Authors: Aristeidis Tsaris, Chengming Zhang, Xiao Wang, Junqi Yin, Siyan Liu, Moetasim Ashfaq, Ming Fan, Jong Youl Choi, Mohamed Wahib, Dan Lu, Prasanna Balaprakash, Feiyi Wang
Abstract: Vision Transformers (ViTs) are pivotal for foundational models in scientific imagery, including Earth science applications, due to their capability to process large sequence lengths. While transformers for text has inspired scaling sequence lengths in ViTs, yet adapting these for ViTs introduces unique challenges. We develop distributed sequence parallelism for ViTs, enabling them to handle up to 1M tokens. Our approach, leveraging DeepSpeed-Ulysses and Long-Sequence-Segmentation with model sharding, is the first to apply sequence parallelism in ViT training, achieving a 94% batch scaling efficiency on 2,048 AMD-MI250X GPUs. Evaluating sequence parallelism in ViTs, particularly in models up to 10B parameters, highlighted substantial bottlenecks. We countered these with hybrid sequence, pipeline, tensor parallelism, and flash attention strategies, to scale beyond single GPU memory limits. Our method significantly enhances climate modeling accuracy by 20% in temperature predictions, marking the first training of a transformer model on a full-attention matrix over 188K sequence length.
Authors: Muhammad Bilal Shaikh, Syed Mohammed Shamsul Islam, Douglas Chai, Naveed Akhtar
Abstract: Due to its widespread applications, human action recognition is one of the most widely studied research problems in Computer Vision. Recent studies have shown that addressing it using multimodal data leads to superior performance as compared to relying on a single data modality. During the adoption of deep learning for visual modelling in the last decade, action recognition approaches have mainly relied on Convolutional Neural Networks (CNNs). However, the recent rise of Transformers in visual modelling is now also causing a paradigm shift for the action recognition task. This survey captures this transition while focusing on Multimodal Human Action Recognition (MHAR). Unique to the induction of multimodal computational models is the process of "fusing" the features of the individual data modalities. Hence, we specifically focus on the fusion design aspects of the MHAR approaches. We analyze the classic and emerging techniques in this regard, while also highlighting the popular trends in the adaption of CNN and Transformer building blocks for the overall problem. In particular, we emphasize on recent design choices that have led to more efficient MHAR models. Unlike existing reviews, which discuss Human Action Recognition from a broad perspective, this survey is specifically aimed at pushing the boundaries of MHAR research by identifying promising architectural and fusion design choices to train practicable models. We also provide an outlook of the multimodal datasets from their scale and evaluation viewpoint. Finally, building on the reviewed literature, we discuss the challenges and future avenues for MHAR.
Authors: Yesian Rohn
Abstract: This paper addresses the limitations of physical models in the current field of image dehazing by proposing an innovative dehazing network (CL2S). Building on the DM2F model, it identifies issues in its ablation experiments and replaces the original logarithmic function model with a trigonometric (sine) model. This substitution aims to better fit the complex and variable distribution of haze. The approach also integrates the atmospheric scattering model and other elementary functions to enhance dehazing performance. Experimental results demonstrate that CL2S achieves outstanding performance on multiple dehazing datasets, particularly in maintaining image details and color authenticity. Additionally, systematic ablation experiments supplementing DM2F validate the concerns raised about DM2F and confirm the necessity and effectiveness of the functional components in the proposed CL2S model. Our code is available at \url{https://github.com/YesianRohn/CL2S}, where the corresponding pre-trained models can also be accessed.
Authors: Dening Lu, Jun Zhou, Kyle Gao, Linlin Xu, Jonathan Li
Abstract: 3D Transformers have achieved great success in point cloud understanding and representation. However, there is still considerable scope for further development in effective and efficient Transformers for large-scale LiDAR point cloud scene segmentation. This paper proposes a novel 3D Transformer framework, named 3D Learnable Supertoken Transformer (3DLST). The key contributions are summarized as follows. Firstly, we introduce the first Dynamic Supertoken Optimization (DSO) block for efficient token clustering and aggregating, where the learnable supertoken definition avoids the time-consuming pre-processing of traditional superpoint generation. Since the learnable supertokens can be dynamically optimized by multi-level deep features during network learning, they are tailored to the semantic homogeneity-aware token clustering. Secondly, an efficient Cross-Attention-guided Upsampling (CAU) block is proposed for token reconstruction from optimized supertokens. Thirdly, the 3DLST is equipped with a novel W-net architecture instead of the common U-net design, which is more suitable for Transformer-based feature learning. The SOTA performance on three challenging LiDAR datasets (airborne MultiSpectral LiDAR (MS-LiDAR) (89.3% of the average F1 score), DALES (80.2% of mIoU), and Toronto-3D dataset (80.4% of mIoU)) demonstrate the superiority of 3DLST and its strong adaptability to various LiDAR point cloud data (airborne MS-LiDAR, aerial LiDAR, and vehicle-mounted LiDAR data). Furthermore, 3DLST also achieves satisfactory results in terms of algorithm efficiency, which is up to 5x faster than previous best-performing methods.
Authors: Dening Lu (Yilin), Jun Zhou (Yilin), Kyle (Yilin), Gao, Linlin Xu, Jonathan Li
Abstract: Recently, point cloud processing and analysis have made great progress due to the development of 3D Transformers. However, existing 3D Transformer methods usually are computationally expensive and inefficient due to their huge and redundant attention maps. They also tend to be slow due to requiring time-consuming point cloud sampling and grouping processes. To address these issues, we propose an efficient point TransFormer with Dynamic Token Aggregating (DTA-Former) for point cloud representation and processing. Firstly, we propose an efficient Learnable Token Sparsification (LTS) block, which considers both local and global semantic information for the adaptive selection of key tokens. Secondly, to achieve the feature aggregation for sparsified tokens, we present the first Dynamic Token Aggregating (DTA) block in the 3D Transformer paradigm, providing our model with strong aggregated features while preventing information loss. After that, a dual-attention Transformer-based Global Feature Enhancement (GFE) block is used to improve the representation capability of the model. Equipped with LTS, DTA, and GFE blocks, DTA-Former achieves excellent classification results via hierarchical feature learning. Lastly, a novel Iterative Token Reconstruction (ITR) block is introduced for dense prediction whereby the semantic features of tokens and their semantic relationships are gradually optimized during iterative reconstruction. Based on ITR, we propose a new W-net architecture, which is more suitable for Transformer-based feature learning than the common U-net design. Extensive experiments demonstrate the superiority of our method. It achieves SOTA performance with up to 30$\times$ faster than prior point Transformers on ModelNet40, ShapeNet, and airborne MultiSpectral LiDAR (MS-LiDAR) datasets.
Authors: Louis Foucard, Samar Khanna, Yi Shi, Chi-Kuei Liu, Quinn Z Shen, Thuyen Ngo, Zi-Xiang Xia
Abstract: In this paper, we propose SpotNet: a fast, single stage, image-centric but LiDAR anchored approach for long range 3D object detection. We demonstrate that our approach to LiDAR/image sensor fusion, combined with the joint learning of 2D and 3D detection tasks, can lead to accurate 3D object detection with very sparse LiDAR support. Unlike more recent bird's-eye-view (BEV) sensor-fusion methods which scale with range $r$ as $O(r^2)$, SpotNet scales as $O(1)$ with range. We argue that such an architecture is ideally suited to leverage each sensor's strength, i.e. semantic understanding from images and accurate range finding from LiDAR data. Finally we show that anchoring detections on LiDAR points removes the need to regress distances, and so the architecture is able to transfer from 2MP to 8MP resolution images without re-training.
Authors: Chak Fong Chong, Jielong Guo, Xu Yang, Wei Ke, Yapeng Wang
Abstract: Multi-label image classification datasets are often partially labeled where many labels are missing, posing a significant challenge to training accurate deep classifiers. However, the powerful Mixup sample-mixing data augmentation cannot be well utilized to address this challenge, as it cannot perform linear interpolation on the unknown labels to construct augmented samples. In this paper, we propose LogicMix, a Mixup variant designed for such partially labeled datasets. LogicMix mixes the sample labels by logical OR so that the unknown labels can be correctly mixed by utilizing OR's logical equivalences, including the domination and identity laws. Unlike Mixup, which mixes exactly two samples, LogicMix can mix multiple ($\geq2$) partially labeled samples, constructing visually more confused augmented samples to regularize training. LogicMix is more general and effective than other compared Mixup variants in the experiments on various partially labeled dataset scenarios. Moreover, it is plug-and-play and only requires minimal computation, hence it can be easily inserted into existing frameworks to collaborate with other methods to improve model performance with a negligible impact on training time, as demonstrated through extensive experiments. In particular, through the collaboration of LogicMix, RandAugment, Curriculum Labeling, and Category-wise Fine-Tuning, we attain state-of-the-art performance on MS-COCO, VG-200, and Pascal VOC 2007 benchmarking datasets. The remarkable generality, effectiveness, collaboration, and simplicity suggest that LogicMix promises to be a popular and vital data augmentation method.
Authors: Shentong Mo, Yapeng Tian
Abstract: In recent developments, the Mamba architecture, known for its selective state space approach, has shown potential in the efficient modeling of long sequences. However, its application in image generation remains underexplored. Traditional diffusion transformers (DiT), which utilize self-attention blocks, are effective but their computational complexity scales quadratically with the input length, limiting their use for high-resolution images. To address this challenge, we introduce a novel diffusion architecture, Diffusion Mamba (DiM), which foregoes traditional attention mechanisms in favor of a scalable alternative. By harnessing the inherent efficiency of the Mamba architecture, DiM achieves rapid inference times and reduced computational load, maintaining linear complexity with respect to sequence length. Our architecture not only scales effectively but also outperforms existing diffusion transformers in both image and video generation tasks. The results affirm the scalability and efficiency of DiM, establishing a new benchmark for image and video generation techniques. This work advances the field of generative models and paves the way for further applications of scalable architectures.
