Abstract: The trend of employing training-free methods for point cloud recognition is becoming increasingly popular due to its significant reduction in computational resources and time costs. However, existing approaches are limited as they typically extract either geometric or semantic features. To address this limitation, we are the first to propose a novel training-free method that integrates both geometric and semantic features. For the geometric branch, we adopt a non-parametric strategy to extract geometric features. In the semantic branch, we leverage a model aligned with text features to obtain semantic features. Additionally, we introduce the GFE module to complement the geometric information of point clouds and the MFF module to improve performance in few-shot settings. Experimental results demonstrate that our method outperforms existing state-of-the-art training-free approaches on mainstream benchmark datasets, including ModelNet and ScanObiectNN.
Authors: Jiaxin Cheng, Zixu Zhao, Tong He, Tianjun Xiao, Yicong Zhou, Zheng Zhang
Abstract: Recent advancements in generative models have significantly enhanced their capacity for image generation, enabling a wide range of applications such as image editing, completion and video editing. A specialized area within generative modeling is layout-to-image (L2I) generation, where predefined layouts of objects guide the generative process. In this study, we introduce a novel regional cross-attention module tailored to enrich layout-to-image generation. This module notably improves the representation of layout regions, particularly in scenarios where existing methods struggle with highly complex and detailed textual descriptions. Moreover, while current open-vocabulary L2I methods are trained in an open-set setting, their evaluations often occur in closed-set environments. To bridge this gap, we propose two metrics to assess L2I performance in open-vocabulary scenarios. Additionally, we conduct a comprehensive user study to validate the consistency of these metrics with human preferences.
Authors: Abhijay Ghildyal, Nabajeet Barman, Saman Zadtootaghaj
Abstract: For full-reference image quality assessment (FR-IQA) using deep-learning approaches, the perceptual similarity score between a distorted image and a reference image is typically computed as a distance measure between features extracted from a pretrained CNN or more recently, a Transformer network. Often, these intermediate features require further fine-tuning or processing with additional neural network layers to align the final similarity scores with human judgments. So far, most IQA models based on foundation models have primarily relied on the final layer or the embedding for the quality score estimation. In contrast, this work explores the potential of utilizing the intermediate features of these foundation models, which have largely been unexplored so far in the design of low-level perceptual similarity metrics. We demonstrate that the intermediate features are comparatively more effective. Moreover, without requiring any training, these metrics can outperform both traditional and state-of-the-art learned metrics by utilizing distance measures between the features.
Authors: Jianxiong Gao, Yanwei Fu, Yuqian Fu, Yun Wang, Xuelin Qian, Jianfeng Feng
Abstract: Reconstructing 3D visuals from functional Magnetic Resonance Imaging (fMRI) data, introduced as Recon3DMind, is of significant interest to both cognitive neuroscience and computer vision. To advance this task, we present the fMRI-3D dataset, which includes data from 15 participants and showcases a total of 4,768 3D objects. The dataset consists of two components: fMRI-Shape, previously introduced and available at https://huggingface.co/datasets/Fudan-fMRI/fMRI-Shape, and fMRI-Objaverse, proposed in this paper and available at https://huggingface.co/datasets/Fudan-fMRI/fMRI-Objaverse. fMRI-Objaverse includes data from 5 subjects, 4 of whom are also part of the core set in fMRI-Shape. Each subject views 3,142 3D objects across 117 categories, all accompanied by text captions. This significantly enhances the diversity and potential applications of the dataset. Moreover, we propose MinD-3D++, a novel framework for decoding textured 3D visual information from fMRI signals. The framework evaluates the feasibility of not only reconstructing 3D objects from the human mind but also generating, for the first time, 3D textured meshes with detailed textures from fMRI data. We establish new benchmarks by designing metrics at the semantic, structural, and textured levels to evaluate model performance. Furthermore, we assess the model's effectiveness in out-of-distribution settings and analyze the attribution of the proposed 3D pari fMRI dataset in visual regions of interest (ROIs) in fMRI signals. Our experiments demonstrate that MinD-3D++ not only reconstructs 3D objects with high semantic and spatial accuracy but also provides deeper insights into how the human brain processes 3D visual information. Project page: https://jianxgao.github.io/MinD-3D.
URLs: https://huggingface.co/datasets/Fudan-fMRI/fMRI-Shape,, https://huggingface.co/datasets/Fudan-fMRI/fMRI-Objaverse., https://jianxgao.github.io/MinD-3D.
Authors: Wenbo Wei, Jun Wang, Abhir Bhalerao
Abstract: To help address the occlusion problem in panoptic segmentation and image understanding, this paper proposes a new large-scale dataset named COCO-OLAC (COCO Occlusion Labels for All Computer Vision Tasks), which is derived from the COCO dataset by manually labelling images into three perceived occlusion levels. Using COCO-OLAC, we systematically assess and quantify the impact of occlusion on panoptic segmentation on samples having different levels of occlusion. Comparative experiments with SOTA panoptic models demonstrate that the presence of occlusion significantly affects performance, with higher occlusion levels resulting in notably poorer performance. Additionally, we propose a straightforward yet effective method as an initial attempt to leverage the occlusion annotation using contrastive learning to render a model that learns a more robust representation capturing different severities of occlusion. Experimental results demonstrate that the proposed approach boosts the performance of the baseline model and achieves SOTA performance on the proposed COCO-OLAC dataset.
Authors: Nanqing Liu, Xun Xu, Yongyi Su, Haojie Zhang, Heng-Chao Li
Abstract: Segment Anything Model (SAM) is an advanced foundational model for image segmentation, which is gradually being applied to remote sensing images (RSIs). Due to the domain gap between RSIs and natural images, traditional methods typically use SAM as a source pre-trained model and fine-tune it with fully supervised masks. Unlike these methods, our work focuses on fine-tuning SAM using more convenient and challenging point annotations. Leveraging SAM's zero-shot capabilities, we adopt a self-training framework that iteratively generates pseudo-labels for training. However, if the pseudo-labels contain noisy labels, there is a risk of error accumulation. To address this issue, we extract target prototypes from the target dataset and use the Hungarian algorithm to match them with prediction prototypes, preventing the model from learning in the wrong direction. Additionally, due to the complex backgrounds and dense distribution of objects in RSI, using point prompts may result in multiple objects being recognized as one. To solve this problem, we propose a negative prompt calibration method based on the non-overlapping nature of instance masks. In brief, we use the prompts of overlapping masks as corresponding negative signals, resulting in refined masks. Combining the above methods, we propose a novel Pointly-supervised Segment Anything Model named PointSAM. We conduct experiments on RSI datasets, including WHU, HRSID, and NWPU VHR-10, and the results show that our method significantly outperforms direct testing with SAM, SAM2, and other comparison methods. Furthermore, we introduce PointSAM as a point-to-box converter and achieve encouraging results, suggesting that this method can be extended to other point-supervised tasks. The code is available at https://github.com/Lans1ng/PointSAM.
Authors: Huafeng Qin, Hongyu Zhu, Xin Jin, Xin Yu, Mounim A. El-Yacoubi, Shuqiang Yang
Abstract: Eye movement biometrics has received increasing attention thanks to its highly secure identification. Although deep learning (DL) models have shown success in eye movement recognition, their architectures largely rely on human prior knowledge. Differentiable Neural Architecture Search (DARTS) automates the manual process of architecture design with high search efficiency. However, DARTS typically stacks multiple cells to form a convolutional network, which limits the diversity of architecture. Furthermore, DARTS generally searches for architectures using shallower networks than those used in the evaluation, creating a significant disparity in architecture depth between the search and evaluation phases. To address this issue, we propose EM-DARTS, a hierarchical differentiable architecture search algorithm to automatically design the DL architecture for eye movement recognition. First, we define a supernet and propose a global and local alternate Neural Architecture Search method to search the optimal architecture alternately with a differentiable neural architecture search. The local search strategy aims to find an optimal architecture for different cells while the global search strategy is responsible for optimizing the architecture of the target network. To minimize redundancy, transfer entropy is proposed to compute the information amount of each layer, thereby further simplifying the network search process. Experimental results on three public datasets demonstrate that the proposed EM-DARTS is capable of producing an optimal architecture that leads to state-of-the-art recognition performance, {Specifically, the recognition models developed using EM-DARTS achieved the lowest EERs of 0.0453 on the GazeBase dataset, 0.0377 on the JuDo1000 dataset, and 0.1385 on the EMglasses dataset.
Authors: Junzhuo Liu, Xuzheng Yang, Weiwei Li, Peng Wang
Abstract: Referring Expression Comprehension (REC) is a crucial cross-modal task that objectively evaluates the capabilities of language understanding, image comprehension, and language-to-image grounding. Consequently, it serves as an ideal testing ground for Multi-modal Large Language Models (MLLMs). In pursuit of this goal, we have established a new REC dataset characterized by two key features: Firstly, it is designed with controllable varying levels of difficulty, necessitating multi-level fine-grained reasoning across object categories, attributes, and multi-hop relationships. Secondly, it includes negative text and images created through fine-grained editing and generation based on existing data, thereby testing the model's ability to correctly reject scenarios where the target object is not visible in the image--an essential aspect often overlooked in existing datasets and approaches. Utilizing this high-quality dataset, we conducted comprehensive evaluations of both state-of-the-art specialist models and MLLMs. Our findings indicate that there remains a significant gap in achieving satisfactory grounding performance. We anticipate that our dataset will inspire new approaches to enhance visual reasoning and develop more advanced cross-modal interaction strategies, ultimately unlocking the full potential of MLLMs. Our code and the datasets are available at https://github.com/liujunzhuo/FineCops-Ref.
Authors: Liang Shi, Boyu Jiang, Tong Zeng, Feng Guo
Abstract: Accurately identifying, understanding and describing traffic safety-critical events (SCEs), including crashes, tire strikes, and near-crashes, is crucial for advanced driver assistance systems, automated driving systems, and traffic safety. As SCEs are rare events, most general vision-language models (VLMs) have not been trained sufficiently to link SCE videos and narratives, which could lead to hallucinations and missing key safety characteristics. Here, we introduce ScVLM, a novel hybrid methodology that integrates supervised and contrastive learning techniques to classify the severity and types of SCEs, as well as to generate narrative descriptions of SCEs. This approach utilizes classification to enhance VLMs' comprehension of driving videos and improve the rationality of event descriptions. The proposed approach is trained on and evaluated by more than 8,600 SCEs from the Second Strategic Highway Research Program Naturalistic Driving Study dataset, the largest publicly accessible driving dataset with videos and SCE annotations. The results demonstrate the superiority of the proposed approach in generating contextually accurate event descriptions and mitigating VLM hallucinations. The code will be available at https://github.com/datadrivenwheels/ScVLM.
Authors: Xavier Juanola, Gloria Haro, Magdalena Fuentes
Abstract: The task of Visual Sound Source Localization (VSSL) involves identifying the location of sound sources in visual scenes, integrating audio-visual data for enhanced scene understanding. Despite advancements in state-of-the-art (SOTA) models, we observe three critical flaws: i) The evaluation of the models is mainly focused in sounds produced by objects that are visible in the image, ii) The evaluation often assumes a prior knowledge of the size of the sounding object, and iii) No universal threshold for localization in real-world scenarios is established, as previous approaches only consider positive examples without accounting for both positive and negative cases. In this paper, we introduce a novel test set and metrics designed to complete the current standard evaluation of VSSL models by testing them in scenarios where none of the objects in the image corresponds to the audio input, i.e. a negative audio. We consider three types of negative audio: silence, noise and offscreen. Our analysis reveals that numerous SOTA models fail to appropriately adjust their predictions based on audio input, suggesting that these models may not be leveraging audio information as intended. Additionally, we provide a comprehensive analysis of the range of maximum values in the estimated audio-visual similarity maps, in both positive and negative audio cases, and show that most of the models are not discriminative enough, making them unfit to choose a universal threshold appropriate to perform sound localization without any a priori information of the sounding object, that is, object size and visibility.