Authors: Parth Padalkar, Natalia \'Slusarz, Ekaterina Komendantskaya, Gopal Gupta
Abstract: Recent efforts in interpreting Convolutional Neural Networks (CNNs) focus on translating the activation of CNN filters into stratified Answer Set Programming (ASP) rule-sets. The CNN filters are known to capture high-level image concepts, thus the predicates in the rule-set are mapped to the concept that their corresponding filter represents. Hence, the rule-set effectively exemplifies the decision-making process of the CNN in terms of the concepts that it learns for any image classification task. These rule-sets help expose and understand the biases in CNNs, although correcting the biases effectively remains a challenge. We introduce a neurosymbolic framework called NeSyBiCor for bias correction in a trained CNN. Given symbolic concepts that the CNN is biased towards, expressed as ASP constraints, we convert the undesirable and desirable concepts to their corresponding vector representations. Then, the CNN is retrained using our novel semantic similarity loss that pushes the filters away from the representations of concepts that are undesirable while pushing them closer to the concepts that are desirable. The final ASP rule-set obtained after retraining, satisfies the constraints to a high degree, thus showing the revision in the knowledge of the CNN for the image classification task. We demonstrate that our NeSyBiCor framework successfully corrects the biases of CNNs trained with subsets of classes from the Places dataset while sacrificing minimal accuracy and improving interpretability, by greatly decreasing the size of the final bias-corrected rule-set w.r.t. the initial rule-set.
Authors: Artem Lukoianov, Haitz S\'aez de Oc\'ariz Borde, Kristjan Greenewald, Vitor Campagnolo Guizilini, Timur Bagautdinov, Vincent Sitzmann, Justin Solomon
Abstract: While 2D diffusion models generate realistic, high-detail images, 3D shape generation methods like Score Distillation Sampling (SDS) built on these 2D diffusion models produce cartoon-like, over-smoothed shapes. To help explain this discrepancy, we show that the image guidance used in Score Distillation can be understood as the velocity field of a 2D denoising generative process, up to the choice of a noise term. In particular, after a change of variables, SDS resembles a high-variance version of Denoising Diffusion Implicit Models (DDIM) with a differently-sampled noise term: SDS introduces noise i.i.d. randomly at each step, while DDIM infers it from the previous noise predictions. This excessive variance can lead to over-smoothing and unrealistic outputs. We show that a better noise approximation can be recovered by inverting DDIM in each SDS update step. This modification makes SDS's generative process for 2D images almost identical to DDIM. In 3D, it removes over-smoothing, preserves higher-frequency detail, and brings the generation quality closer to that of 2D samplers. Experimentally, our method achieves better or similar 3D generation quality compared to other state-of-the-art Score Distillation methods, all without training additional neural networks or multi-view supervision, and providing useful insights into relationship between 2D and 3D asset generation with diffusion models.
Authors: Yumin Zhang, Xingyu Miao, Haoran Duan, Bo Wei, Tejal Shah, Yang Long, Rajiv Ranjan
Abstract: Text-to-3D content creation is a rapidly evolving research area. Given the scarcity of 3D data, current approaches often adapt pre-trained 2D diffusion models for 3D synthesis. Among these approaches, Score Distillation Sampling (SDS) has been widely adopted. However, the issue of over-smoothing poses a significant limitation on the high-fidelity generation of 3D models. To address this challenge, LucidDreamer replaces the Denoising Diffusion Probabilistic Model (DDPM) in SDS with the Denoising Diffusion Implicit Model (DDIM) to construct Interval Score Matching (ISM). However, ISM inevitably inherits inconsistencies from DDIM, causing reconstruction errors during the DDIM inversion process. This results in poor performance in the detailed generation of 3D objects and loss of content. To alleviate these problems, we propose a novel method named Exact Score Matching (ESM). Specifically, ESM leverages auxiliary variables to mathematically guarantee exact recovery in the DDIM reverse process. Furthermore, to effectively capture the dynamic changes of the original and auxiliary variables, the LoRA of a pre-trained diffusion model implements these exact paths. Extensive experiments demonstrate the effectiveness of ESM in text-to-3D generation, particularly highlighting its superiority in detailed generation.
Authors: Jianing Qian, Anastasios Panagopoulos, Dinesh Jayaraman
Abstract: Generic re-usable pre-trained image representation encoders have become a standard component of methods for many computer vision tasks. As visual representations for robots however, their utility has been limited, leading to a recent wave of efforts to pre-train robotics-specific image encoders that are better suited to robotic tasks than their generic counterparts. We propose Scene Objects From Transformers, abbreviated as SOFT, a wrapper around pre-trained vision transformer (PVT) models that bridges this gap without any further training. Rather than construct representations out of only the final layer activations, SOFT individuates and locates object-like entities from PVT attentions, and describes them with PVT activations, producing an object-centric embedding. Across standard choices of generic pre-trained vision transformers PVT, we demonstrate in each case that policies trained on SOFT(PVT) far outstrip standard PVT representations for manipulation tasks in simulated and real settings, approaching the state-of-the-art robotics-aware representations. Code, appendix and videos: https://sites.google.com/view/robot-soft/
Authors: Soumyabrata Kundu, Risi Kondor
Abstract: In this work we introduce Steerable Transformers, an extension of the Vision Transformer mechanism that maintains equivariance to the special Euclidean group $\mathrm{SE}(d)$. We propose an equivariant attention mechanism that operates on features extracted by steerable convolutions. Operating in Fourier space, our network utilizes Fourier space non-linearities. Our experiments in both two and three dimensions show that adding a steerable transformer encoder layer to a steerable convolution network enhances performance.
Authors: Yi-Ting Shen, Hyungtae Lee, Heesung Kwon, Shuvra S. Bhattacharyya
Abstract: We present a framework for diversifying human poses in a synthetic dataset for aerial-view human detection. Our method firstly constructs a set of novel poses using a pose generator and then alters images in the existing synthetic dataset to assume the novel poses while maintaining the original style using an image translator. Since images corresponding to the novel poses are not available in training, the image translator is trained to be applicable only when the input and target poses are similar, thus training does not require the novel poses and their corresponding images. Next, we select a sequence of target novel poses from the novel pose set, using Dijkstra's algorithm to ensure that poses closer to each other are located adjacently in the sequence. Finally, we repeatedly apply the image translator to each target pose in sequence to produce a group of novel pose images representing a variety of different limited body movements from the source pose. Experiments demonstrate that, regardless of how the synthetic data is used for training or the data size, leveraging the pose-diversified synthetic dataset in training generally presents remarkably better accuracy than using the original synthetic dataset on three aerial-view human detection benchmarks (VisDrone, Okutama-Action, and ICG) in the few-shot regime.
Authors: Abdullah Nazhat Abdullah, Tarkan Aydin
Abstract: Transformer architecture currently represents the main driver behind many successes in a variety of tasks addressed by deep learning, especially the recent advances in natural language processing (NLP) culminating with large language models (LLM). In addition, transformer architecture has found a wide spread of interest from computer vision (CV) researchers and practitioners, allowing for many advancements in vision-related tasks and opening the door for multi-task and multi-modal deep learning architectures that share the same principle of operation. One drawback to these architectures is their reliance on the scaled dot product attention mechanism with the softmax activation function, which is computationally expensive and requires large compute capabilities both for training and inference. This paper investigates substituting the attention mechanism usually adopted for transformer architecture with an architecture incorporating gated linear unit (GLU) activation within a multi-layer perceptron (MLP) structure in conjunction with the default MLP incorporated in the traditional transformer design. Another step forward taken by this paper is to eliminate the second non-gated MLP to further reduce the computational cost. Experimental assessments conducted by this research show that both proposed modifications and reductions offer competitive performance in relation to baseline architectures, in support of the aims of this work in establishing a more efficient yet capable alternative to the traditional attention mechanism as the core component in designing transformer architectures.
Authors: Yiran Luo, Joshua Feinglass, Tejas Gokhale, Kuan-Cheng Lee, Chitta Baral, Yezhou Yang
Abstract: Domain Generalization (DG) is a challenging task in machine learning that requires a coherent ability to comprehend shifts across various domains through extraction of domain-invariant features. DG performance is typically evaluated by performing image classification in domains of various image styles. However, current methodology lacks quantitative understanding about shifts in stylistic domain, and relies on a vast amount of pre-training data, such as ImageNet1K, which are predominantly in photo-realistic style with weakly supervised class labels. Such a data-driven practice could potentially result in spurious correlation and inflated performance on DG benchmarks. In this paper, we introduce a new DG paradigm to address these risks. We first introduce two new quantitative measures ICV and IDD to describe domain shifts in terms of consistency of classes within one domain and similarity between two stylistic domains. We then present SuperMarioDomains (SMD), a novel synthetic multi-domain dataset sampled from video game scenes with more consistent classes and sufficient dissimilarity compared to ImageNet1K. We demonstrate our DG method SMOS. SMOS first uses SMD to train a precursor model, which is then used to ground the training on a DG benchmark. We observe that SMOS contributes to state-of-the-art performance across five DG benchmarks, gaining large improvements to performances on abstract domains along with on-par or slight improvements to those on photo-realistic domains. Our qualitative analysis suggests that these improvements can be attributed to reduced distributional divergence between originally distant domains. Our data are available at https://github.com/fpsluozi/SMD-SMOS .
Authors: Haoran Duan, Shidong Wang, Varun Ojha, Shizheng Wang, Yawen Huang, Yang Long, Rajiv Ranjan, Yefeng Zheng
Abstract: While traditional feature engineering for Human Activity Recognition (HAR) involves a trial-anderror process, deep learning has emerged as a preferred method for high-level representations of sensor-based human activities. However, most deep learning-based HAR requires a large amount of labelled data and extracting HAR features from unlabelled data for effective deep learning training remains challenging. We, therefore, introduce a deep semi-supervised HAR approach, MixHAR, which concurrently uses labelled and unlabelled activities. Our MixHAR employs a linear interpolation mechanism to blend labelled and unlabelled activities while addressing both inter- and intra-activity variability. A unique challenge identified is the activityintrusion problem during mixing, for which we propose a mixing calibration mechanism to mitigate it in the feature embedding space. Additionally, we rigorously explored and evaluated the five conventional/popular deep semi-supervised technologies on HAR, acting as the benchmark of deep semi-supervised HAR. Our results demonstrate that MixHAR significantly improves performance, underscoring the potential of deep semi-supervised techniques in HAR.