Authors: Yazhou Zhu, Minxian Li, Qiaolin Ye, Shidong Wang, Tong Xin, Haofeng Zhang
Abstract: Few-shot medical image segmentation (FSMIS) aims to perform the limited annotated data learning in the medical image analysis scope. Despite the progress has been achieved, current FSMIS models are all trained and deployed on the same data domain, as is not consistent with the clinical reality that medical imaging data is always across different data domains (e.g. imaging modalities, institutions and equipment sequences). How to enhance the FSMIS models to generalize well across the different specific medical imaging domains? In this paper, we focus on the matching mechanism of the few-shot semantic segmentation models and introduce an Earth Mover's Distance (EMD) calculation based domain robust matching mechanism for the cross-domain scenario. Specifically, we formulate the EMD transportation process between the foreground support-query features, the texture structure aware weights generation method, which proposes to perform the sobel based image gradient calculation over the nodes, is introduced in the EMD matching flow to restrain the domain relevant nodes. Besides, the point set level distance measurement metric is introduced to calculated the cost for the transportation from support set nodes to query set nodes. To evaluate the performance of our model, we conduct experiments on three scenarios (i.e., cross-modal, cross-sequence and cross-institution), which includes eight medical datasets and involves three body regions, and the results demonstrate that our model achieves the SoTA performance against the compared models.
Authors: Gustavo A. Bas\'ilio, Thiago B. Pereira, Alessandro L. Koerich, Hermano Tavares, Ludmila Dias, Maria das Gra\c{c}as da S. Teixeira, Rafael T. Sousa, Wilian H. Hisatugu, Amanda S. Mota, Anilton S. Garcia, Marco Aur\'elio K. Galletta, Thiago M. Paix\~ao
Abstract: Major Depressive Disorder and anxiety disorders affect millions globally, contributing significantly to the burden of mental health issues. Early screening is crucial for effective intervention, as timely identification of mental health issues can significantly improve treatment outcomes. Artificial intelligence (AI) can be valuable for improving the screening of mental disorders, enabling early intervention and better treatment outcomes. AI-driven screening can leverage the analysis of multiple data sources, including facial features in digital images. However, existing methods often rely on controlled environments or specialized equipment, limiting their broad applicability. This study explores the potential of AI models for ubiquitous depression-anxiety screening given face-centric selfies. The investigation focuses on high-risk pregnant patients, a population that is particularly vulnerable to mental health issues. To cope with limited training data resulting from our clinical setup, pre-trained models were utilized in two different approaches: fine-tuning convolutional neural networks (CNNs) originally designed for facial expression recognition and employing vision-language models (VLMs) for zero-shot analysis of facial expressions. Experimental results indicate that the proposed VLM-based method significantly outperforms CNNs, achieving an accuracy of 77.6%. Although there is significant room for improvement, the results suggest that VLMs can be a promising approach for mental health screening.
Authors: Virmarie Maquiling, Sean Anthony Byrne, Diederick C. Niehorster, Marco Carminati, Enkelejda Kasneci
Abstract: We explore the transformative potential of SAM 2, a vision foundation model, in advancing gaze estimation and eye tracking technologies. By significantly reducing annotation time, lowering technical barriers through its ease of deployment, and enhancing segmentation accuracy, SAM 2 addresses critical challenges faced by researchers and practitioners. Utilizing its zero-shot segmentation capabilities with minimal user input-a single click per video-we tested SAM 2 on over 14 million eye images from diverse datasets, including virtual reality setups and the world's largest unified dataset recorded using wearable eye trackers. Remarkably, in pupil segmentation tasks, SAM 2 matches the performance of domain-specific models trained solely on eye images, achieving competitive mean Intersection over Union (mIoU) scores of up to 93% without fine-tuning. Additionally, we provide our code and segmentation masks for these widely used datasets to promote further research.
Authors: Alaa Awad, Salah A. Aly
Abstract: Leukemia, a severe form of blood cancer, claims thousands of lives each year. This study focuses on the detection of Acute Lymphoblastic Leukemia (ALL) using advanced image processing and deep learning techniques. By leveraging recent advancements in artificial intelligence, the research evaluates the reliability of these methods in practical, real-world scenarios. Specifically, it examines the performance of state-of-the-art YOLO models, including YOLOv8 and YOLOv11, to distinguish between malignant and benign white blood cells and accurately identify different stages of ALL, including early stages. Moreover, the models demonstrate the ability to detect hematogones, which are frequently misclassified as ALL. With accuracy rates reaching 98.8%, this study highlights the potential of these algorithms to provide robust and precise leukemia detection across diverse datasets and conditions.
Authors: Di Qiu, Zheng Chen, Rui Wang, Mingyuan Fan, Changqian Yu, Junshi Huang, Xiang Wen
Abstract: Recent advancements in character video synthesis still depend on extensive fine-tuning or complex 3D modeling processes, which can restrict accessibility and hinder real-time applicability. To address these challenges, we propose a simple yet effective tuning-free framework for character video synthesis, named MovieCharacter, designed to streamline the synthesis process while ensuring high-quality outcomes. Our framework decomposes the synthesis task into distinct, manageable modules: character segmentation and tracking, video object removal, character motion imitation, and video composition. This modular design not only facilitates flexible customization but also ensures that each component operates collaboratively to effectively meet user needs. By leveraging existing open-source models and integrating well-established techniques, MovieCharacter achieves impressive synthesis results without necessitating substantial resources or proprietary datasets. Experimental results demonstrate that our framework enhances the efficiency, accessibility, and adaptability of character video synthesis, paving the way for broader creative and interactive applications.
Authors: Junjie Ni, Guofeng Zhang, Guanglin Li, Yijin Li, Xinyang Liu, Zhaoyang Huang, Hujun Bao
Abstract: We tackle the efficiency problem of learning local feature matching. Recent advancements have given rise to purely CNN-based and transformer-based approaches, each augmented with deep learning techniques. While CNN-based methods often excel in matching speed, transformer-based methods tend to provide more accurate matches. We propose an efficient transformer-based network architecture for local feature matching. This technique is built on constructing multiple homography hypotheses to approximate the continuous correspondence in the real world and uni-directional cross-attention to accelerate the refinement. On the YFCC100M dataset, our matching accuracy is competitive with LoFTR, a state-of-the-art transformer-based architecture, while the inference speed is boosted to 4 times, even outperforming the CNN-based methods. Comprehensive evaluations on other open datasets such as Megadepth, ScanNet, and HPatches demonstrate our method's efficacy, highlighting its potential to significantly enhance a wide array of downstream applications.
Authors: Chinthani Sugandhika, Chen Li, Deepu Rajan, Basura Fernando
Abstract: Graph based representation has been widely used in modelling spatio-temporal relationships in video understanding. Although effective, existing graph-based approaches focus on capturing the human-object relationships while ignoring fine-grained semantic properties of the action components. These semantic properties are crucial for understanding the current situation, such as where does the action takes place, what tools are used and functional properties of the objects. In this work, we propose a graph-based representation called Situational Scene Graph (SSG) to encode both human-object relationships and the corresponding semantic properties. The semantic details are represented as predefined roles and values inspired by situation frame, which is originally designed to represent a single action. Based on our proposed representation, we introduce the task of situational scene graph generation and propose a multi-stage pipeline Interactive and Complementary Network (InComNet) to address the task. Given that the existing datasets are not applicable to the task, we further introduce a SSG dataset whose annotations consist of semantic role-value frames for human, objects and verb predicates of human-object relations. Finally, we demonstrate the effectiveness of our proposed SSG representation by testing on different downstream tasks. Experimental results show that the unified representation can not only benefit predicate classification and semantic role-value classification, but also benefit reasoning tasks on human-centric situation understanding. We will release the code and the dataset soon.
Authors: Qishuai Wen, Chun-Guang Li
Abstract: State-of-the-art methods for Transformer-based semantic segmentation typically adopt Transformer decoders that are used to extract additional embeddings from image embeddings via cross-attention, refine either or both types of embeddings via self-attention, and project image embeddings onto the additional embeddings via dot-product. Despite their remarkable success, these empirical designs still lack theoretical justifications or interpretations, thus hindering potentially principled improvements. In this paper, we argue that there are fundamental connections between semantic segmentation and compression, especially between the Transformer decoders and Principal Component Analysis (PCA). From such a perspective, we derive a white-box, fully attentional DEcoder for PrIncipled semantiC segemenTation (DEPICT), with the interpretations as follows: 1) the self-attention operator refines image embeddings to construct an ideal principal subspace that aligns with the supervision and retains most information; 2) the cross-attention operator seeks to find a low-rank approximation of the refined image embeddings, which is expected to be a set of orthonormal bases of the principal subspace and corresponds to the predefined classes; 3) the dot-product operation yields compact representation for image embeddings as segmentation masks. Experiments conducted on dataset ADE20K find that DEPICT consistently outperforms its black-box counterpart, Segmenter, and it is light weight and more robust.
Authors: Jiaxuan Chen, Bo Zhang, Qingdong He, Jinlong Peng, Li Niu
Abstract: Generative image composition aims to regenerate the given foreground object in the background image to produce a realistic composite image. In this work, we propose an effective finetuning strategy for generative image composition model, in which we finetune a pretrained model using one or more images containing the same foreground object. Moreover, we propose a multi-reference strategy, which allows the model to take in multiple reference images of the foreground object. The experiments on MureCOM dataset verify the effectiveness of our method.
Authors: Piyush Bagad, Makarand Tapaswi, Cees G. M. Snoek, Andrew Zisserman
Abstract: We study the connection between audio-visual observations and the underlying physics of a mundane yet intriguing everyday activity: pouring liquids. Given only the sound of liquid pouring into a container, our objective is to automatically infer physical properties such as the liquid level, the shape and size of the container, the pouring rate and the time to fill. To this end, we: (i) show in theory that these properties can be determined from the fundamental frequency (pitch); (ii) train a pitch detection model with supervision from simulated data and visual data with a physics-inspired objective; (iii) introduce a new large dataset of real pouring videos for a systematic study; (iv) show that the trained model can indeed infer these physical properties for real data; and finally, (v) we demonstrate strong generalization to various container shapes, other datasets, and in-the-wild YouTube videos. Our work presents a keen understanding of a narrow yet rich problem at the intersection of acoustics, physics, and learning. It opens up applications to enhance multisensory perception in robotic pouring.
Authors: Defan Chen, Luchan Zhang
Abstract: Detecting small objects in complex scenes, such as those captured by drones, is a daunting challenge due to the difficulty in capturing the complex features of small targets. While the YOLO family has achieved great success in large target detection, its performance is less than satisfactory when faced with small targets. Because of this, this paper proposes a revolutionary model SL-YOLO (Stronger and Lighter YOLO) that aims to break the bottleneck of small target detection. We propose the Hierarchical Extended Path Aggregation Network (HEPAN), a pioneering cross-scale feature fusion method that can ensure unparalleled detection accuracy even in the most challenging environments. At the same time, without sacrificing detection capabilities, we design the C2fDCB lightweight module and add the SCDown downsampling module to greatly reduce the model's parameters and computational complexity. Our experimental results on the VisDrone2019 dataset reveal a significant improvement in performance, with mAP@0.5 jumping from 43.0% to 46.9% and mAP@0.5:0.95 increasing from 26.0% to 28.9%. At the same time, the model parameters are reduced from 11.1M to 9.6M, and the FPS can reach 132, making it an ideal solution for real-time small object detection in resource-constrained environments.