Authors: Haiyu Wu, Sicong Tian, Aman Bhatta, Jacob Gutierrez, Grace Bezold, Genesis Argueta, Karl Ricanek Jr., Michael C. King, Kevin W. Bowyer
Abstract: Face Recognition models are commonly trained with web-scraped datasets containing millions of images and evaluated on test sets emphasizing pose, age and mixed attributes. With train and test sets both assembled from web-scraped images, it is critical to ensure disjoint sets of identities between train and test sets. However, existing train and test sets have not considered this. Moreover, as accuracy levels become saturated, such as LFW $>99.8\%$, more challenging test sets are needed. We show that current train and test sets are generally not identity- or even image-disjoint, and that this results in an optimistic bias in the estimated accuracy. In addition, we show that identity-disjoint folds are important in the 10-fold cross-validation estimate of test accuracy. To better support continued advances in face recognition, we introduce two "Goldilocks" test sets, Hadrian and Eclipse. The former emphasizes challenging facial hairstyles and latter emphasizes challenging over- and under-exposure conditions. Images in both datasets are from a large, controlled-acquisition (not web-scraped) dataset, so they are identity- and image-disjoint with all popular training sets. Accuracy for these new test sets generally falls below that observed on LFW, CPLFW, CALFW, CFP-FP and AgeDB-30, showing that these datasets represent important dimensions for improvement of face recognition. The datasets are available at: \url{https://github.com/HaiyuWu/SOTA-Face-Recognition-Train-and-Test}
URLs: https://github.com/HaiyuWu/SOTA-Face-Recognition-Train-and-Test
Authors: Xiyao Wang, Jiuhai Chen, Zhaoyang Wang, Yuhang Zhou, Yiyang Zhou, Huaxiu Yao, Tianyi Zhou, Tom Goldstein, Parminder Bhatia, Furong Huang, Cao Xiao
Abstract: Large vision-language models (LVLMs) have achieved impressive results in various visual question-answering and reasoning tasks through vision instruction tuning on specific datasets. However, there is still significant room for improvement in the alignment between visual and language modalities. Previous methods to enhance this alignment typically require external models or data, heavily depending on their capabilities and quality, which inevitably sets an upper bound on performance. In this paper, we propose SIMA, a framework that enhances visual and language modality alignment through self-improvement, eliminating the needs for external models or data. SIMA leverages prompts from existing vision instruction tuning datasets to self-generate responses and employs an in-context self-critic mechanism to select response pairs for preference tuning. The key innovation is the introduction of three vision metrics during the in-context self-critic process, which can guide the LVLM in selecting responses that enhance image comprehension. Through experiments across 14 hallucination and comprehensive benchmarks, we demonstrate that SIMA not only improves model performance across all benchmarks but also achieves superior modality alignment, outperforming previous approaches.
Authors: Aleksandr Algasov, Ekaterina Nepovinnykh, Tuomas Eerola, Heikki K\"alvi\"ainen, Charles V. Stewart, Lasha Otarashvili, Jason A. Holmberg
Abstract: Recent advancements in the automatic re-identification of animal individuals from images have opened up new possibilities for studying wildlife through camera traps and citizen science projects. Existing methods leverage distinct and permanent visual body markings, such as fur patterns or scars, and typically employ one of two strategies: local features or end-to-end learning. In this study, we delve into the impact of training set size by conducting comprehensive experiments across six different methods and five animal species. While it is well known that end-to-end learning-based methods surpass local feature-based methods given a sufficient amount of good-quality training data, the challenge of gathering such datasets for wildlife animals means that local feature-based methods remain a more practical approach for many species. We demonstrate the benefits of both local feature and end-to-end learning-based approaches and show that species-specific characteristics, particularly intra-individual variance, have a notable effect on training data requirements.
Authors: Uche Ochuba
Abstract: This paper addresses the critical issue of deforestation by exploring the application of vision transformers (ViTs) for classifying the drivers of deforestation using satellite imagery from Indonesian forests. Motivated by the urgency of this problem, I propose an approach that leverages ViTs and machine learning techniques. The input to my algorithm is a 332x332-pixel satellite image, and I employ a ViT architecture to predict the deforestation driver class; grassland shrubland, other, plantation, or smallholder agriculture. My methodology involves fine-tuning a pre-trained ViT on a dataset from the Stanford ML Group, and I experiment with rotational data augmentation techniques (among others) and embedding of longitudinal data to improve classification accuracy. I also tried training a ViT from scratch. Results indicate a significant improvement over baseline models, achieving a test accuracy of 72.9%. I conduct a comprehensive analysis, including error patterns and metrics, to highlight the strengths and limitations of my approach. This research contributes to the ongoing efforts to address deforestation challenges through advanced computer vision techniques.
Authors: Peiyao Wang, Yuewei Lin, Erik Blasch, Jie Wei, Haibin Ling
Abstract: Although the performance of Temporal Action Segmentation (TAS) has improved in recent years, achieving promising results often comes with a high computational cost due to dense inputs, complex model structures, and resource-intensive post-processing requirements. To improve the efficiency while keeping the performance, we present a novel perspective centered on per-segment classification. By harnessing the capabilities of Transformers, we tokenize each video segment as an instance token, endowed with intrinsic instance segmentation. To realize efficient action segmentation, we introduce BaFormer, a boundary-aware Transformer network. It employs instance queries for instance segmentation and a global query for class-agnostic boundary prediction, yielding continuous segment proposals. During inference, BaFormer employs a simple yet effective voting strategy to classify boundary-wise segments based on instance segmentation. Remarkably, as a single-stage approach, BaFormer significantly reduces the computational costs, utilizing only 6% of the running time compared to state-of-the-art method DiffAct, while producing better or comparable accuracy over several popular benchmarks. The code for this project is publicly available at https://github.com/peiyao-w/BaFormer.
Authors: Kazutaka Kiuchi, Shimpei Imamura, Norihiko Kawai
Abstract: Recent studies have shown that visually impaired people have desires to take selfies in the same way as sighted people do to record their photos and share them with others. Although support applications using sound and vibration have been developed to help visually impaired people take selfies using smartphone cameras, it is still difficult to capture everyone in the angle of view, and it is also difficult to confirm that they all have good expressions in the photo. To mitigate these issues, we propose a method to take selfies with multiple people using an omni-directional camera. Specifically, a user takes a few seconds of video with an omni-directional camera, followed by face detection on all frames. The proposed method then eliminates false face detections and complements undetected ones considering the consistency across all frames. After performing facial expression recognition on all the frames, the proposed method finally extracts the frame in which the participants are happiest, and generates a perspective projection image in which all the participants are in the angle of view from the omni-directional frame. In experiments, we use several scenes with different number of people taken to demonstrate the effectiveness of the proposed method.
Authors: Junyi Wu, Haoxuan Wang, Yuzhang Shang, Mubarak Shah, Yan Yan
Abstract: The recent introduction of Diffusion Transformers (DiTs) has demonstrated exceptional capabilities in image generation by using a different backbone architecture, departing from traditional U-Nets and embracing the scalable nature of transformers. Despite their advanced capabilities, the wide deployment of DiTs, particularly for real-time applications, is currently hampered by considerable computational demands at the inference stage. Post-training Quantization (PTQ) has emerged as a fast and data-efficient solution that can significantly reduce computation and memory footprint by using low-bit weights and activations. However, its applicability to DiTs has not yet been explored and faces non-trivial difficulties due to the unique design of DiTs. In this paper, we propose PTQ4DiT, a specifically designed PTQ method for DiTs. We discover two primary quantization challenges inherent in DiTs, notably the presence of salient channels with extreme magnitudes and the temporal variability in distributions of salient activation over multiple timesteps. To tackle these challenges, we propose Channel-wise Salience Balancing (CSB) and Spearmen's $\rho$-guided Salience Calibration (SSC). CSB leverages the complementarity property of channel magnitudes to redistribute the extremes, alleviating quantization errors for both activations and weights. SSC extends this approach by dynamically adjusting the balanced salience to capture the temporal variations in activation. Additionally, to eliminate extra computational costs caused by PTQ4DiT during inference, we design an offline re-parameterization strategy for DiTs. Experiments demonstrate that our PTQ4DiT successfully quantizes DiTs to 8-bit precision (W8A8) while preserving comparable generation ability and further enables effective quantization to 4-bit weight precision (W4A8) for the first time.
Authors: Hakim Ikebayashi, Norihiko Kawai
Abstract: Augmented reality (AR) using camera images in mobile devices is becoming popular for tourism promotion. However, obstructions such as tourists appearing in the camera images may cause the camera pose estimation error, resulting in CG misalignment and reduced visibility of the contents. To avoid this problem, Indirect AR (IAR), which does not use real-time camera images, has been proposed. In this method, an omnidirectional image is captured and virtual objects are synthesized on the image in advance. Users can experience AR by viewing a scene extracted from the synthesized omnidirectional image according to the device's sensor. This enables robustness and high visibility. However, if the weather conditions and season in the pre-captured 360 images differs from the current weather conditions and season when AR is experienced, the realism of the AR experience is reduced. To overcome the problem, we propose a method for correcting the intensity and texture of a past omnidirectional image using camera images from mobile devices. We first perform semantic segmentation. We then reproduce the current sky pattern by panoramic image composition and inpainting. For the other areas, we correct the intensity by histogram matching. In experiments, we show the effectiveness of the proposed method using various scenes.
Authors: Rui Qian, Xiaoyi Dong, Pan Zhang, Yuhang Zang, Shuangrui Ding, Dahua Lin, Jiaqi Wang
Abstract: This paper presents VideoStreaming, an advanced vision-language large model (VLLM) for video understanding, that capably understands arbitrary-length video with a constant number of video tokens streamingly encoded and adaptively selected. The challenge of video understanding in the vision language area mainly lies in the significant computational burden caused by the great number of tokens extracted from long videos. Previous works rely on sparse sampling or frame compression to reduce tokens. However, such approaches either disregard temporal information in a long time span or sacrifice spatial details, resulting in flawed compression. To address these limitations, our VideoStreaming has two core designs: Memory-Propagated Streaming Encoding and Adaptive Memory Selection. The Memory-Propagated Streaming Encoding architecture segments long videos into short clips and sequentially encodes each clip with a propagated memory. In each iteration, we utilize the encoded results of the preceding clip as historical memory, which is integrated with the current clip to distill a condensed representation that encapsulates the video content up to the current timestamp. After the encoding process, the Adaptive Memory Selection strategy selects a constant number of question-related memories from all the historical memories and feeds them into the LLM to generate informative responses. The question-related selection reduces redundancy within the memories, enabling efficient and precise video understanding. Meanwhile, the disentangled video extraction and reasoning design allows the LLM to answer different questions about a video by directly selecting corresponding memories, without the need to encode the whole video for each question. Our model achieves superior performance and higher efficiency on long video benchmarks, showcasing precise temporal comprehension for detailed question answering.