Authors: Zhendong Liu, Yuanbi Nie, Yingshui Tan, Jiaheng Liu, Xiangyu Yue, Qiushi Cui, Chongjun Wang, Xiaoyong Zhu, Bo Zheng
Abstract: Benefiting from the powerful capabilities of Large Language Models (LLMs), pre-trained visual encoder models connected to LLMs form Vision Language Models (VLMs). However, recent research shows that the visual modality in VLMs is highly vulnerable, allowing attackers to bypass safety alignment in LLMs through visually transmitted content, launching harmful attacks. To address this challenge, we propose a progressive concept-based alignment strategy, PSA-VLM, which incorporates safety modules as concept bottlenecks to enhance visual modality safety alignment. By aligning model predictions with specific safety concepts, we improve defenses against risky images, enhancing explainability and controllability while minimally impacting general performance. Our method is obtained through two-stage training. The low computational cost of the first stage brings very effective performance improvement, and the fine-tuning of the language model in the second stage further improves the safety performance. Our method achieves state-of-the-art results on popular VLM safety benchmark.
Authors: Zhicheng Zhao, Changfu Zhou, Yu Zhang, Chenglong Li, Xiaoliang Ma, Jin Tang
Abstract: Remote Sensing Visual Question Answering (RSVQA) has gained significant research interest. However, current RSVQA methods are limited by the imaging mechanisms of optical sensors, particularly under challenging conditions such as cloud-covered and low-light scenarios. Given the all-time and all-weather imaging capabilities of Synthetic Aperture Radar (SAR), it is crucial to investigate the integration of optical-SAR images to improve RSVQA performance. In this work, we propose a Text-guided Coarse-to-Fine Fusion Network (TGFNet), which leverages the semantic relationships between question text and multi-source images to guide the network toward complementary fusion at the feature level. Specifically, we develop a Text-guided Coarse-to-Fine Attention Refinement (CFAR) module to focus on key areas related to the question in complex remote sensing images. This module progressively directs attention from broad areas to finer details through key region routing, enhancing the model's ability to focus on relevant regions. Furthermore, we propose an Adaptive Multi-Expert Fusion (AMEF) module that dynamically integrates different experts, enabling the adaptive fusion of optical and SAR features. In addition, we create the first large-scale benchmark dataset for evaluating optical-SAR RSVQA methods, comprising 6,008 well-aligned optical-SAR image pairs and 1,036,694 well-labeled question-answer pairs across 16 diverse question types, including complex relational reasoning questions. Extensive experiments on the proposed dataset demonstrate that our TGFNet effectively integrates complementary information between optical and SAR images, significantly improving the model's performance in challenging scenarios. The dataset is available at: https://github.com/mmic-lcl/. Index Terms: Remote Sensing Visual Question Answering, Multi-source Data Fusion, Multimodal, Remote Sensing, OPT-SAR.
Authors: Wentao Qu, Jing Wang, YongShun Gong, Xiaoshui Huang, Liang Xiao
Abstract: Existing conditional Denoising Diffusion Probabilistic Models (DDPMs) with a Noise-Conditional Framework (NCF) remain challenging for 3D scene understanding tasks, as the complex geometric details in scenes increase the difficulty of fitting the gradients of the data distribution (the scores) from semantic labels. This also results in longer training and inference time for DDPMs compared to non-DDPMs. From a different perspective, we delve deeply into the model paradigm dominated by the Conditional Network. In this paper, we propose an end-to-end robust semantic Segmentation Network based on a Conditional-Noise Framework (CNF) of DDPMs, named CDSegNet. Specifically, CDSegNet models the Noise Network (NN) as a learnable noise-feature generator. This enables the Conditional Network (CN) to understand 3D scene semantics under multi-level feature perturbations, enhancing the generalization in unseen scenes. Meanwhile, benefiting from the noise system of DDPMs, CDSegNet exhibits strong noise and sparsity robustness in experiments. Moreover, thanks to CNF, CDSegNet can generate the semantic labels in a single-step inference like non-DDPMs, due to avoiding directly fitting the scores from semantic labels in the dominant network of CDSegNet. On public indoor and outdoor benchmarks, CDSegNet significantly outperforms existing methods, achieving state-of-the-art performance.
Authors: Hugo Resende, Isabela Borlido, Victor Sundermann, Eduardo B. Neto, Silvio Jamil F. Guimar\~aes, Fabio Faria, Alvaro Luiz Fazenda
Abstract: Tropical forests play an essential role in the planet's ecosystem, making the conservation of these biomes a worldwide priority. However, ongoing deforestation and degradation pose a significant threat to their existence, necessitating effective monitoring and the proposal of actions to mitigate the damage caused by these processes. In this regard, initiatives range from government and private sector monitoring programs to solutions based on citizen science campaigns, for example. Particularly in the context of citizen science campaigns, the segmentation of remote sensing images to identify deforested areas and subsequently submit them to analysis by non-specialized volunteers is necessary. Thus, segmentation using superpixel-based techniques proves to be a viable solution for this important task. Therefore, this paper presents an analysis of 22 superpixel-based segmentation methods applied to remote sensing images, aiming to identify which of them are more suitable for generating segments for citizen science campaigns. The results reveal that seven of the segmentation methods outperformed the baseline method (SLIC) currently employed in the ForestEyes citizen science project, indicating an opportunity for improvement in this important stage of campaign development.
Authors: Bohai Gu, Hao Luo, Song Guo, Peiran Dong
Abstract: Recently, diffusion-based methods have achieved great improvements in the video inpainting task. However, these methods still face many challenges, such as maintaining temporal consistency and the time-consuming issue. This paper proposes an advanced video inpainting framework using optical Flow-guided Efficient Diffusion, called FloED. Specifically, FloED employs a dual-branch architecture, where a flow branch first restores corrupted flow and a multi-scale flow adapter provides motion guidance to the main inpainting branch. Additionally, a training-free latent interpolation method is proposed to accelerate the multi-step denoising process using flow warping. Further introducing a flow attention cache mechanism, FLoED efficiently reduces the computational cost brought by incorporating optical flow. Comprehensive experiments in both background restoration and object removal tasks demonstrate that FloED outperforms state-of-the-art methods from the perspective of both performance and efficiency.
Authors: Gorkem Polat, \"Umit Mert \c{C}a\u{g}lar, Alptekin Temizel
Abstract: Assessing disease severity with ordinal classes, where each class reflects increasing severity levels, benefits from loss functions designed for this ordinal structure. Traditional categorical loss functions, like Cross-Entropy (CE), often perform suboptimally in these scenarios. To address this, we propose a novel loss function, Class Distance Weighted Cross-Entropy (CDW-CE), which penalizes misclassifications more severely when the predicted and actual classes are farther apart. We evaluated CDW-CE using various deep architectures, comparing its performance against several categorical and ordinal loss functions. To assess the quality of latent representations, we used t-distributed stochastic neighbor embedding (t-SNE) and uniform manifold approximation and projection (UMAP) visualizations, quantified the clustering quality using the Silhouette Score, and compared Class Activation Maps (CAM) generated by models trained with CDW-CE and CE loss. Feedback from domain experts was incorporated to evaluate how well model attention aligns with expert opinion. Our results show that CDW-CE consistently improves performance in ordinal image classification tasks. It achieves higher Silhouette Scores, indicating better class discrimination capability, and its CAM visualizations show a stronger focus on clinically significant regions, as validated by domain experts. Receiver operator characteristics (ROC) curves and the area under the curve (AUC) scores highlight that CDW-CE outperforms other loss functions, including prominent ordinal loss functions from the literature.
Authors: Zhe Chen, Weiyun Wang, Yue Cao, Yangzhou Liu, Zhangwei Gao, Erfei Cui, Jinguo Zhu, Shenglong Ye, Hao Tian, Zhaoyang Liu, Lixin Gu, Xuehui Wang, Qingyun Li, Yimin Ren, Zixuan Chen, Jiapeng Luo, Jiahao Wang, Tan Jiang, Bo Wang, Conghui He, Botian Shi, Xingcheng Zhang, Han Lv, Yi Wang, Wenqi Shao, Pei Chu, Zhongying Tu, Tong He, Zhiyong Wu, Huipeng Deng, Jiaye Ge, Kai Chen, Kaipeng Zhang, Limin Wang, Min Dou, Lewei Lu, Xizhou Zhu, Tong Lu, Dahua Lin, Yu Qiao, Jifeng Dai, Wenhai Wang
Abstract: We introduce InternVL 2.5, an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0, maintaining its core model architecture while introducing significant enhancements in training and testing strategies as well as data quality. In this work, we delve into the relationship between model scaling and performance, systematically exploring the performance trends in vision encoders, language models, dataset sizes, and test-time configurations. Through extensive evaluations on a wide range of benchmarks, including multi-discipline reasoning, document understanding, multi-image / video understanding, real-world comprehension, multimodal hallucination detection, visual grounding, multilingual capabilities, and pure language processing, InternVL 2.5 exhibits competitive performance, rivaling leading commercial models such as GPT-4o and Claude-3.5-Sonnet. Notably, our model is the first open-source MLLMs to surpass 70% on the MMMU benchmark, achieving a 3.7-point improvement through Chain-of-Thought (CoT) reasoning and showcasing strong potential for test-time scaling. We hope this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems. HuggingFace demo see https://huggingface.co/spaces/OpenGVLab/InternVL
Authors: Xin Zhao, Xiaojun Chen, Yuexin Xuan, Zhendong Zhao, Xiaojun Jia, Xinfeng Li, Xiaofeng Wang
Abstract: The rise of deep learning models in the digital era has raised substantial concerns regarding the generation of Not-Safe-for-Work (NSFW) content. Existing defense methods primarily involve model fine-tuning and post-hoc content moderation. Nevertheless, these approaches largely lack scalability in eliminating harmful content, degrade the quality of benign image generation, or incur high inference costs. To address these challenges, we propose an innovative framework named \textit{Buster}, which injects backdoors into the text encoder to prevent NSFW content generation. Buster leverages deep semantic information rather than explicit prompts as triggers, redirecting NSFW prompts towards targeted benign prompts. Additionally, Buster employs energy-based training data generation through Langevin dynamics for adversarial knowledge augmentation, thereby ensuring robustness in harmful concept definition. This approach demonstrates exceptional resilience and scalability in mitigating NSFW content. Particularly, Buster fine-tunes the text encoder of Text-to-Image models within merely five minutes, showcasing its efficiency. Our extensive experiments denote that Buster outperforms nine state-of-the-art baselines, achieving a superior NSFW content removal rate of at least 91.2\% while preserving the quality of harmless images.
Authors: Pablo Morales-\'Alvarez, Stergios Christodoulidis, Maria Vakalopoulou, Pablo Piantanida, Jose Dolz
Abstract: The emergence of large pre-trained vision-language models (VLMs) represents a paradigm shift in machine learning, with unprecedented results in a broad span of visual recognition tasks. CLIP, one of the most popular VLMs, has exhibited remarkable zero-shot and transfer learning capabilities in classification. To transfer CLIP to downstream tasks, adapters constitute a parameter-efficient approach that avoids backpropagation through the large model (unlike related prompt learning methods). However, CLIP adapters have been developed to target discriminative performance, and the quality of their uncertainty estimates has been overlooked. In this work we show that the discriminative performance of state-of-the-art CLIP adapters does not always correlate with their uncertainty estimation capabilities, which are essential for a safe deployment in real-world scenarios. We also demonstrate that one of such adapters is obtained through MAP inference from a more general probabilistic framework. Based on this observation we introduce BayesAdapter, which leverages Bayesian inference to estimate a full probability distribution instead of a single point, better capturing the variability inherent in the parameter space. In a comprehensive empirical evaluation we show that our approach obtains high quality uncertainty estimates in the predictions, standing out in calibration and selective classification. Our code will be publicly available upon acceptance of the paper.