Authors: Yusuke Akamatsu, Terumi Umematsu, Hitoshi Imaoka, Shizuko Gomi, Hideo Tsurushima
Abstract: Daily monitoring of intra-personal facial changes associated with health and emotional conditions has great potential to be useful for medical, healthcare, and emotion recognition fields. However, the approach for capturing intra-personal facial changes is relatively unexplored due to the difficulty of collecting temporally changing face images. In this paper, we propose a facial representation learning method using synthetic images for comparing faces, called ComFace, which is designed to capture intra-personal facial changes. For effective representation learning, ComFace aims to acquire two feature representations, i.e., inter-personal facial differences and intra-personal facial changes. The key point of our method is the use of synthetic face images to overcome the limitations of collecting real intra-personal face images. Facial representations learned by ComFace are transferred to three extensive downstream tasks for comparing faces: estimating facial expression changes, weight changes, and age changes from two face images of the same individual. Our ComFace, trained using only synthetic data, achieves comparable to or better transfer performance than general pre-training and state-of-the-art representation learning methods trained using real images.
Authors: Xiangyu Chen, Zhenzhen Liu, Katie Z Luo, Siddhartha Datta, Adhitya Polavaram, Yan Wang, Yurong You, Boyi Li, Marco Pavone, Wei-Lun Chao, Mark Campbell, Bharath Hariharan, Kilian Q. Weinberger
Abstract: Ensuring robust 3D object detection and localization is crucial for many applications in robotics and autonomous driving. Recent models, however, face difficulties in maintaining high performance when applied to domains with differing sensor setups or geographic locations, often resulting in poor localization accuracy due to domain shift. To overcome this challenge, we introduce a novel diffusion-based box refinement approach. This method employs a domain-agnostic diffusion model, conditioned on the LiDAR points surrounding a coarse bounding box, to simultaneously refine the box's location, size, and orientation. We evaluate this approach under various domain adaptation settings, and our results reveal significant improvements across different datasets, object classes and detectors.
Authors: Xue Zhang, Si-Yuan Cao, Fang Wang, Runmin Zhang, Zhe Wu, Xiaohan Zhang, Xiaokai Bai, Hui-Liang Shen
Abstract: Most recent multispectral object detectors employ a two-branch structure to extract features from RGB and thermal images. While the two-branch structure achieves better performance than a single-branch structure, it overlooks inference efficiency. This conflict is increasingly aggressive, as recent works solely pursue higher performance rather than both performance and efficiency. In this paper, we address this issue by improving the performance of efficient single-branch structures. We revisit the reasons causing the performance gap between these structures. For the first time, we reveal the information interference problem in the naive early-fusion strategy adopted by previous single-branch structures. Besides, we find that the domain gap between multispectral images, and weak feature representation of the single-branch structure are also key obstacles for performance. Focusing on these three problems, we propose corresponding solutions, including a novel shape-priority early-fusion strategy, a weakly supervised learning method, and a core knowledge distillation technique. Experiments demonstrate that single-branch networks equipped with these three contributions achieve significant performance enhancements while retaining high efficiency. Our code will be available at \url{https://github.com/XueZ-phd/Efficient-RGB-T-Early-Fusion-Detection}.
URLs: https://github.com/XueZ-phd/Efficient-RGB-T-Early-Fusion-Detection
Authors: Yuzhong Zhao, Feng Liu, Yue Liu, Mingxiang Liao, Chen Gong, Qixiang Ye, Fang Wan
Abstract: Region-level multi-modality methods can translate referred image regions to human preferred language descriptions. Unfortunately, most of existing methods using fixed visual inputs remain lacking the resolution adaptability to find out precise language descriptions. In this study, we propose a dynamic resolution approach, referred to as DynRefer, to pursue high-accuracy region-level referring through mimicking the resolution adaptability of human visual cognition. DynRefer first implements stochastic vision-language alignment. It aligns desired language descriptions of multi-modality tasks with images of stochastic resolution, which are constructed by nesting a set of views around the referred region. DynRefer then implements dynamic multi-modality referring, which is realized by selecting views based on image and language priors. This allows the visual information used for referring to better match human preferences, thereby improving the representational adaptability of region-level multi-modality models. Extensive experiments show that DynRefer brings mutual improvement upon tasks including region-level captioning, open-vocabulary region recognition and attribute detection. Last but not least, DynRefer achieves new state-of-the-art on multiple region-level multi-modality tasks using a single model. Code is available at https://github.com/callsys/DynRefer.
Authors: Hyekyoung Hwang, Jitae Shin
Abstract: Deep Learning (DL) has made remarkable achievements in computer vision and adopted in safety critical domains such as medical imaging or autonomous drive. Thus, it is necessary to understand the uncertainty of the model to effectively reduce accidents and losses due to misjudgment of the Deep Neural Networks (DNN). This can start by efficiently selecting data that could potentially malfunction to the model. Traditionally, data collection and labeling have been done manually, but recently test data selection methods have emerged that focus on capturing samples that are not relevant to what the model had been learned. They're selected based on the activation pattern of neurons in DNN, entropy minimization based on softmax output of the DL. However, these methods cannot quantitatively analyze the extent to which unseen samples are extrapolated from the training data. Therefore, we propose To-hull Uncertainty and Closure Ratio, which measures an uncertainty of trained model based on the convex hull of training data. It can observe the positional relation between the convex hull of the learned data and an unseen sample and infer how extrapolate the sample is from the convex hull. To evaluate the proposed method, we conduct empirical studies on popular datasets and DNN models, compared to state-of-the art test selection metrics. As a result of the experiment, the proposed To-hull Uncertainty is effective in finding samples with unusual patterns (e.g. adversarial attack) compared to the existing test selection metric.
Authors: Junjie Gao, Chongjian Wang, Zhongjun Ding, Shuangmin Chen, Shiqing Xin, Changhe Tu, Wenping Wang
Abstract: In the realm of point cloud registration, the most prevalent pose evaluation approaches are statistics-based, identifying the optimal transformation by maximizing the number of consistent correspondences. However, registration recall decreases significantly when point clouds exhibit a low overlap rate, despite efforts in designing feature descriptors and establishing correspondences. In this paper, we introduce Deep-PE, a lightweight, learning-based pose evaluator designed to enhance the accuracy of pose selection, especially in challenging point cloud scenarios with low overlap. Our network incorporates a Pose-Aware Attention (PAA) module to simulate and learn the alignment status of point clouds under various candidate poses, alongside a Pose Confidence Prediction (PCP) module that predicts the likelihood of successful registration. These two modules facilitate the learning of both local and global alignment priors. Extensive tests across multiple benchmarks confirm the effectiveness of Deep-PE. Notably, on 3DLoMatch with a low overlap rate, Deep-PE significantly outperforms state-of-the-art methods by at least 8% and 11% in registration recall under handcrafted FPFH and learning-based FCGF descriptors, respectively. To the best of our knowledge, this is the first study to utilize deep learning to select the optimal pose without the explicit need for input correspondences.
Authors: Myong Chol Jung, He Zhao, Joanna Dipnall, Belinda Gabbe, Lan Du
Abstract: Prompt learning has shown to be an efficient and effective fine-tuning method for vision-language models like CLIP. While numerous studies have focused on the generalisation of these models in few-shot classification, their capability in near out-of-distribution (OOD) detection has been overlooked. A few recent works have highlighted the promising performance of prompt learning in far OOD detection. However, the more challenging task of few-shot near OOD detection has not yet been addressed. In this study, we investigate the near OOD detection capabilities of prompt learning models and observe that commonly used OOD scores have limited performance in near OOD detection. To enhance the performance, we propose a fast and simple post-hoc method that complements existing logit-based scores, improving near OOD detection AUROC by up to 11.67% with minimal computational cost. Our method can be easily applied to any prompt learning model without change in architecture or re-training the models. Comprehensive empirical evaluations across 13 datasets and 8 models demonstrate the effectiveness and adaptability of our method.
Authors: Qikai Wang, Rundong He, Yongshun Gong, Chunxiao Ren, Haoliang Sun, Xiaoshui Huang, Yilong Yin
Abstract: Semi-supervised learning can significantly boost model performance by leveraging unlabeled data, particularly when labeled data is scarce. However, real-world unlabeled data often contain unseen-class samples, which can hinder the classification of seen classes. To address this issue, mainstream safe SSL methods suggest detecting and discarding unseen-class samples from unlabeled data. Nevertheless, these methods typically employ a single-model strategy to simultaneously tackle both the classification of seen classes and the detection of unseen classes. Our research indicates that such an approach may lead to conflicts during training, resulting in suboptimal model optimization. Inspired by this, we introduce a novel framework named Diverse Teacher-Students (\textbf{DTS}), which uniquely utilizes dual teacher-student models to individually and effectively handle these two tasks. DTS employs a novel uncertainty score to softly separate unseen-class and seen-class data from the unlabeled set, and intelligently creates an additional ($K$+1)-th class supervisory signal for training. By training both teacher-student models with all unlabeled samples, DTS can enhance the classification of seen classes while simultaneously improving the detection of unseen classes. Comprehensive experiments demonstrate that DTS surpasses baseline methods across a variety of datasets and configurations. Our code and models can be publicly accessible on the link https://github.com/Zhanlo/DTS.
Authors: Zhaochen Liu, Limeng Qiao, Xiangxiang Chu, Tingting Jiang
Abstract: Aiming to predict the complete shapes of partially occluded objects, amodal segmentation is an important step towards visual intelligence. With crucial significance, practical prior knowledge derives from sufficient training, while limited amodal annotations pose challenges to achieve better performance. To tackle this problem, utilizing the mighty priors accumulated in the foundation model, we propose the first SAM-based amodal segmentation approach, PLUG. Methodologically, a novel framework with hierarchical focus is presented to better adapt the task characteristics and unleash the potential capabilities of SAM. In the region level, due to the association and division in visible and occluded areas, inmodal and amodal regions are assigned as the focuses of distinct branches to avoid mutual disturbance. In the point level, we introduce the concept of uncertainty to explicitly assist the model in identifying and focusing on ambiguous points. Guided by the uncertainty map, a computation-economic point loss is applied to improve the accuracy of predicted boundaries. Experiments are conducted on several prominent datasets, and the results show that our proposed method outperforms existing methods with large margins. Even with fewer total parameters, our method still exhibits remarkable advantages.