Authors: Tony Chang, Kiarie Ndegwa, Andreas Gros, Vincent A. Landau, Luke J. Zachmann, Bogdan State, Mitchell A. Gritts, Colton W. Miller, Nathan E. Rutenbeck, Scott Conway, Guy Bayes
Abstract: This paper explores the application of a novel multi-task vision transformer (ViT) model for the estimation of canopy height models (CHMs) using 4-band National Agriculture Imagery Program (NAIP) imagery across the western United States. We compare the effectiveness of this model in terms of accuracy and precision aggregated across ecoregions and class heights versus three other benchmark peer-reviewed models. Key findings suggest that, while other benchmark models can provide high precision in localized areas, the VibrantVS model has substantial advantages across a broad reach of ecoregions in the western United States with higher accuracy, higher precision, the ability to generate updated inference at a cadence of three years or less, and high spatial resolution. The VibrantVS model provides significant value for ecological monitoring and land management decisions for wildfire mitigation.
Authors: Kun Guo, Qiang Ling
Abstract: Multi-camera 3D object detection aims to detect and localize objects in 3D space using multiple cameras, which has attracted more attention due to its cost-effectiveness trade-off. However, these methods often struggle with the lack of accurate depth estimation caused by the natural weakness of the camera in ranging. Recently, multi-modal fusion and knowledge distillation methods for 3D object detection have been proposed to solve this problem, which are time-consuming during the training phase and not friendly to memory cost. In light of this, we propose PromptDet, a lightweight yet effective 3D object detection framework motivated by the success of prompt learning in 2D foundation model. Our proposed framework, PromptDet, comprises two integral components: a general camera-based detection module, exemplified by models like BEVDet and BEVDepth, and a LiDAR-assisted prompter. The LiDAR-assisted prompter leverages the LiDAR points as a complementary signal, enriched with a minimal set of additional trainable parameters. Notably, our framework is flexible due to our prompt-like design, which can not only be used as a lightweight multi-modal fusion method but also as a camera-only method for 3D object detection during the inference phase. Extensive experiments on nuScenes validate the effectiveness of the proposed PromptDet. As a multi-modal detector, PromptDet improves the mAP and NDS by at most 22.8\% and 21.1\% with fewer than 2\% extra parameters compared with the camera-only baseline. Without LiDAR points, PromptDet still achieves an improvement of at most 2.4\% mAP and 4.0\% NDS with almost no impact on camera detection inference time.
Authors: Kangning Li, Zheyang Jia, Anyu Ying
Abstract: In recent years, large-scale models have achieved significant advancements, accompanied by the emergence of numerous high-quality benchmarks for evaluating various aspects of their comprehension abilities. However, most existing benchmarks primarily focus on spatial understanding in static image tasks. While some benchmarks extend evaluations to temporal tasks, they fall short in assessing text generation under complex contexts involving long videos and rich auxiliary information. To address this limitation, we propose a novel benchmark: the Multi-modal Story Generation Benchmark (MSBench), designed to evaluate text generation capabilities in scenarios enriched with auxiliary information. Our work introduces an innovative automatic dataset generation method to ensure the availability of accurate auxiliary information. On one hand, we leverage existing datasets and apply automated processes to generate new evaluation datasets, significantly reducing manual efforts. On the other hand, we refine auxiliary data through systematic filtering and utilize state-of-the-art models to ensure the fairness and accuracy of the ground-truth datasets. Our experiments reveal that current Multi-modal Large Language Models (MLLMs) perform suboptimally under the proposed evaluation metrics, highlighting significant gaps in their capabilities. To address these challenges, we propose a novel model architecture and methodology to better handle the overall process, demonstrating improvements on our benchmark.
Authors: Chen Duan, Qianyi Jiang, Pei Fu, Jiamin Chen, Shengxi Li, Zining Wang, Shan Guo, Junfeng Luo
Abstract: In the field of scene text spotting, previous OCR methods primarily relied on image encoders and pre-trained text information, but they often overlooked the advantages of incorporating human language instructions. To address this gap, we propose InstructOCR, an innovative instruction-based scene text spotting model that leverages human language instructions to enhance the understanding of text within images. Our framework employs both text and image encoders during training and inference, along with instructions meticulously designed based on text attributes. This approach enables the model to interpret text more accurately and flexibly. Extensive experiments demonstrate the effectiveness of our model and we achieve state-of-the-art results on widely used benchmarks. Furthermore, the proposed framework can be seamlessly applied to scene text VQA tasks. By leveraging instruction strategies during pre-training, the performance on downstream VQA tasks can be significantly improved, with a 2.6% increase on the TextVQA dataset and a 2.1% increase on the ST-VQA dataset. These experimental results provide insights into the benefits of incorporating human language instructions for OCR-related tasks.
Authors: Yue Guo, Haoxiang Liao, Haibin Ling, Bingyao Huang
Abstract: Underwater image restoration aims to remove geometric and color distortions due to water refraction, absorption and scattering. Previous studies focus on restoring either color or the geometry, but to our best knowledge, not both. However, in practice it may be cumbersome to address the two rectifications one-by-one. In this paper, we propose NeuroPump, a self-supervised method to simultaneously optimize and rectify underwater geometry and color as if water were pumped out. The key idea is to explicitly model refraction, absorption and scattering in Neural Radiance Field (NeRF) pipeline, such that it not only performs simultaneous geometric and color rectification, but also enables to synthesize novel views and optical effects by controlling the decoupled parameters. In addition, to address issue of lack of real paired ground truth images, we propose an underwater 360 benchmark dataset that has real paired (i.e., with and without water) images. Our method clearly outperforms other baselines both quantitatively and qualitatively. Our project page is available at: https://ygswu.github.io/NeuroPump.github.io/.
Authors: Jiale Huang, Dehong Gao, Jinxia Zhang, Zechao Zhan, Yang Hu, Xin Wang
Abstract: Large-scale Vision-Language Pre-training (VLP) has demonstrated remarkable success in the general domain. However, in the fashion domain, items are distinguished by fine-grained attributes like texture and material, which are crucial for tasks such as retrieval. Existing models often fail to leverage these fine-grained attributes from both text and image modalities. To address the above issues, we propose a novel approach for the fashion domain, Fine-grained Attributes Enhanced VLP (FashionFAE), which focuses on the detailed characteristics of fashion data. An attribute-emphasized text prediction task is proposed to predict fine-grained attributes of the items. This forces the model to focus on the salient attributes from the text modality. Additionally, a novel attribute-promoted image reconstruction task is proposed, which further enhances the fine-grained ability of the model by leveraging the representative attributes from the image modality. Extensive experiments show that FashionFAE significantly outperforms State-Of-The-Art (SOTA) methods, achieving 2.9% and 5.2% improvements in retrieval on sub-test and full test sets, respectively, and a 1.6% average improvement in recognition tasks.
Authors: Wenkun He, Yun Liu, Ruitao Liu, Li Yi
Abstract: Synthesizing realistic human-object interaction motions is a critical problem in VR/AR and human animation. Unlike the commonly studied scenarios involving a single human or hand interacting with one object, we address a more generic multi-body setting with arbitrary numbers of humans, hands, and objects. This complexity introduces significant challenges in synchronizing motions due to the high correlations and mutual influences among bodies. To address these challenges, we introduce SyncDiff, a novel method for multi-body interaction synthesis using a synchronized motion diffusion strategy. SyncDiff employs a single diffusion model to capture the joint distribution of multi-body motions. To enhance motion fidelity, we propose a frequency-domain motion decomposition scheme. Additionally, we introduce a new set of alignment scores to emphasize the synchronization of different body motions. SyncDiff jointly optimizes both data sample likelihood and alignment likelihood through an explicit synchronization strategy. Extensive experiments across four datasets with various multi-body configurations demonstrate the superiority of SyncDiff over existing state-of-the-art motion synthesis methods.
Authors: Chengjie Wang, Xi Jiang, Bin-Bin Gao, Zhenye Gan, Yong Liu, Feng Zheng, Lizhuang Ma
Abstract: Although mainstream unsupervised anomaly detection (AD) (including image-level classification and pixel-level segmentation)algorithms perform well in academic datasets, their performance is limited in practical application due to the ideal experimental setting of clean training data. Training with noisy data is an inevitable problem in real-world anomaly detection but is seldom discussed. This paper is the first to consider fully unsupervised industrial anomaly detection (i.e., unsupervised AD with noisy data). To solve this problem, we proposed memory-based unsupervised AD methods, SoftPatch and SoftPatch+, which efficiently denoise the data at the patch level. Noise discriminators are utilized to generate outlier scores for patch-level noise elimination before coreset construction. The scores are then stored in the memory bank to soften the anomaly detection boundary. Compared with existing methods, SoftPatch maintains a strong modeling ability of normal data and alleviates the overconfidence problem in coreset, and SoftPatch+ has more robust performance which is articularly useful in real-world industrial inspection scenarios with high levels of noise (from 10% to 40%). Comprehensive experiments conducted in diverse noise scenarios demonstrate that both SoftPatch and SoftPatch+ outperform the state-of-the-art AD methods on the MVTecAD, ViSA, and BTAD benchmarks. Furthermore, the performance of SoftPatch and SoftPatch+ is comparable to that of the noise-free methods in conventional unsupervised AD setting. The code of the proposed methods can be found at https://github.com/TencentYoutuResearch/AnomalyDetection-SoftPatch.
URLs: https://github.com/TencentYoutuResearch/AnomalyDetection-SoftPatch.
Authors: Zhiqiang Yuan, Ting Zhang, Ying Deng, Jiapei Zhang, Yeshuang Zhu, Zexi Jia, Jie Zhou, Jinchao Zhang
Abstract: Approximately 200 million individuals around the world suffer from varying degrees of visual impairment, making it crucial to leverage AI technology to offer walking assistance for these people. With the recent progress of vision-language models (VLMs), employing VLMs to improve this field has emerged as a popular research topic. However, most existing methods are studied on self-built question-answering datasets, lacking a unified training and testing benchmark for walk guidance. Moreover, in blind walking task, it is necessary to perform real-time streaming video parsing and generate concise yet informative reminders, which poses a great challenge for VLMs that suffer from redundant responses and low inference efficiency. In this paper, we firstly release a diverse, extensive, and unbiased walking awareness dataset, containing 12k video-manual annotation pairs from Europe and Asia to provide a fair training and testing benchmark for blind walking task. Furthermore, a WalkVLM model is proposed, which employs chain of thought for hierarchical planning to generate concise but informative reminders and utilizes temporal-aware adaptive prediction to reduce the temporal redundancy of reminders. Finally, we have established a solid benchmark for blind walking task and verified the advantages of WalkVLM in stream video processing for this task compared to other VLMs. Our dataset and code will be released at anonymous link https://walkvlm2024.github.io.
Authors: Nathanael L. Baisa
Abstract: In this work, we investigate four different fusion methods for associating detections to tracklets in multi-object visual tracking. In addition to considering strong cues such as motion and appearance information, we also consider weak cues such as height intersection-over-union (height-IoU) and tracklet confidence information in the data association using different fusion methods. These fusion methods include minimum, weighted sum based on IoU, Kalman filter (KF) gating, and hadamard product of costs due to the different cues. We conduct extensive evaluations on validation sets of MOT17, MOT20 and DanceTrack datasets, and find out that the choice of a fusion method is key for data association in multi-object visual tracking. We hope that this investigative work helps the computer vision research community to use the right fusion method for data association in multi-object visual tracking.
Authors: Hao Wang, Xiwen Chen, Ashish Bastola, Jiayou Qin, Abolfazl Razi
Abstract: The emergence of generative AI and controllable diffusion has made image-to-image synthesis increasingly practical and efficient. However, when input images exhibit low entropy and sparse, the inherent characteristics of diffusion models often result in limited diversity. This constraint significantly interferes with data augmentation. To address this, we propose Diffusion Prism, a training-free framework that efficiently transforms binary masks into realistic and diverse samples while preserving morphological features. We explored that a small amount of artificial noise will significantly assist the image-denoising process. To prove this novel mask-to-image concept, we use nano-dendritic patterns as an example to demonstrate the merit of our method compared to existing controllable diffusion models. Furthermore, we extend the proposed framework to other biological patterns, highlighting its potential applications across various fields.
Authors: Bohang Sun, Pietro Li\`o
Abstract: In this study, we introduce the Multi-Head Explainer (MHEX), a versatile and modular framework that enhances both the explainability and accuracy of Convolutional Neural Networks (CNNs) and Transformer-based models. MHEX consists of three core components: an Attention Gate that dynamically highlights task-relevant features, Deep Supervision that guides early layers to capture fine-grained details pertinent to the target class, and an Equivalent Matrix that unifies refined local and global representations to generate comprehensive saliency maps. Our approach demonstrates superior compatibility, enabling effortless integration into existing residual networks like ResNet and Transformer architectures such as BERT with minimal modifications. Extensive experiments on benchmark datasets in medical imaging and text classification show that MHEX not only improves classification accuracy but also produces highly interpretable and detailed saliency scores.
Authors: Deguo Xia, Weiming Zhang, Xiyan Liu, Wei Zhang, Chenting Gong, Xiao Tan, Jizhou Huang, Mengmeng Yang, Diange Yang
Abstract: An up-to-date city-scale lane-level map is an indispensable infrastructure and a key enabling technology for ensuring the safety and user experience of autonomous driving systems. In industrial scenarios, reliance on manual annotation for map updates creates a critical bottleneck. Lane-level updates require precise change information and must ensure consistency with adjacent data while adhering to strict standards. Traditional methods utilize a three-stage approach-construction, change detection, and updating-which often necessitates manual verification due to accuracy limitations. This results in labor-intensive processes and hampers timely updates. To address these challenges, we propose LDMapNet-U, which implements a new end-to-end paradigm for city-scale lane-level map updating. By reconceptualizing the update task as an end-to-end map generation process grounded in historical map data, we introduce a paradigm shift in map updating that simultaneously generates vectorized maps and change information. To achieve this, a Prior-Map Encoding (PME) module is introduced to effectively encode historical maps, serving as a critical reference for detecting changes. Additionally, we incorporate a novel Instance Change Prediction (ICP) module that learns to predict associations with historical maps. Consequently, LDMapNet-U simultaneously achieves vectorized map element generation and change detection. To demonstrate the superiority and effectiveness of LDMapNet-U, extensive experiments are conducted using large-scale real-world datasets. In addition, LDMapNet-U has been successfully deployed in production at Baidu Maps since April 2024, supporting map updating for over 360 cities and significantly shortening the update cycle from quarterly to weekly. The updated maps serve hundreds of millions of users and are integrated into the autonomous driving systems of several leading vehicle companies.
Authors: Xuyang Wang, Ziang Cheng, Zhenyu Li, Jiayu Yang, Haorui Ji, Pan Ji, Mehrtash Harandi, Richard Hartley, Hongdong Li
Abstract: This paper proposes DoubleDiffusion, a novel framework that combines heat dissipation diffusion and denoising diffusion for direct generative learning on 3D mesh surfaces. Our approach addresses the challenges of generating continuous signal distributions residing on a curve manifold surface. Unlike previous methods that rely on unrolling 3D meshes into 2D or adopting field representations, DoubleDiffusion leverages the Laplacian-Beltrami operator to process features respecting the mesh structure. This combination enables effective geometry-aware signal diffusion across the underlying geometry. As shown in Fig.1, we demonstrate that DoubleDiffusion has the ability to generate RGB signal distributions on complex 3D mesh surfaces and achieves per-category shape-conditioned texture generation across different shape geometry. Our work contributes a new direction in diffusion-based generative modeling on 3D surfaces, with potential applications in the field of 3D asset generation.
Authors: Ngoc Dung Huynh, Mohamed Reda Bouadjenek, Sunil Aryal, Imran Razzak, Hakim Hacid
Abstract: Visual Question Answering (VQA) is an evolving research field aimed at enabling machines to answer questions about visual content by integrating image and language processing techniques such as feature extraction, object detection, text embedding, natural language understanding, and language generation. With the growth of multimodal data research, VQA has gained significant attention due to its broad applications, including interactive educational tools, medical image diagnosis, customer service, entertainment, and social media captioning. Additionally, VQA plays a vital role in assisting visually impaired individuals by generating descriptive content from images. This survey introduces a taxonomy of VQA architectures, categorizing them based on design choices and key components to facilitate comparative analysis and evaluation. We review major VQA approaches, focusing on deep learning-based methods, and explore the emerging field of Large Visual Language Models (LVLMs) that have demonstrated success in multimodal tasks like VQA. The paper further examines available datasets and evaluation metrics essential for measuring VQA system performance, followed by an exploration of real-world VQA applications. Finally, we highlight ongoing challenges and future directions in VQA research, presenting open questions and potential areas for further development. This survey serves as a comprehensive resource for researchers and practitioners interested in the latest advancements and future
Authors: Wisdom O. Ikezogwo, Kevin Zhang, Mehmet Saygin Seyfioglu, Fatemeh Ghezloo, Linda Shapiro, Ranjay Krishna
Abstract: We propose MedicalNarratives, a dataset curated from medical pedagogical videos similar in nature to data collected in Think-Aloud studies and inspired by Localized Narratives, which collects grounded image-text data by curating instructors' speech and mouse cursor movements synchronized in time. MedicalNarratives enables pretraining of both semantic and dense objectives, alleviating the need to train medical semantic and dense tasks disparately due to the lack of reasonably sized datasets. Our dataset contains 4.7M image-text pairs from videos and articles, with 1M samples containing dense annotations in the form of traces and bounding boxes. To evaluate the utility of MedicalNarratives, we train GenMedClip based on the CLIP architecture using our dataset spanning 12 medical domains and demonstrate that it outperforms previous state-of-the-art models on a newly constructed medical imaging benchmark that comprehensively evaluates performance across all modalities. Data, demo, code and models available at https://medical-narratives.github.io
Authors: Hafiz Mughees Ahmad, Dario Morle, Afshin Rahimi
Abstract: Deep learning models have demonstrated remarkable performance across various computer vision tasks, yet their vulnerability to distribution shifts remains a critical challenge. Despite sophisticated neural network architectures, existing models often struggle to maintain consistent performance when confronted with Out-of-Distribution (OOD) samples, including natural corruptions, adversarial perturbations, and anomalous patterns. We introduce LayerMix, an innovative data augmentation approach that systematically enhances model robustness through structured fractal-based image synthesis. By meticulously integrating structural complexity into training datasets, our method generates semantically consistent synthetic samples that significantly improve neural network generalization capabilities. Unlike traditional augmentation techniques that rely on random transformations, LayerMix employs a structured mixing pipeline that preserves original image semantics while introducing controlled variability. Extensive experiments across multiple benchmark datasets, including CIFAR-10, CIFAR-100, ImageNet-200, and ImageNet-1K demonstrate LayerMixs superior performance in classification accuracy and substantially enhances critical Machine Learning (ML) safety metrics, including resilience to natural image corruptions, robustness against adversarial attacks, improved model calibration and enhanced prediction consistency. LayerMix represents a significant advancement toward developing more reliable and adaptable artificial intelligence systems by addressing the fundamental challenges of deep learning generalization. The code is available at https://github.com/ahmadmughees/layermix.
Authors: Ludwic Leonard, Nils Thuerey, Ruediger Westermann
Abstract: We introduce a single-view reconstruction technique of volumetric fields in which multiple light scattering effects are omnipresent, such as in clouds. We model the unknown distribution of volumetric fields using an unconditional diffusion model trained on a novel benchmark dataset comprising 1,000 synthetically simulated volumetric density fields. The neural diffusion model is trained on the latent codes of a novel, diffusion-friendly, monoplanar representation. The generative model is used to incorporate a tailored parametric diffusion posterior sampling technique into different reconstruction tasks. A physically-based differentiable volume renderer is employed to provide gradients with respect to light transport in the latent space. This stands in contrast to classic NeRF approaches and makes the reconstructions better aligned with observed data. Through various experiments, we demonstrate single-view reconstruction of volumetric clouds at a previously unattainable quality.
Authors: Jiaxuan Peng, Mengshi Qi, Dong Zhao, Huadong Ma
Abstract: 3D human pose estimation (3D HPE) has emerged as a prominent research topic, particularly in the realm of RGB-based methods. However, RGB images are susceptible to limitations such as sensitivity to lighting conditions and potential user discomfort. Consequently, multi-modal sensing, which leverages non-intrusive sensors, is gaining increasing attention. Nevertheless, multi-modal 3D HPE still faces challenges, including modality imbalance and the imperative for continual learning. In this work, we introduce a novel balanced continual multi-modal learning method for 3D HPE, which harnesses the power of RGB, LiDAR, mmWave, and WiFi. Specifically, we propose a Shapley value-based contribution algorithm to quantify the contribution of each modality and identify modality imbalance. To address this imbalance, we employ a re-learning strategy. Furthermore, recognizing that raw data is prone to noise contamination, we develop a novel denoising continual learning approach. This approach incorporates a noise identification and separation module to mitigate the adverse effects of noise and collaborates with the balanced learning strategy to enhance optimization. Additionally, an adaptive EWC mechanism is employed to alleviate catastrophic forgetting. We conduct extensive experiments on the widely-adopted multi-modal dataset, MM-Fi, which demonstrate the superiority of our approach in boosting 3D pose estimation and mitigating catastrophic forgetting in complex scenarios. We will release our codes.
Authors: Dimitrios Gerogiannis, Foivos Paraperas Papantoniou, Rolandos Alexandros Potamias, Alexandros Lattas, Stefanos Zafeiriou
Abstract: Inspired by the effectiveness of 3D Gaussian Splatting (3DGS) in reconstructing detailed 3D scenes within multi-view setups and the emergence of large 2D human foundation models, we introduce Arc2Avatar, the first SDS-based method utilizing a human face foundation model as guidance with just a single image as input. To achieve that, we extend such a model for diverse-view human head generation by fine-tuning on synthetic data and modifying its conditioning. Our avatars maintain a dense correspondence with a human face mesh template, allowing blendshape-based expression generation. This is achieved through a modified 3DGS approach, connectivity regularizers, and a strategic initialization tailored for our task. Additionally, we propose an optional efficient SDS-based correction step to refine the blendshape expressions, enhancing realism and diversity. Experiments demonstrate that Arc2Avatar achieves state-of-the-art realism and identity preservation, effectively addressing color issues by allowing the use of very low guidance, enabled by our strong identity prior and initialization strategy, without compromising detail. Please visit https://arc2avatar.github.io for more resources.
Authors: Ziheng Wu, Zhenghao Chen, Ruipu Luo, Can Zhang, Yuan Gao, Zhentao He, Xian Wang, Haoran Lin, Minghui Qiu
Abstract: Recently, vision-language models have made remarkable progress, demonstrating outstanding capabilities in various tasks such as image captioning and video understanding. We introduce Valley2, a novel multimodal large language model designed to enhance performance across all domains and extend the boundaries of practical applications in e-commerce and short video scenarios. Notably, Valley2 achieves state-of-the-art (SOTA) performance on e-commerce benchmarks, surpassing open-source models of similar size by a large margin (79.66 vs. 72.76). Additionally, Valley2 ranks second on the OpenCompass leaderboard among models with fewer than 10B parameters, with an impressive average score of 67.4. The code and model weights are open-sourced at https://github.com/bytedance/Valley.