Authors: Kunye Shen, Xiaofei Zhou, Zhi Liu
Abstract: The automated surface defect detection is a fundamental task in industrial production, and the existing saliencybased works overcome the challenging scenes and give promising detection results. However, the cutting-edge efforts often suffer from large parameter size, heavy computational cost, and slow inference speed, which heavily limits the practical applications. To this end, we devise a multi-scale interactive (MI) module, which employs depthwise convolution (DWConv) and pointwise convolution (PWConv) to independently extract and interactively fuse features of different scales, respectively. Particularly, the MI module can provide satisfactory characterization for defect regions with fewer parameters. Embarking on this module, we propose a lightweight Multi-scale Interactive Network (MINet) to conduct real-time salient object detection of strip steel surface defects. Comprehensive experimental results on SD-Saliency-900 dataset, which contains three kinds of strip steel surface defect detection images (i.e., inclusion, patches, and scratches), demonstrate that the proposed MINet presents comparable detection accuracy with the state-of-the-art methods while running at a GPU speed of 721FPS and a CPU speed of 6.3FPS for 368*368 images with only 0.28M parameters. The code is available at https://github.com/Kunye-Shen/MINet.
Authors: Zizhao Hu, Mohammad Rostami
Abstract: The Transformer architecture has dominated machine learning in a wide range of tasks. The specific characteristic of this architecture is an expensive scaled dot-product attention mechanism that models the inter-token interactions, which is known to be the reason behind its success. However, such a mechanism does not have a direct parallel to the human brain which brings the question if the scaled-dot product is necessary for intelligence with strong expressive power. Inspired by the lateralization of the human brain, we propose a new simple but effective architecture called the Lateralization MLP (L-MLP). Stacking L-MLP blocks can generate complex architectures. Each L-MLP block is based on a multi-layer perceptron (MLP) that permutes data dimensions, processes each dimension in parallel, merges them, and finally passes through a joint MLP. We discover that this specific design outperforms other MLP variants and performs comparably to a transformer-based architecture in the challenging diffusion task while being highly efficient. We conduct experiments using text-to-image generation tasks to demonstrate the effectiveness and efficiency of L-MLP. Further, we look into the model behavior and discover a connection to the function of the human brain. Our code is publicly available: \url{https://github.com/zizhao-hu/L-MLP}
Authors: Huizhou Chen, Jiangyi Wang, Yuxin Li, Na Zhao, Jun Cheng, Xulei Yang
Abstract: 3D environment recognition is essential for autonomous driving systems, as autonomous vehicles require a comprehensive understanding of surrounding scenes. Recently, the predominant approach to define this real-life problem is through 3D occupancy prediction. It attempts to predict the occupancy states and semantic labels for all voxels in 3D space, which enhances the perception capability. Birds-Eye-View(BEV)-based perception has achieved the SOTA performance for this task. Nonetheless, this architecture fails to represent various scales of BEV features. In this paper, inspired by the success of UNet in semantic segmentation tasks, we introduce a novel UNet-like Multi-scale Occupancy Head module to relieve this issue. Furthermore, we propose the class-balancing loss to compensate for rare classes in the dataset. The experimental results on nuScenes 3D occupancy challenge dataset show the superiority of our proposed approach over baseline and SOTA methods.
Authors: Jiangwei Weng, Zhiqiang Yan, Ying Tai, Jianjun Qian, Jian Yang, Jun Li
Abstract: Recent advances in low light image enhancement have been dominated by Retinex-based learning framework, leveraging convolutional neural networks (CNNs) and Transformers. However, the vanilla Retinex theory primarily addresses global illumination degradation and neglects local issues such as noise and blur in dark conditions. Moreover, CNNs and Transformers struggle to capture global degradation due to their limited receptive fields. While state space models (SSMs) have shown promise in the long-sequence modeling, they face challenges in combining local invariants and global context in visual data. In this paper, we introduce MambaLLIE, an implicit Retinex-aware low light enhancer featuring a global-then-local state space design. We first propose a Local-Enhanced State Space Module (LESSM) that incorporates an augmented local bias within a 2D selective scan mechanism, enhancing the original SSMs by preserving local 2D dependency. Additionally, an Implicit Retinex-aware Selective Kernel module (IRSK) dynamically selects features using spatially-varying operations, adapting to varying inputs through an adaptive kernel selection process. Our Global-then-Local State Space Block (GLSSB) integrates LESSM and IRSK with LayerNorm as its core. This design enables MambaLLIE to achieve comprehensive global long-range modeling and flexible local feature aggregation. Extensive experiments demonstrate that MambaLLIE significantly outperforms state-of-the-art CNN and Transformer-based methods. Project Page: https://mamballie.github.io/anon/
Authors: Yuanhuiyi Lyu, Xu Zheng, Dahun Kim, Lin Wang
Abstract: Research on multi-modal learning dominantly aligns the modalities in a unified space at training, and only a single one is taken for prediction at inference. However, for a real machine, e.g., a robot, sensors could be added or removed at any time. Thus, it is crucial to enable the machine to tackle the mismatch and unequal-scale problems of modality combinations between training and inference. In this paper, we tackle these problems from a new perspective: "Modalities Help Modalities". Intuitively, we present OmniBind, a novel two-stage learning framework that can achieve any modality combinations and interaction. It involves teaching data-constrained, a.k.a, student, modalities to be aligned with the well-trained data-abundant, a.k.a, teacher, modalities. This subtly enables the adaptive fusion of any modalities to build a unified representation space for any combinations. Specifically, we propose Cross-modal Alignment Distillation (CAD) to address the unequal-scale problem between student and teacher modalities and effectively align student modalities into the teacher modalities' representation space in stage one. We then propose an Adaptive Fusion (AF) module to fuse any modality combinations and learn a unified representation space in stage two. To address the mismatch problem, we aggregate existing datasets and combine samples from different modalities by the same semantics. This way, we build the first dataset for training and evaluation that consists of teacher (image, text) and student (touch, thermal, event, point cloud, audio) modalities and enables omni-bind for any of them. Extensive experiments on the recognition task show performance gains over prior arts by an average of 4.05 % on the arbitrary modality combination setting. It also achieves state-of-the-art performance for a single modality, e.g., touch, with a 4.34 % gain.
Authors: Ma\"elic Neau, Paulo E. Santos, Karl Sammut, Anne-Gwenn Bosser, C\'edric Buche
Abstract: Scene Graph Generation (SGG) can extract abstract semantic relations between entities in images as graph representations. This task holds strong promises for other downstream tasks such as the embodied cognition of an autonomous agent. However, to power such applications, SGG needs to solve the gap of real-time latency. In this work, we propose to investigate the bottlenecks of current approaches for real-time constraint applications. Then, we propose a simple yet effective implementation of a real-time SGG approach using YOLOV8 as an object detection backbone. Our implementation is the first to obtain more than 48 FPS for the task with no loss of accuracy, successfully outperforming any other lightweight approaches. Our code is freely available at https://github.com/Maelic/SGG-Benchmark.
Authors: Mingli Zhu, Siyuan Liang, Baoyuan Wu
Abstract: Deep neural networks face persistent challenges in defending against backdoor attacks, leading to an ongoing battle between attacks and defenses. While existing backdoor defense strategies have shown promising performance on reducing attack success rates, can we confidently claim that the backdoor threat has truly been eliminated from the model? To address it, we re-investigate the characteristics of the backdoored models after defense (denoted as defense models). Surprisingly, we find that the original backdoors still exist in defense models derived from existing post-training defense strategies, and the backdoor existence is measured by a novel metric called backdoor existence coefficient. It implies that the backdoors just lie dormant rather than being eliminated. To further verify this finding, we empirically show that these dormant backdoors can be easily re-activated during inference, by manipulating the original trigger with well-designed tiny perturbation using universal adversarial attack. More practically, we extend our backdoor reactivation to black-box scenario, where the defense model can only be queried by the adversary during inference, and develop two effective methods, i.e., query-based and transfer-based backdoor re-activation attacks. The effectiveness of the proposed methods are verified on both image classification and multimodal contrastive learning (i.e., CLIP) tasks. In conclusion, this work uncovers a critical vulnerability that has never been explored in existing defense strategies, emphasizing the urgency of designing more robust and advanced backdoor defense mechanisms in the future.
Authors: Hong-Shuo Chen, Yao Zhu, Suya You, Azad M. Madni, C. -C. Jay Kuo
Abstract: We introduce GreenCOD, a green method for detecting camouflaged objects, distinct in its avoidance of backpropagation techniques. GreenCOD leverages gradient boosting and deep features extracted from pre-trained Deep Neural Networks (DNNs). Traditional camouflaged object detection (COD) approaches often rely on complex deep neural network architectures, seeking performance improvements through backpropagation-based fine-tuning. However, such methods are typically computationally demanding and exhibit only marginal performance variations across different models. This raises the question of whether effective training can be achieved without backpropagation. Addressing this, our work proposes a new paradigm that utilizes gradient boosting for COD. This approach significantly simplifies the model design, resulting in a system that requires fewer parameters and operations and maintains high performance compared to state-of-the-art deep learning models. Remarkably, our models are trained without backpropagation and achieve the best performance with fewer than 20G Multiply-Accumulate Operations (MACs). This new, more efficient paradigm opens avenues for further exploration in green, backpropagation-free model training.
Authors: Xinyi Chen, Yaohui Li, Haoxing Chen
Abstract: We study the problem of few-shot out-of-distribution (OOD) detection, which aims to detect OOD samples from unseen categories during inference time with only a few labeled in-domain (ID) samples. Existing methods mainly focus on training task-aware prompts for OOD detection. However, training on few-shot data may cause severe overfitting and textual prompts alone may not be enough for effective detection. To tackle these problems, we propose a prior-based Training-free Dual Adaptation method (Dual-Adapter) to detect OOD samples from both textual and visual perspectives. Specifically, Dual-Adapter first extracts the most significant channels as positive features and designates the remaining less relevant channels as negative features. Then, it constructs both a positive adapter and a negative adapter from a dual perspective, thereby better leveraging previously outlooked or interfering features in the training dataset. In this way, Dual-Adapter can inherit the advantages of CLIP not having to train, but also excels in distinguishing between ID and OOD samples. Extensive experimental results on four benchmark datasets demonstrate the superiority of Dual-Adapter.