Authors: Yinghao Zhang, Peng Li, Yue Hu
Abstract: The tensor low-rank prior has attracted considerable attention in dynamic MR reconstruction. Tensor low-rank methods preserve the inherent high-dimensional structure of data, allowing for improved extraction and utilization of intrinsic low-rank characteristics. However, most current methods are still confined to utilizing low-rank structures either in the image domain or predefined transformed domains. Designing an optimal transformation adaptable to dynamic MRI reconstruction through manual efforts is inherently challenging. In this paper, we propose a deep unrolling network that utilizes the convolutional neural network (CNN) to adaptively learn the transformed domain for leveraging tensor low-rank priors. Under the supervised mechanism, the learning of the tensor low-rank domain is directly guided by the reconstruction accuracy. Specifically, we generalize the traditional t-SVD to a transformed version based on arbitrary high-dimensional unitary transformations and introduce a novel unitary transformed tensor nuclear norm (UTNN). Subsequently, we present a dynamic MRI reconstruction model based on UTNN and devise an efficient iterative optimization algorithm using ADMM, which is finally unfolded into the proposed T2LR-Net. Experiments on two dynamic cardiac MRI datasets demonstrate that T2LR-Net outperforms the state-of-the-art optimization-based and unrolling network-based methods.
Authors: John Chiang
Abstract: In this paper, we present a practical solution to implement privacy-preserving CNN training based on mere Homomorphic Encryption (HE) technique. To our best knowledge, this is the first attempt successfully to crack this nut and no work ever before has achieved this goal. Several techniques combine to accomplish the task:: (1) with transfer learning, privacy-preserving CNN training can be reduced to homomorphic neural network training, or even multiclass logistic regression (MLR) training; (2) via a faster gradient variant called $\texttt{Quadratic Gradient}$, an enhanced gradient method for MLR with a state-of-the-art performance in convergence speed is applied in this work to achieve high performance; (3) we employ the thought of transformation in mathematics to transform approximating Softmax function in the encryption domain to the approximation of the Sigmoid function. A new type of loss function termed $\texttt{Squared Likelihood Error}$ has been developed alongside to align with this change.; and (4) we use a simple but flexible matrix-encoding method named $\texttt{Volley Revolver}$ to manage the data flow in the ciphertexts, which is the key factor to complete the whole homomorphic CNN training. The complete, runnable C++ code to implement our work can be found at: \href{https://github.com/petitioner/HE.CNNtraining}{$\texttt{https://github.com/petitioner/HE.CNNtraining}$}. We select $\texttt{REGNET\_X\_400MF}$ as our pre-trained model for transfer learning. We use the first 128 MNIST training images as training data and the whole MNIST testing dataset as the testing data. The client only needs to upload 6 ciphertexts to the cloud and it takes $\sim 21$ mins to perform 2 iterations on a cloud with 64 vCPUs, resulting in a precision of $21.49\%$.
URLs: https://github.com/petitioner/HE.CNNtraining, https://github.com/petitioner/HE.CNNtraining
Authors: Xinyuan Song
Abstract: With the spread of COVID-19 around the globe over the past year, the usage of artificial intelligence (AI) algorithms and image processing methods to analyze the X-ray images of patients' chest with COVID-19 has become essential. The COVID-19 virus recognition in the lung area of a patient is one of the basic and essential needs of clicical centers and hospitals. Most research in this field has been devoted to papers on the basis of deep learning methods utilizing CNNs (Convolutional Neural Network), which mainly deal with the screening of sick and healthy people.In this study, a new structure of a 19-layer CNN has been recommended for accurately recognition of the COVID-19 from the X-ray pictures of chest. The offered CNN is developed to serve as a precise diagnosis system for a three class (viral pneumonia, Normal, COVID) and a four classclassification (Lung opacity, Normal, COVID-19, and pneumonia). A comparison is conducted among the outcomes of the offered procedure and some popular pretrained networks, including Inception, Alexnet, ResNet50, Squeezenet, and VGG19 and based on Specificity, Accuracy, Precision, Sensitivity, Confusion Matrix, and F1-score. The experimental results of the offered CNN method specify its dominance over the existing published procedures. This method can be a useful tool for clinicians in deciding properly about COVID-19.
Authors: Sahil Verma, Gantavya Bhatt, Avi Schwarzschild, Soumye Singhal, Arnav Mohanty Das, Chirag Shah, John P Dickerson, Pin-Yu Chen, Jeff Bilmes
Abstract: Despite the advanced capabilities of contemporary machine learning (ML) models, they remain vulnerable to adversarial and backdoor attacks. This vulnerability is particularly concerning in real-world deployments, where compromised models may exhibit unpredictable behavior in critical scenarios. Such risks are heightened by the prevalent practice of collecting massive, internet-sourced datasets for training multimodal models, as these datasets may harbor backdoors. Various techniques have been proposed to mitigate the effects of backdooring in multimodal models, such as CleanCLIP, which is the current state-of-the-art approach. In this work, we demonstrate that the efficacy of CleanCLIP in mitigating backdoors is highly dependent on the particular objective used during model pre-training. We observe that stronger pre-training objectives that lead to higher zero-shot classification performance correlate with harder to remove backdoors behaviors. We show this by training multimodal models on two large datasets consisting of 3 million (CC3M) and 6 million (CC6M) datapoints, under various pre-training objectives, followed by poison removal using CleanCLIP. We find that CleanCLIP, even with extensive hyperparameter tuning, is ineffective in poison removal when stronger pre-training objectives are used. Our findings underscore critical considerations for ML practitioners who train models using large-scale web-curated data and are concerned about potential backdoor threats.
Authors: Anirudh Prabhakaran, YeKun Xiao, Ching-Yu Cheng, Dianbo Liu
Abstract: Ocular diseases, including diabetic retinopathy and glaucoma, present a significant public health challenge due to their high prevalence and potential for causing vision impairment. Early and accurate diagnosis is crucial for effective treatment and management. In recent years, deep learning models have emerged as powerful tools for analysing medical images, such as retina imaging. However, challenges persist in model relibability and uncertainty estimation, which are critical for clinical decision-making. This study leverages the probabilistic framework of Generative Flow Networks (GFlowNets) to learn the posterior distribution over latent discrete dropout masks for the classification and analysis of ocular diseases using fundus images. We develop a robust and generalizable method that utilizes GFlowOut integrated with ResNet18 and ViT models as the backbone in identifying various ocular conditions. This study employs a unique set of dropout masks - none, random, bottomup, and topdown - to enhance model performance in analyzing these fundus images. Our results demonstrate that our learnable probablistic latents significantly improves accuracy, outperforming the traditional dropout approach. We utilize a gradient map calculation method, Grad-CAM, to assess model explainability, observing that the model accurately focuses on critical image regions for predictions. The integration of GFlowOut in neural networks presents a promising advancement in the automated diagnosis of ocular diseases, with implications for improving clinical workflows and patient outcomes.
Authors: Xinjie Zhang, Shenyuan Gao, Zhening Liu, Jiawei Shao, Xingtong Ge, Dailan He, Tongda Xu, Yan Wang, Jun Zhang
Abstract: Existing learning-based stereo image codec adopt sophisticated transformation with simple entropy models derived from single image codecs to encode latent representations. However, those entropy models struggle to effectively capture the spatial-disparity characteristics inherent in stereo images, which leads to suboptimal rate-distortion results. In this paper, we propose a stereo image compression framework, named CAMSIC. CAMSIC independently transforms each image to latent representation and employs a powerful decoder-free Transformer entropy model to capture both spatial and disparity dependencies, by introducing a novel content-aware masked image modeling (MIM) technique. Our content-aware MIM facilitates efficient bidirectional interaction between prior information and estimated tokens, which naturally obviates the need for an extra Transformer decoder. Experiments show that our stereo image codec achieves state-of-the-art rate-distortion performance on two stereo image datasets Cityscapes and InStereo2K with fast encoding and decoding speed. Code is available at https://github.com/Xinjie-Q/CAMSIC.
Authors: Zewen Xu, Yijia He, Hao Wei, Bo Xu, BinJian Xie, Yihong Wu
Abstract: Line features are valid complements for point features in man-made environments. 3D-2D constraints provided by line features have been widely used in Visual Odometry (VO) and Structure-from-Motion (SfM) systems. However, how to accurately solve three-view relative motion only with 2D observations of points and lines in real time has not been fully explored. In this paper, we propose a novel three-view pose solver based on rotation-translation decoupled estimation. First, a high-precision rotation estimation method based on normal vector coplanarity constraints that consider the uncertainty of observations is proposed, which can be solved by Levenberg-Marquardt (LM) algorithm efficiently. Second, a robust linear translation constraint that minimizes the degree of the rotation components and feature observation components in equations is elaborately designed for estimating translations accurately. Experiments on synthetic data and real-world data show that the proposed approach improves both rotation and translation accuracy compared to the classical trifocal-tensor-based method and the state-of-the-art two-view algorithm in outdoor and indoor environments.
Authors: Billel Essaid, Hamza Kheddar, Noureddine Batel, Muhammad E. H. Chowdhury, Abderrahmane Lakas
Abstract: Automatic speech recognition (ASR) plays a pivotal role in our daily lives, offering utility not only for interacting with machines but also for facilitating communication for individuals with partial or profound hearing impairments. The process involves receiving the speech signal in analog form, followed by various signal processing algorithms to make it compatible with devices of limited capacities, such as cochlear implants (CIs). Unfortunately, these implants, equipped with a finite number of electrodes, often result in speech distortion during synthesis. Despite efforts by researchers to enhance received speech quality using various state-of-the-art (SOTA) signal processing techniques, challenges persist, especially in scenarios involving multiple sources of speech, environmental noise, and other adverse conditions. The advent of new artificial intelligence (AI) methods has ushered in cutting-edge strategies to address the limitations and difficulties associated with traditional signal processing techniques dedicated to CIs. This review aims to comprehensively cover advancements in CI-based ASR and speech enhancement, among other related aspects. The primary objective is to provide a thorough overview of metrics and datasets, exploring the capabilities of AI algorithms in this biomedical field, and summarizing and commenting on the best results obtained. Additionally, the review will delve into potential applications and suggest future directions to bridge existing research gaps in this domain.
Authors: Jaeill Kim, Wonseok Lee, Moonjung Eo, Wonjong Rhee
Abstract: Class Incremental Learning (CIL) constitutes a pivotal subfield within continual learning, aimed at enabling models to progressively learn new classification tasks while retaining knowledge obtained from prior tasks. Although previous studies have predominantly focused on backward compatible approaches to mitigate catastrophic forgetting, recent investigations have introduced forward compatible methods to enhance performance on novel tasks and complement existing backward compatible methods. In this study, we introduce an effective-Rank based Feature Richness enhancement (RFR) method, designed for improving forward compatibility. Specifically, this method increases the effective rank of representations during the base session, thereby facilitating the incorporation of more informative features pertinent to unseen novel tasks. Consequently, RFR achieves dual objectives in backward and forward compatibility: minimizing feature extractor modifications and enhancing novel task performance, respectively. To validate the efficacy of our approach, we establish a theoretical connection between effective rank and the Shannon entropy of representations. Subsequently, we conduct comprehensive experiments by integrating RFR into eleven well-known CIL methods. Our results demonstrate the effectiveness of our approach in enhancing novel-task performance while mitigating catastrophic forgetting. Furthermore, our method notably improves the average incremental accuracy across all eleven cases examined.
Authors: Siddarth Venkatraman, Moksh Jain, Luca Scimeca, Minsu Kim, Marcin Sendera, Mohsin Hasan, Luke Rowe, Sarthak Mittal, Pablo Lemos, Emmanuel Bengio, Alexandre Adam, Jarrid Rector-Brooks, Yoshua Bengio, Glen Berseth, Nikolay Malkin
Abstract: Diffusion models have emerged as effective distribution estimators in vision, language, and reinforcement learning, but their use as priors in downstream tasks poses an intractable posterior inference problem. This paper studies amortized sampling of the posterior over data, $\mathbf{x}\sim p^{\rm post}(\mathbf{x})\propto p(\mathbf{x})r(\mathbf{x})$, in a model that consists of a diffusion generative model prior $p(\mathbf{x})$ and a black-box constraint or likelihood function $r(\mathbf{x})$. We state and prove the asymptotic correctness of a data-free learning objective, relative trajectory balance, for training a diffusion model that samples from this posterior, a problem that existing methods solve only approximately or in restricted cases. Relative trajectory balance arises from the generative flow network perspective on diffusion models, which allows the use of deep reinforcement learning techniques to improve mode coverage. Experiments illustrate the broad potential of unbiased inference of arbitrary posteriors under diffusion priors: in vision (classifier guidance), language (infilling under a discrete diffusion LLM), and multimodal data (text-to-image generation). Beyond generative modeling, we apply relative trajectory balance to the problem of continuous control with a score-based behavior prior, achieving state-of-the-art results on benchmarks in offline reinforcement learning.
Authors: Ziqiang Liu, Feiteng Fang, Xi Feng, Xinrun Du, Chenhao Zhang, Zekun Wang, Yuelin Bai, Qixuan Zhao, Liyang Fan, Chengguang Gan, Hongquan Lin, Jiaming Li, Yuansheng Ni, Haihong Wu, Yaswanth Narsupalli, Zhigang Zheng, Chengming Li, Xiping Hu, Ruifeng Xu, Xiaojun Chen, Min Yang, Jiaheng Liu, Ruibo Liu, Wenhao Huang, Ge Zhang, Shiwen Ni
Abstract: The rapid advancements in the development of multimodal large language models (MLLMs) have consistently led to new breakthroughs on various benchmarks. In response, numerous challenging and comprehensive benchmarks have been proposed to more accurately assess the capabilities of MLLMs. However, there is a dearth of exploration of the higher-order perceptual capabilities of MLLMs. To fill this gap, we propose the Image Implication understanding Benchmark, II-Bench, which aims to evaluate the model's higher-order perception of images. Through extensive experiments on II-Bench across multiple MLLMs, we have made significant findings. Initially, a substantial gap is observed between the performance of MLLMs and humans on II-Bench. The pinnacle accuracy of MLLMs attains 74.8%, whereas human accuracy averages 90%, peaking at an impressive 98%. Subsequently, MLLMs perform worse on abstract and complex images, suggesting limitations in their ability to understand high-level semantics and capture image details. Finally, it is observed that most models exhibit enhanced accuracy when image sentiment polarity hints are incorporated into the prompts. This observation underscores a notable deficiency in their inherent understanding of image sentiment. We believe that II-Bench will inspire the community to develop the next generation of MLLMs, advancing the journey towards expert artificial general intelligence (AGI). II-Bench is publicly available at https://huggingface.co/datasets/m-a-p/II-Bench.
Authors: Mehmet Can Yavuz, Yang Yang
Abstract: Deep learning classifiers face significant challenges when dealing with heterogeneous multi-modal and multi-organ biomedical datasets. The low-level feature distinguishability limited to imaging-modality hinders the classifiers' ability to learn high-level semantic relationships, resulting in sub-optimal performance. To address this issue, image augmentation strategies are employed as regularization techniques. While additive noise input during network training is a well-established augmentation as regularization method, modern pipelines often favor more robust techniques such as dropout and weight decay. This preference stems from the observation that combining these established techniques with noise input can adversely affect model performance. In this study, we propose a novel pretraining pipeline that learns to generate conditional noise mask specifically tailored to improve performance on multi-modal and multi-organ datasets. As a reinforcement learning algorithm, our approach employs a dual-component system comprising a very light-weight policy network that learns to sample conditional noise using a differentiable beta distribution as well as a classifier network. The policy network is trained using the reinforce algorithm to generate image-specific noise masks that regularize the classifier during pretraining. A key aspect is that the policy network's role is limited to obtaining an intermediate (or heated) model before fine-tuning. During inference, the policy network is omitted, allowing direct comparison between the baseline and noise-regularized models. We conducted experiments and related analyses on RadImageNet datasets. Results demonstrate that fine-tuning the intermediate models consistently outperforms conventional training algorithms on both classification and generalization to unseen concept tasks.
Authors: Sao Mai Nguyen, Maxime Devanne, Olivier Remy-Neris, Mathieu Lempereur, Andr\'e Thepaut
Abstract: While automatic monitoring and coaching of exercises are showing encouraging results in non-medical applications, they still have limitations such as errors and limited use contexts. To allow the development and assessment of physical rehabilitation by an intelligent tutoring system, we identify in this article four challenges to address and propose a medical dataset of clinical patients carrying out low back-pain rehabilitation exercises. The dataset includes 3D Kinect skeleton positions and orientations, RGB videos, 2D skeleton data, and medical annotations to assess the correctness, and error classification and localisation of body part and timespan. Along this dataset, we perform a complete research path, from data collection to processing, and finally a small benchmark. We evaluated on the dataset two baseline movement recognition algorithms, pertaining to two different approaches: the probabilistic approach with a Gaussian Mixture Model (GMM), and the deep learning approach with a Long-Short Term Memory (LSTM). This dataset is valuable because it includes rehabilitation relevant motions in a clinical setting with patients in their rehabilitation program, using a cost-effective, portable, and convenient sensor, and because it shows the potential for improvement on these challenges.
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: Jamie Burke, Justin Engelmann, Samuel Gibbon, Charlene Hamid, Diana Moukaddem, Dan Pugh, Tariq Farrah, Niall Strang, Neeraj Dhaun, Tom MacGillivray, Stuart King, Ian J. C. MacCormick
Abstract: Optical coherence tomography (OCT) and scanning laser ophthalmoscopy (SLO) of the eye has become essential to ophthalmology and the emerging field of oculomics, thus requiring a need for transparent, reproducible, and rapid analysis of this data for clinical research and the wider research community. Here, we introduce OCTolyzer, the first open-source toolkit for retinochoroidal analysis in OCT/SLO data. It features two analysis suites for OCT and SLO data, facilitating deep learning-based anatomical segmentation and feature extraction of the cross-sectional retinal and choroidal layers and en face retinal vessels. We describe OCTolyzer and evaluate the reproducibility of its OCT choroid analysis. At the population level, metrics for choroid region thickness were highly reproducible, with a mean absolute error (MAE)/Pearson correlation for macular volume choroid thickness (CT) of 6.7$\mu$m/0.99, macular B-scan CT of 11.6$\mu$m/0.99, and peripapillary CT of 5.0$\mu$m/0.99. Macular choroid vascular index (CVI) also showed strong reproducibility, with MAE/Pearson for volume CVI yielding 0.0271/0.97 and B-scan CVI 0.0130/0.91. At the eye level, measurement noise for regional and vessel metrics was below 5% and 20% of the population's variability, respectively. Outliers were caused by poor-quality B-scans with thick choroids and invisible choroid-sclera boundary. Processing times on a laptop CPU were under three seconds for macular/peripapillary B-scans and 85 seconds for volume scans. OCTolyzer can convert OCT/SLO data into reproducible and clinically meaningful retinochoroidal features and will improve the standardisation of ocular measurements in OCT/SLO image analysis, requiring no specialised training or proprietary software to be used. OCTolyzer is freely available here: https://github.com/jaburke166/OCTolyzer.
Authors: Hongbo Liu
Abstract: Contrastive Language-Image Pre-training (CLIP) has attracted a surge of attention for its superior zero-shot performance and excellent transferability to downstream tasks. However, training such large-scale models usually requires substantial computation and storage, which poses barriers for general users with consumer-level computers. Motivated by this observation, in this paper we investigate how to achieve competitive performance on only one Nvidia RTX3090 GPU and with one terabyte for storing dataset. On one hand, we simplify the transformer block structure and combine Weight Inheritance with multi-stage Knowledge Distillation (WIKD), thereby reducing the parameters and improving the inference speed during training along with deployment. On the other hand, confronted with the convergence challenge posed by small dataset, we generate synthetic captions for each sample as data augmentation, and devise a novel Pair Matching (PM) loss to fully exploit the distinguishment among positive and negative image-text pairs. Extensive experiments demonstrate that our model can achieve a new state-of-the-art datascale-parameter-accuracy tradeoff, which could further popularize the CLIP model in the related research community.
Authors: Navid Salami Pargoo, Mahshid Ghasemi, Shuren Xia, Mehmet Kerem Turkcan, Taqiya Ehsan, Chengbo Zang, Yuan Sun, Javad Ghaderi, Gil Zussman, Zoran Kostic, Jorge Ortiz
Abstract: As urban populations grow, cities are becoming more complex, driving the deployment of interconnected sensing systems to realize the vision of smart cities. These systems aim to improve safety, mobility, and quality of life through applications that integrate diverse sensors with real-time decision-making. Streetscape applications-focusing on challenges like pedestrian safety and adaptive traffic management-depend on managing distributed, heterogeneous sensor data, aligning information across time and space, and enabling real-time processing. These tasks are inherently complex and often difficult to scale. The Streetscape Application Services Stack (SASS) addresses these challenges with three core services: multimodal data synchronization, spatiotemporal data fusion, and distributed edge computing. By structuring these capabilities as clear, composable abstractions with clear semantics, SASS allows developers to scale streetscape applications efficiently while minimizing the complexity of multimodal integration. We evaluated SASS in two real-world testbed environments: a controlled parking lot and an urban intersection in a major U.S. city. These testbeds allowed us to test SASS under diverse conditions, demonstrating its practical applicability. The Multimodal Data Synchronization service reduced temporal misalignment errors by 88%, achieving synchronization accuracy within 50 milliseconds. Spatiotemporal Data Fusion service improved detection accuracy for pedestrians and vehicles by over 10%, leveraging multicamera integration. The Distributed Edge Computing service increased system throughput by more than an order of magnitude. Together, these results show how SASS provides the abstractions and performance needed to support real-time, scalable urban applications, bridging the gap between sensing infrastructure and actionable streetscape intelligence.
Authors: Patrick Styll, Dowon Kim, Jiook Cha
Abstract: Brain development in the first few months of human life is a critical phase characterized by rapid structural growth and functional organization. Accurately predicting developmental outcomes during this time is crucial for identifying delays and enabling timely interventions. This study introduces the SwiFT (Swin 4D fMRI Transformer) model, designed to predict Bayley-III composite scores using neonatal fMRI data from the Developing Human Connectome Project (dHCP). To enhance predictive accuracy, we apply dimensionality reduction via group independent component analysis (ICA) and pretrain SwiFT on large adult fMRI datasets to address the challenges of limited neonatal data. Our analysis shows that SwiFT significantly outperforms baseline models in predicting cognitive, motor, and language outcomes, leveraging both single-label and multi-label prediction strategies. The model's attention-based architecture processes spatiotemporal data end-to-end, delivering superior predictive performance. Additionally, we use Integrated Gradients with Smoothgrad sQuare (IG-SQ) to interpret predictions, identifying neural spatial representations linked to early cognitive and behavioral development. These findings underscore the potential of Transformer models to advance neurodevelopmental research and clinical practice.