Authors: Jiawei Fang, Haishan Song, Chengxu Zuo, Xiaoxia Gao, Xiaowei Chen, Shihui Guo, Yipeng Qin
Abstract: Flexible sensors hold promise for human motion capture (MoCap), offering advantages such as wearability, privacy preservation, and minimal constraints on natural movement. However, existing flexible sensor-based MoCap methods rely on deep learning and necessitate large and diverse labeled datasets for training. These data typically need to be collected in MoCap studios with specialized equipment and substantial manual labor, making them difficult and expensive to obtain at scale. Thanks to the high-linearity of flexible sensors, we address this challenge by proposing a novel Sim2Real Mocap solution based on domain adaptation, eliminating the need for labeled data yet achieving comparable accuracy to supervised learning. Our solution relies on a novel Support-based Domain Adaptation method, namely SuDA, which aligns the supports of the predictive functions rather than the instance-dependent distributions between the source and target domains. Extensive experimental results demonstrate the effectiveness of our method andits superiority over state-of-the-art distribution-based domain adaptation methods in our task.
Authors: Chunlin Qiu, Yiheng Duan, Lingchen Zhao, Qian Wang
Abstract: Transfer-based attacks craft adversarial examples utilizing a white-box surrogate model to compromise various black-box target models, posing significant threats to many real-world applications. However, existing transfer attacks suffer from either weak transferability or expensive computation. To bridge the gap, we propose a novel sample-based attack, named neighborhood conditional sampling (NCS), which enjoys high transferability with lightweight computation. Inspired by the observation that flat maxima result in better transferability, NCS is formulated as a max-min bi-level optimization problem to seek adversarial regions with high expected adversarial loss and small standard deviations. Specifically, due to the inner minimization problem being computationally intensive to resolve, and affecting the overall transferability, we propose a momentum-based previous gradient inversion approximation (PGIA) method to effectively solve the inner problem without any computation cost. In addition, we prove that two newly proposed attacks, which achieve flat maxima for better transferability, are actually specific cases of NCS under particular conditions. Extensive experiments demonstrate that NCS efficiently generates highly transferable adversarial examples, surpassing the current best method in transferability while requiring only 50% of the computational cost. Additionally, NCS can be seamlessly integrated with other methods to further enhance transferability.
Authors: Fuheng Zhou, Dikai Wei, Ye Fan, Yulong Huang, Yonggang Zhang
Abstract: Although deep learning based models for underwater image enhancement have achieved good performance, they face limitations in both lightweight and effectiveness, which prevents their deployment and application on resource-constrained platforms. Moreover, most existing deep learning based models use data compression to get high-level semantic information in latent space instead of using the original information. Therefore, they require decoder blocks to generate the details of the output. This requires additional computational cost. In this paper, a lightweight network named lightweight selective attention network (LSNet) based on the top-k selective attention and transmission maps mechanism is proposed. The proposed model achieves a PSNR of 97\% with only 7K parameters compared to a similar attention-based model. Extensive experiments show that the proposed LSNet achieves excellent performance in state-of-the-art models with significantly fewer parameters and computational resources. The code is available at https://github.com/FuhengZhou/LSNet}{https://github.com/FuhengZhou/LSNet.
URLs: https://github.com/FuhengZhou/LSNet, https://github.com/FuhengZhou/LSNet.
Authors: Lan Wu, Xuebin Wang, Ruijuan Chu, Guangyi Liu, Yingchun Chen, Jing Zhang, Linyu Wang
Abstract: Accurate multi-step flight trajectory prediction plays an important role in Air Traffic Control, which can ensure the safety of air transportation. Two main issues limit the flight trajectory prediction performance of existing works. The first issue is the negative impact on prediction accuracy caused by the significant differences in data range. The second issue is that real-world flight trajectories involve underlying temporal dependencies, and existing methods fail to reveal the hidden complex temporal variations and only extract features from one single time scale. To address the above issues, we propose FlightPatchNet, a multi-scale patch network with differential coding for flight trajectory prediction. Specifically, FlightPatchNet first utilizes the differential coding to encode the original values of longitude and latitude into first-order differences and generates embeddings for all variables at each time step. Then, a global temporal attention is introduced to explore the dependencies between different time steps. To fully explore the diverse temporal patterns in flight trajectories, a multi-scale patch network is delicately designed to serve as the backbone. The multi-scale patch network exploits stacked patch mixer blocks to capture inter- and intra-patch dependencies under different time scales, and further integrates multi-scale temporal features across different scales and variables. Finally, FlightPatchNet ensembles multiple predictors to make direct multi-step prediction. Extensive experiments on ADS-B datasets demonstrate that our model outperforms the competitive baselines. Code is available at: https://github.com/FlightTrajectoryResearch/FlightPatchNet.
URLs: https://github.com/FlightTrajectoryResearch/FlightPatchNet.
Authors: Phong Tran, Egor Zakharov, Long-Nhat Ho, Liwen Hu, Adilbek Karmanov, Aviral Agarwal, McLean Goldwhite, Ariana Bermudez Venegas, Anh Tuan Tran, Hao Li
Abstract: We introduce VOODOO XP: a 3D-aware one-shot head reenactment method that can generate highly expressive facial expressions from any input driver video and a single 2D portrait. Our solution is real-time, view-consistent, and can be instantly used without calibration or fine-tuning. We demonstrate our solution on a monocular video setting and an end-to-end VR telepresence system for two-way communication. Compared to 2D head reenactment methods, 3D-aware approaches aim to preserve the identity of the subject and ensure view-consistent facial geometry for novel camera poses, which makes them suitable for immersive applications. While various facial disentanglement techniques have been introduced, cutting-edge 3D-aware neural reenactment techniques still lack expressiveness and fail to reproduce complex and fine-scale facial expressions. We present a novel cross-reenactment architecture that directly transfers the driver's facial expressions to transformer blocks of the input source's 3D lifting module. We show that highly effective disentanglement is possible using an innovative multi-stage self-supervision approach, which is based on a coarse-to-fine strategy, combined with an explicit face neutralization and 3D lifted frontalization during its initial training stage. We further integrate our novel head reenactment solution into an accessible high-fidelity VR telepresence system, where any person can instantly build a personalized neural head avatar from any photo and bring it to life using the headset. We demonstrate state-of-the-art performance in terms of expressiveness and likeness preservation on a large set of diverse subjects and capture conditions.
Authors: Gabriel Moreira, Alexander Hauptmann, Manuel Marques, Jo\~ao Paulo Costeira
Abstract: Learning representations that capture rich semantic relationships and accommodate propositional calculus poses a significant challenge. Existing approaches are either contrastive, lacking theoretical guarantees, or fall short in effectively representing the partial orders inherent to rich visual-semantic hierarchies. In this paper, we propose a novel approach for learning visual representations that not only conform to a specified semantic structure but also facilitate probabilistic propositional reasoning. Our approach is based on a new nuclear norm-based loss. We show that its minimum encodes the spectral geometry of the semantics in a subspace lattice, where logical propositions can be represented by projection operators.
Authors: Shuaixin Liu, Kunqian Li, Yilin Ding
Abstract: In this paper, we propose a novel underwater image enhancement method, by utilizing the multi-guided diffusion model for iterative enhancement. Unlike other image enhancement tasks, underwater images suffer from the unavailability of real reference images. Although existing works exploit synthetic images, manually selected well-enhanced images as reference images, to train enhancement networks, their enhancement performance always comes with subjective preferences that are inherited from the manual selection. To address this issue, we also use the image synthesis strategy, but the synthetic images derive from in-air natural images degraded into corresponding underwater images, guided by the underwater domain. Based on this strategy, the diffusion model can learn the prior knowledge of image enhancement from the underwater degradation domain to the real in-air natural domain. However, it is inevitable to fine-tune the model to suit downstream tasks, and this may erase the prior knowledge. To mitigate this, we combine the prior knowledge from the in-air natural domain with Contrastive Language-Image Pretraining (CLIP) to train a classifier for controlling the diffusion model generation process. Moreover, for image enhancement tasks, we find that the image-to-image diffusion model and the CLIP-Classifier mainly act in the high-frequency region during the fine-tuning process. Therefore, we propose a fast fine-tuning strategy focusing on the high-frequency region, which can be up to 10 times faster than the traditional strategy. Extensive experiments demonstrate that our method, abbreviated as CLIP-UIE, exhibit a more natural appearance.
Authors: Yuzhuo Chen, Zetong Chen, Yunuo An, Chenyang Lu, Xu Qiao
Abstract: The precise categorization of white blood cell (WBC) is crucial for diagnosing blood-related disorders. However, manual analysis in clinical settings is time-consuming, labor-intensive, and prone to errors. Numerous studies have employed machine learning and deep learning techniques to achieve objective WBC classification, yet these studies have not fully utilized the information of WBC images. Therefore, our motivation is to comprehensively utilize the morphological information and high-level semantic information of WBC images to achieve accurate classification of WBC. In this study, we propose a novel dual-branch network Dual Attention Feature Fusion Network (DAFFNet), which for the first time integrates the high-level semantic features with morphological features of WBC to achieve accurate classification. Specifically, we introduce a dual attention mechanism, which enables the model to utilize the channel features and spatially localized features of the image more comprehensively. Morphological Feature Extractor (MFE), comprising Morphological Attributes Predictor (MAP) and Morphological Attributes Encoder (MAE), is proposed to extract the morphological features of WBC. We also implement Deep-supervised Learning (DSL) and Semi-supervised Learning (SSL) training strategies for MAE to enhance its performance. Our proposed network framework achieves 98.77%, 91.30%, 98.36%, 99.71%, 98.45%, and 98.85% overall accuracy on the six public datasets PBC, LISC, Raabin-WBC, BCCD, LDWBC, and Labelled, respectively, demonstrating superior effectiveness compared to existing studies. The results indicate that the WBC classification combining high-level semantic features and low-level morphological features is of great significance, which lays the foundation for objective and accurate classification of WBC in microscopic blood cell images.
Authors: Qian Wang, Chen Li, Yuchen Luo, Hefei Ling, Ping Li, Jiazhong Chen, Shijuan Huang, Ning Yu
Abstract: As a defense strategy against adversarial attacks, adversarial detection aims to identify and filter out adversarial data from the data flow based on discrepancies in distribution and noise patterns between natural and adversarial data. Although previous detection methods achieve high performance in detecting gradient-based adversarial attacks, new attacks based on generative models with imbalanced and anisotropic noise patterns evade detection. Even worse, existing techniques either necessitate access to attack data before deploying a defense or incur a significant time cost for inference, rendering them impractical for defending against newly emerging attacks that are unseen by defenders. In this paper, we explore the proximity relationship between adversarial noise distributions and demonstrate the existence of an open covering for them. By learning to distinguish this open covering from the distribution of natural data, we can develop a detector with strong generalization capabilities against all types of adversarial attacks. Based on this insight, we heuristically propose Perturbation Forgery, which includes noise distribution perturbation, sparse mask generation, and pseudo-adversarial data production, to train an adversarial detector capable of detecting unseen gradient-based, generative-model-based, and physical adversarial attacks, while remaining agnostic to any specific models. Comprehensive experiments conducted on multiple general and facial datasets, with a wide spectrum of attacks, validate the strong generalization of our method.