Authors: Ayush Deshmukh
Abstract: The global outbreak of the Mpox virus, classified as a Public Health Emergency of International Concern (PHEIC) by the World Health Organization, presents significant diagnostic challenges due to its visual similarity to other skin lesion diseases. Traditional diagnostic methods for Mpox, which rely on clinical symptoms and laboratory tests, are slow and labor intensive. Deep learning-based approaches for skin lesion classification offer a promising alternative. However, developing a model that balances efficiency with accuracy is crucial to ensure reliable and timely diagnosis without compromising performance. This study introduces the Cascaded Atrous Group Attention (CAGA) framework to address these challenges, combining the Cascaded Atrous Attention module and the Cascaded Group Attention mechanism. The Cascaded Atrous Attention module utilizes dilated convolutions and cascades the outputs to enhance multi-scale representation. This is integrated into the Cascaded Group Attention mechanism, which reduces redundancy in Multi-Head Self-Attention. By integrating the Cascaded Atrous Group Attention module with EfficientViT-L1 as the backbone architecture, this approach achieves state-of-the-art performance, reaching an accuracy of 98% on the Mpox Close Skin Image (MCSI) dataset while reducing model parameters by 37.5% compared to the original EfficientViT-L1. The model's robustness is demonstrated through extensive validation on two additional benchmark datasets, where it consistently outperforms existing approaches.
Authors: Manman Yuan, Weiming Jia, Xiong Luo, Jiazhen Ye, Peican Zhu, Junlin Li
Abstract: The precise detection of mild cognitive impairment (MCI) is of significant importance in preventing the deterioration of patients in a timely manner. Although hypergraphs have enhanced performance by learning and analyzing brain networks, they often only depend on vector distances between features at a single scale to infer interactions. In this paper, we deal with a more arduous challenge, hypergraph modelling with synchronization between brain regions, and design a novel framework, i.e., A Multi-scale Hypergraph Network for MCI Detection via Synchronous and Attentive Fusion (MHSA), to tackle this challenge. Specifically, our approach employs the Phase-Locking Value (PLV) to calculate the phase synchronization relationship in the spectrum domain of regions of interest (ROIs) and designs a multi-scale feature fusion mechanism to integrate dynamic connectivity features of functional magnetic resonance imaging (fMRI) from both the temporal and spectrum domains. To evaluate and optimize the direct contribution of each ROI to phase synchronization in the temporal domain, we structure the PLV coefficients dynamically adjust strategy, and the dynamic hypergraph is modelled based on a comprehensive temporal-spectrum fusion matrix. Experiments on the real-world dataset indicate the effectiveness of our strategy. The code is available at https://github.com/Jia-Weiming/MHSA.
Authors: Zhongyi Zhou, Yaxin Peng, Pin Yi, Minjie Zhu, Chaomin Shen
Abstract: Continual Learning enables models to learn and adapt to new tasks while retaining prior knowledge. Introducing new tasks, however, can naturally lead to feature entanglement across tasks, limiting the model's capability to distinguish between new domain data. In this work, we propose a method called Feature Realignment through Experts on hyperSpHere in Continual Learning (Fresh-CL). By leveraging predefined and fixed simplex equiangular tight frame (ETF) classifiers on a hypersphere, our model improves feature separation both intra and inter tasks. However, the projection to a simplex ETF shifts with new tasks, disrupting structured feature representation of previous tasks and degrading performance. Therefore, we propose a dynamic extension of ETF through mixture of experts, enabling adaptive projections onto diverse subspaces to enhance feature representation. Experiments on 11 datasets demonstrate a 2% improvement in accuracy compared to the strongest baseline, particularly in fine-grained datasets, confirming the efficacy of combining ETF and MoE to improve feature distinction in continual learning scenarios.
Authors: Xianhao Zhou, Jianghao Wu, Huangxuan Zhao, Lei Chen, Shaoting Zhang, Guotai Wang
Abstract: Generating synthetic Computed Tomography (CT) images from Cone Beam Computed Tomography (CBCT) is desirable for improving the image quality of CBCT. Existing synthetic CT (sCT) generation methods using Convolutional Neural Networks (CNN) and Transformers often face difficulties in effectively capturing both global and local features and contrasts for high-quality sCT generation. In this work, we propose a Global-Local Feature and Contrast learning (GLFC) framework for sCT generation. First, a Mamba-Enhanced UNet (MEUNet) is introduced by integrating Mamba blocks into the skip connections of a high-resolution UNet for effective global and local feature learning. Second, we propose a Multiple Contrast Loss (MCL) that calculates synthetic loss at different intensity windows to improve quality for both soft tissues and bone regions. Experiments on the SynthRAD2023 dataset demonstrate that GLFC improved the SSIM of sCT from 77.91% to 91.50% compared with the original CBCT, and significantly outperformed several existing methods for sCT generation. The code is available at https://github.com/HiLab-git/GLFC
Authors: Runci Bai
Abstract: Brain tumors can result in neurological dysfunction, alterations in cognitive and psychological states, increased intracranial pressure, and the occurrence of seizures, thereby presenting a substantial risk to human life and health. The You Only Look Once(YOLO) series models have demonstrated superior accuracy in object detection for medical imaging. In this paper, we develop a novel SCC-YOLO architecture by integrating the SCConv attention mechanism into YOLOv9. The SCConv module reconstructs an efficient convolutional module by reducing spatial and channel redundancy among features, thereby enhancing the learning of image features. We investigate the impact of intergrating different attention mechanisms with the YOLOv9 model on brain tumor image detection using both the Br35H dataset and our self-made dataset(Brain_Tumor_Dataset). Experimental results show that on the Br35H dataset, SCC-YOLO achieved a 0.3% improvement in mAp50 compared to YOLOv9, while on our self-made dataset, SCC-YOLO exhibited a 0.5% improvement over YOLOv9. SCC-YOLO has reached state-of-the-art performance in brain tumor detection. Source code is available at : https://jihulab.com/healthcare-information-studio/SCC-YOLO/-/tree/master
URLs: https://jihulab.com/healthcare-information-studio/SCC-YOLO/-/tree/master
Authors: Ying Chen, Rami Al-Maskari, Izabela Horvath, Mayar Ali, Luciano Hoher, Kaiyuan Yang, Zengming Lin, Zhiwei Zhai, Mengzhe Shen, Dejin Xun, Yi Wang, Tony Xu, Maged Goubran, Yunheng Wu, Kensaku Mori, Johannes C. Paetzold, Ali Erturk
Abstract: Recent innovations in light sheet microscopy, paired with developments in tissue clearing techniques, enable the 3D imaging of large mammalian tissues with cellular resolution. Combined with the progress in large-scale data analysis, driven by deep learning, these innovations empower researchers to rapidly investigate the morphological and functional properties of diverse biological samples. Segmentation, a crucial preliminary step in the analysis process, can be automated using domain-specific deep learning models with expert-level performance. However, these models exhibit high sensitivity to domain shifts, leading to a significant drop in accuracy when applied to data outside their training distribution. To address this limitation, and inspired by the recent success of self-supervised learning in training generalizable models, we organized the SELMA3D Challenge during the MICCAI 2024 conference. SELMA3D provides a vast collection of light-sheet images from cleared mice and human brains, comprising 35 large 3D images-each with over 1000^3 voxels-and 315 annotated small patches for finetuning, preliminary testing and final testing. The dataset encompasses diverse biological structures, including vessel-like and spot-like structures. Five teams participated in all phases of the challenge, and their proposed methods are reviewed in this paper. Quantitative and qualitative results from most participating teams demonstrate that self-supervised learning on large datasets improves segmentation model performance and generalization. We will continue to support and extend SELMA3D as an inaugural MICCAI challenge focused on self-supervised learning for 3D microscopy image segmentation.
Authors: Chia-Ming Lee, Yu-Fan Lin, Yu-Hao Ho, Li-Wei Kang, Chih-Chung Hsu
Abstract: Hyperspectral image (HSI) fusion addresses the challenge of reconstructing High-Resolution HSIs (HR-HSIs) from High-Resolution Multispectral images (HR-MSIs) and Low-Resolution HSIs (LR-HSIs), a critical task given the high costs and hardware limitations associated with acquiring high-quality HSIs. While existing methods leverage spatial and spectral relationships, they often suffer from limited receptive fields and insufficient feature utilization, leading to suboptimal performance. Furthermore, the scarcity of high-quality HSI data highlights the importance of efficient data utilization to maximize reconstruction quality. To address these issues, we propose HyFusion, a novel Dual-Coupled Network (DCN) framework designed to enhance cross-domain feature extraction and enable effective feature map reusing. The framework first processes HR-MSI and LR-HSI inputs through specialized subnetworks that mutually enhance each other during feature extraction, preserving complementary spatial and spectral details. At its core, HyFusion utilizes an Enhanced Reception Field Block (ERFB), which combines shifting-window attention and dense connections to expand the receptive field, effectively capturing long-range dependencies while minimizing information loss. Extensive experiments demonstrate that HyFusion achieves state-of-the-art performance in HR-MSI/LR-HSI fusion, significantly improving reconstruction quality while maintaining a compact model size and computational efficiency. By integrating enhanced receptive fields and feature map reusing into a coupled network architecture, HyFusion provides a practical and effective solution for HSI fusion in resource-constrained scenarios, setting a new benchmark in hyperspectral imaging. Our code will be publicly available.
Authors: You Li, Heyu Huang, Chi Chen, Kaiyu Huang, Chao Huang, Zonghao Guo, Zhiyuan Liu, Jinan Xu, Yuhua Li, Ruixuan Li, Maosong Sun
Abstract: The recent advancement of Multimodal Large Language Models (MLLMs) has significantly improved their fine-grained perception of single images and general comprehension across multiple images. However, existing MLLMs still face challenges in achieving precise grounding in complex multi-image scenarios. To address this, we first explore a Chain-of-Thought (CoT) framework that integrates single-image grounding with multi-image comprehension. While partially effective, it remains unstable and struggles to capture abstract visual information due to its non-end-to-end nature. Therefore, we introduce Migician, the first multi-image grounding model capable of performing free-form and accurate grounding across multiple images. To support this, we present the MGrounding-630k dataset, which comprises data for several multi-image grounding tasks derived from existing datasets, along with newly generated free-form grounding instruction-following data. Furthermore, we propose MIG-Bench, a comprehensive benchmark specifically designed for evaluating multi-image grounding capabilities. Experimental results demonstrate that our model achieves significantly superior multi-image grounding capabilities, outperforming the best existing MLLMs by 21.61% and even surpassing much larger 70B models. Our code, model, dataset, and benchmark are fully open-sourced at https://migician-vg.github.io/.
Authors: Amit Kr Dey, Pradeep Walia, Girish Somvanshi, Abrar Ali, Sagarnil Das, Pallabi Paul, Minakhi Ghosh
Abstract: Purpose: Diabetic retinopathy (DR) is a major cause of vision loss, particularly in India, where access to retina specialists is limited in rural areas. This study aims to evaluate the Artificial Intelligence-based Diabetic Retinopathy Screening System (AIDRSS) for DR detection and prevalence assessment, addressing the growing need for scalable, automated screening solutions in resource-limited settings. Approach: A multicentric, cross-sectional study was conducted in Kolkata, India, involving 5,029 participants and 10,058 macula-centric retinal fundus images. The AIDRSS employed a deep learning algorithm with 50 million trainable parameters, integrated with Contrast Limited Adaptive Histogram Equalization (CLAHE) preprocessing for enhanced image quality. DR was graded using the International Clinical Diabetic Retinopathy (ICDR) Scale, categorizing disease into five stages (DR0 to DR4). Statistical metrics including sensitivity, specificity, and prevalence rates were evaluated against expert retina specialist assessments. Results: The prevalence of DR in the general population was 13.7%, rising to 38.2% among individuals with elevated random blood glucose levels. The AIDRSS achieved an overall sensitivity of 92%, specificity of 88%, and 100% sensitivity for detecting referable DR (DR3 and DR4). These results demonstrate the system's robust performance in accurately identifying and grading DR in a diverse population. Conclusions: AIDRSS provides a reliable, scalable solution for early DR detection in resource-constrained environments. Its integration of advanced AI techniques ensures high diagnostic accuracy, with potential to significantly reduce the burden of diabetes-related vision loss in underserved regions.