Authors: Shiyu Xia, Junyu Xiong, Haoyu Dong, Jianbo Zhao, Yuzhang Tian, Mengyu Zhou, Yeye He, Shi Han, Dongmei Zhang
Abstract: This paper explores capabilities of Vision Language Models on spreadsheet comprehension. We propose three self-supervised challenges with corresponding evaluation metrics to comprehensively evaluate VLMs on Optical Character Recognition (OCR), spatial perception, and visual format recognition. Additionally, we utilize the spreadsheet table detection task to assess the overall performance of VLMs by integrating these challenges. To probe VLMs more finely, we propose three spreadsheet-to-image settings: column width adjustment, style change, and address augmentation. We propose variants of prompts to address the above tasks in different settings. Notably, to leverage the strengths of VLMs in understanding text rather than two-dimensional positioning, we propose to decode cell values on the four boundaries of the table in spreadsheet boundary detection. Our findings reveal that VLMs demonstrate promising OCR capabilities but produce unsatisfactory results due to cell omission and misalignment, and they notably exhibit insufficient spatial and format recognition skills, motivating future work to enhance VLMs' spreadsheet data comprehension capabilities using our methods to generate extensive spreadsheet-image pairs in various settings.
Authors: Shelly Golan, Roy Ganz, Michael Elad
Abstract: The recently introduced Consistency models pose an efficient alternative to diffusion algorithms, enabling rapid and good quality image synthesis. These methods overcome the slowness of diffusion models by directly mapping noise to data, while maintaining a (relatively) simpler training. Consistency models enable a fast one- or few-step generation, but they typically fall somewhat short in sample quality when compared to their diffusion origins. In this work we propose a novel and highly effective technique for post-processing Consistency-based generated images, enhancing their perceptual quality. Our approach utilizes a joint classifier-discriminator model, in which both portions are trained adversarially. While the classifier aims to grade an image based on its assignment to a designated class, the discriminator portion of the very same network leverages the softmax values to assess the proximity of the input image to the targeted data manifold, thereby serving as an Energy-based Model. By employing example-specific projected gradient iterations under the guidance of this joint machine, we refine synthesized images and achieve an improved FID scores on the ImageNet 64x64 dataset for both Consistency-Training and Consistency-Distillation techniques.
Authors: Tianyi Chen, Jianfu Zhang, Yan Hong, Yiyi Zhang, Liqing Zhang
Abstract: Image inpainting, the task of reconstructing missing segments in corrupted images using available data, faces challenges in ensuring consistency and fidelity, especially under information-scarce conditions. Traditional evaluation methods, heavily dependent on the existence of unmasked reference images, inherently favor certain inpainting outcomes, introducing biases. Addressing this issue, we introduce an innovative evaluation paradigm that utilizes a self-supervised metric based on multiple re-inpainting passes. This approach, diverging from conventional reliance on direct comparisons in pixel or feature space with original images, emphasizes the principle of self-consistency to enable the exploration of various viable inpainting solutions, effectively reducing biases. Our extensive experiments across numerous benchmarks validate the alignment of our evaluation method with human judgment.
Authors: Mingshuang Luo, Ruibing Hou, Hong Chang, Zimo Liu, Yaowei Wang, Shiguang Shan
Abstract: This paper presents $M^3$GPT, an advanced \textbf{M}ultimodal, \textbf{M}ultitask framework for \textbf{M}otion comprehension and generation. $M^3$GPT operates on three fundamental principles. The first focuses on creating a unified representation space for various motion-relevant modalities. We employ discrete vector quantization for multimodal control and generation signals, such as text, music and motion/dance, enabling seamless integration into a large language model (LLM) with a single vocabulary. The second involves modeling model generation directly in the raw motion space. This strategy circumvents the information loss associated with discrete tokenizer, resulting in more detailed and comprehensive model generation. Third, $M^3$GPT learns to model the connections and synergies among various motion-relevant tasks. Text, the most familiar and well-understood modality for LLMs, is utilized as a bridge to establish connections between different motion tasks, facilitating mutual reinforcement. To our knowledge, $M^3$GPT is the first model capable of comprehending and generating motions based on multiple signals. Extensive experiments highlight $M^3$GPT's superior performance across various motion-relevant tasks and its powerful zero-shot generalization capabilities for extremely challenging tasks.
Authors: Jhen Hsieh
Abstract: In this paper, we present a neural network-based approach for tracking and reconstructing the trajectories of baseball pitches from 2D video footage to 3D coordinates. We utilize OpenCV's CSRT algorithm to accurately track the baseball and fixed reference points in 2D video frames. These tracked pixel coordinates are then used as input features for our neural network model, which comprises multiple fully connected layers to map the 2D coordinates to 3D space. The model is trained on a dataset of labeled trajectories using a mean squared error loss function and the Adam optimizer, optimizing the network to minimize prediction errors. Our experimental results demonstrate that this approach achieves high accuracy in reconstructing 3D trajectories from 2D inputs. This method shows great potential for applications in sports analysis, coaching, and enhancing the accuracy of trajectory predictions in various sports.
Authors: Dae Ung Jo, Kyuewang Lee, JaeHo Chung, Jin Young Choi
Abstract: Securing a sufficient amount of paired data is important to train an image-text retrieval (ITR) model, but collecting paired data is very expensive. To address this issue, in this paper, we propose an active learning algorithm for ITR that can collect paired data cost-efficiently. Previous studies assume that image-text pairs are given and their category labels are asked to the annotator. However, in the recent ITR studies, the importance of category label is decreased since a retrieval model can be trained with only image-text pairs. For this reason, we set up an active learning scenario where unpaired images (or texts) are given and the annotator provides corresponding texts (or images) to make paired data. The key idea of the proposed AL algorithm is to select unpaired images (or texts) that can be hard negative samples for existing texts (or images). To this end, we introduce a novel scoring function to choose hard negative samples. We validate the effectiveness of the proposed method on Flickr30K and MS-COCO datasets.
Authors: Xiaoyang Chen, Hao Zheng, Yifang Xie, Yuncong Ma, Tengfei Li
Abstract: Current methods for developing foundation models in medical image segmentation rely on two primary assumptions: a fixed set of classes and the immediate availability of a substantial and diverse training dataset. However, this can be impractical due to the evolving nature of imaging technology and patient demographics, as well as labor-intensive data curation, limiting their practical applicability and scalability. To address these challenges, we introduce a novel segmentation paradigm enabling the segmentation of a variable number of classes within a single classifier-free network, featuring an architecture independent of class number. This network is trained using contrastive learning and produces discriminative feature representations that facilitate straightforward interpretation. Additionally, we integrate this strategy into a knowledge distillation-based incremental learning framework, facilitating the gradual assimilation of new information from non-stationary data streams while avoiding catastrophic forgetting. Our approach provides a unified solution for tackling both class- and domain-incremental learning scenarios. We demonstrate the flexibility of our method in handling varying class numbers within a unified network and its capacity for incremental learning. Experimental results on an incompletely annotated, multi-modal, multi-source dataset for medical image segmentation underscore its superiority over state-of-the-art alternative approaches.
Authors: Silky Singh, Surgan Jandial, Simra Shahid, Abhinav Java
Abstract: Text-conditioned style transfer enables users to communicate their desired artistic styles through text descriptions, offering a new and expressive means of achieving stylization. In this work, we evaluate the text-conditioned image editing and style transfer techniques on their fine-grained understanding of user prompts for precise "local" style transfer. We find that current methods fail to accomplish localized style transfers effectively, either failing to localize style transfer to certain regions in the image, or distorting the content and structure of the input image. To this end, we carefully design an end-to-end pipeline that guarantees local style transfer according to users' intent. Further, we substantiate the effectiveness of our approach through quantitative and qualitative analysis. The project code is available at: https://github.com/silky1708/local-style-transfer.
Authors: Changhoon Kim, Kyle Min, Yezhou Yang
Abstract: In the evolving landscape of text-to-image (T2I) diffusion models, the remarkable capability to generate high-quality images from textual descriptions faces challenges with the potential misuse of reproducing sensitive content. To address this critical issue, we introduce Robust Adversarial Concept Erase (RACE), a novel approach designed to mitigate these risks by enhancing the robustness of concept erasure method for T2I models. RACE utilizes a sophisticated adversarial training framework to identify and mitigate adversarial text embeddings, significantly reducing the Attack Success Rate (ASR). Impressively, RACE achieves a 30 percentage point reduction in ASR for the ``nudity'' concept against the leading white-box attack method. Our extensive evaluations demonstrate RACE's effectiveness in defending against both white-box and black-box attacks, marking a significant advancement in protecting T2I diffusion models from generating inappropriate or misleading imagery. This work underlines the essential need for proactive defense measures in adapting to the rapidly advancing field of adversarial challenges.
Authors: James Maier, Nishanth Mohankumar
Abstract: This study explores the potential of open-source video conditional generation models as encoders for downstream tasks, focusing on instance segmentation using the BAIR Robot Pushing Dataset. The researchers propose using video prediction models as general visual encoders, leveraging their ability to capture critical spatial and temporal information which is essential for tasks such as instance segmentation. Inspired by human vision studies, particularly Gestalts principle of common fate, the approach aims to develop a latent space representative of motion from images to effectively discern foreground from background information. The researchers utilize a 3D Vector-Quantized Variational Autoencoder 3D VQVAE video generative encoder model conditioned on an input frame, coupled with downstream segmentation tasks. Experiments involve adapting pre-trained video generative models, analyzing their latent spaces, and training custom decoders for foreground-background segmentation. The findings demonstrate promising results in leveraging generative pretext learning for downstream tasks, working towards enhanced scene analysis and segmentation in computer vision applications.
Authors: Jinlin Liu, Kai Yu, Mengyang Feng, Xiefang Guo, Miaomiao Cui
Abstract: Recent advancements in human video synthesis have enabled the generation of high-quality videos through the application of stable diffusion models. However, existing methods predominantly concentrate on animating solely the human element (the foreground) guided by pose information, while leaving the background entirely static. Contrary to this, in authentic, high-quality videos, backgrounds often dynamically adjust in harmony with foreground movements, eschewing stagnancy. We introduce a technique that concurrently learns both foreground and background dynamics by segregating their movements using distinct motion representations. Human figures are animated leveraging pose-based motion, capturing intricate actions. Conversely, for backgrounds, we employ sparse tracking points to model motion, thereby reflecting the natural interaction between foreground activity and environmental changes. Training on real-world videos enhanced with this innovative motion depiction approach, our model generates videos exhibiting coherent movement in both foreground subjects and their surrounding contexts. To further extend video generation to longer sequences without accumulating errors, we adopt a clip-by-clip generation strategy, introducing global features at each step. To ensure seamless continuity across these segments, we ingeniously link the final frame of a produced clip with input noise to spawn the succeeding one, maintaining narrative flow. Throughout the sequential generation process, we infuse the feature representation of the initial reference image into the network, effectively curtailing any cumulative color inconsistencies that may otherwise arise. Empirical evaluations attest to the superiority of our method in producing videos that exhibit harmonious interplay between foreground actions and responsive background dynamics, surpassing prior methodologies in this regard.
Authors: Neha Kalibhat, Priyatham Kattakinda, Arman Zarei, Nikita Seleznev, Samuel Sharpe, Senthil Kumar, Soheil Feizi
Abstract: Vision transformers have established a precedent of patchifying images into uniformly-sized chunks before processing. We hypothesize that this design choice may limit models in learning comprehensive and compositional representations from visual data. This paper explores the notion of providing semantically-meaningful visual tokens to transformer encoders within a vision-language pre-training framework. Leveraging off-the-shelf segmentation and scene-graph models, we extract representations of instance segmentation masks (referred to as tangible tokens) and relationships and actions (referred to as intangible tokens). Subsequently, we pre-train a vision-side transformer by incorporating these newly extracted tokens and aligning the resultant embeddings with caption embeddings from a text-side encoder. To capture the structural and semantic relationships among visual tokens, we introduce additive attention weights, which are used to compute self-attention scores. Our experiments on COCO demonstrate notable improvements over ViTs in learned representation quality across text-to-image (+47%) and image-to-text retrieval (+44%) tasks. Furthermore, we showcase the advantages on compositionality benchmarks such as ARO (+18%) and Winoground (+10%).
Authors: Huayuan Ye, Shenzhuo Zhang, Shiqi Jiang, Jing Liao, Shuhang Gu, Changbo Wang, Chenhui Li
Abstract: Image steganography can hide information in a host image and obtain a stego image that is perceptually indistinguishable from the original one. This technique has tremendous potential in scenarios like copyright protection, information retrospection, etc. Some previous studies have proposed to enhance the robustness of the methods against image disturbances to increase their applicability. However, they generally cannot achieve a satisfying balance between the steganography quality and robustness. In this paper, we focus on the issue of QR Code steganography that is robust to real-world printing and photography. Different from common image steganography, QR Code steganography aims to embed a non-natural image into a natural image and the restored QR Code is required to be recognizable, which increases the difficulty of data concealing and revealing. Inspired by the recent developments in transformer-based vision models, we discover that the tokenized representation of images is naturally suitable for steganography. In this paper, we propose a novel QR Code embedding framework, called Printing and Photography Robust Steganography (PPRSteg), which is competent to hide QR Code in a host image with unperceivable changes and can restore it even if the stego image is printed out and photoed. We outline a transition process to reduce the artifacts in stego images brought by QR Codes. We also propose a steganography model based on normalizing flow, which combines the attention mechanism to enhance its performance. To our best knowledge, this is the first work that integrates the advantages of transformer models into normalizing flow. We conduct comprehensive and detailed experiments to demonstrate the effectiveness of our method and the result shows that PPRSteg has great potential in robust, secure and high-quality QR Code steganography.
Authors: Lin Zhu, Yifeng Yang, Qinying Gu, Xinbing Wang, Chenghu Zhou, Nanyang Ye
Abstract: Recent vision-language pre-trained models (VL-PTMs) have shown remarkable success in open-vocabulary tasks. However, downstream use cases often involve further fine-tuning of VL-PTMs, which may distort their general knowledge and impair their ability to handle distribution shifts. In real-world scenarios, machine learning systems inevitably encounter both covariate shifts (e.g., changes in image styles) and semantic shifts (e.g., test-time unseen classes). This highlights the importance of enhancing out-of-distribution (OOD) generalization on covariate shifts and simultaneously detecting semantic-shifted unseen classes. Thus a critical but underexplored question arises: How to improve VL-PTMs' generalization ability to closed-set OOD data, while effectively detecting open-set unseen classes during fine-tuning? In this paper, we propose a novel objective function of OOD detection that also serves to improve OOD generalization. We show that minimizing the gradient magnitude of energy scores on training data leads to domain-consistent Hessians of classification loss, a strong indicator for OOD generalization revealed by theoretical analysis. Based on this finding, we have developed a unified fine-tuning framework that allows for concurrent optimization of both tasks. Extensive experiments have demonstrated the superiority of our method. The code is available at https://github.com/LinLLLL/CRoFT.
Authors: Chau Pham, Bryan A. Plummer
Abstract: Multi-Channel Imaging (MCI) contains an array of challenges for encoding useful feature representations not present in traditional images. For example, images from two different satellites may both contain RGB channels, but the remaining channels can be different for each imaging source. Thus, MCI models must support a variety of channel configurations at test time. Recent work has extended traditional visual encoders for MCI, such as Vision Transformers (ViT), by supplementing pixel information with an encoding representing the channel configuration. However, these methods treat each channel equally, i.e., they do not consider the unique properties of each channel type, which can result in needless and potentially harmful redundancies in the learned features. For example, if RGB channels are always present, the other channels can focus on extracting information that cannot be captured by the RGB channels. To this end, we propose DiChaViT, which aims to enhance the diversity in the learned features of MCI-ViT models. This is achieved through a novel channel sampling strategy that encourages the selection of more distinct channel sets for training. Additionally, we employ regularization and initialization techniques to increase the likelihood that new information is learned from each channel. Many of our improvements are architecture agnostic and could be incorporated into new architectures as they are developed. Experiments on both satellite and cell microscopy datasets, CHAMMI, JUMP-CP, and So2Sat, report DiChaViT yields a 1.5-5.0% gain over the state-of-the-art.
Authors: FNU Shivam, Megan Leight, Mary Kate Kelly, Claire Davis, Kelsey Clodfelter, Jacob Thrasher, Yenumula Reddy, Prashnna Gyawali
Abstract: The study of Maya hieroglyphic writing unlocks the rich history of cultural and societal knowledge embedded within this ancient civilization's visual narrative. Artificial Intelligence (AI) offers a novel lens through which we can translate these inscriptions, with the potential to allow non-specialists access to reading these texts and to aid in the decipherment of those hieroglyphs which continue to elude comprehensive interpretation. Toward this, we leverage a foundational model to segment Maya hieroglyphs from an open-source digital library dedicated to Maya artifacts. Despite the initial promise of publicly available foundational segmentation models, their effectiveness in accurately segmenting Maya hieroglyphs was initially limited. Addressing this challenge, our study involved the meticulous curation of image and label pairs with the assistance of experts in Maya art and history, enabling the fine-tuning of these foundational models. This process significantly enhanced model performance, illustrating the potential of fine-tuning approaches and the value of our expanding dataset. We plan to open-source this dataset for encouraging future research, and eventually to help make the hieroglyphic texts legible to a broader community, particularly for Maya heritage community members.
Authors: Yawen Zou, Chunzhi Gu, Jun Yu, Shangce Gao, Chao Zhang
Abstract: Black-Box unsupervised domain adaptation (BBUDA) learns knowledge only with the prediction of target data from the source model without access to the source data and source model, which attempts to alleviate concerns about the privacy and security of data. However, incorrect pseudo-labels are prevalent in the prediction generated by the source model due to the cross-domain discrepancy, which may substantially degrade the performance of the target model. To address this problem, we propose a novel approach that incrementally selects high-confidence pseudo-labels to improve the generalization ability of the target model. Specifically, we first generate pseudo-labels using a source model and train a crude target model by a vanilla BBUDA method. Second, we iteratively select high-confidence data from the low-confidence data pool by thresholding the softmax probabilities, prototype labels, and intra-class similarity. Then, we iteratively train a stronger target network based on the crude target model to correct the wrongly labeled samples to improve the accuracy of the pseudo-label. Experimental results demonstrate that the proposed method achieves state-of-the-art black-box unsupervised domain adaptation performance on three benchmark datasets.
Authors: Taichi Uchida, Yoshihiro Kanamori, Yuki Endo
Abstract: Achieving aesthetically pleasing photography necessitates attention to multiple factors, including composition and capture conditions, which pose challenges to novices. Prior research has explored the enhancement of photo aesthetics post-capture through 2D manipulation techniques; however, these approaches offer limited search space for aesthetics. We introduce a pioneering method that employs 3D operations to simulate the conditions at the moment of capture retrospectively. Our approach extrapolates the input image and then reconstructs the 3D scene from the extrapolated image, followed by an optimization to identify camera parameters and image aspect ratios that yield the best 3D view with enhanced aesthetics. Comparative qualitative and quantitative assessments reveal that our method surpasses traditional 2D editing techniques with superior aesthetics.
Authors: Hanting Li, Hongjing Niu, Feng Zhao
Abstract: Micro-expression recognition (MER) has drawn increasing attention in recent years due to its potential applications in intelligent medical and lie detection. However, the shortage of annotated data has been the major obstacle to further improve deep-learning based MER methods. Intuitively, utilizing sufficient macro-expression data to promote MER performance seems to be a feasible solution. However, the facial patterns of macro-expressions and micro-expressions are significantly different, which makes naive transfer learning methods difficult to deploy directly. To tacle this issue, we propose a generalized transfer learning paradigm, called \textbf{MA}cro-expression \textbf{TO} \textbf{MI}cro-expression (MA2MI). Under our paradigm, networks can learns the ability to represent subtle facial movement by reconstructing future frames. In addition, we also propose a two-branch micro-action network (MIACNet) to decouple facial position features and facial action features, which can help the network more accurately locate facial action locations. Extensive experiments on three popular MER benchmarks demonstrate the superiority of our method.
Authors: Shugo Yamashita, Masaaki Ikehara
Abstract: Removing rain artifacts in images is recognized as a significant issue. In this field, deep learning-based approaches, such as convolutional neural networks (CNNs) and Transformers, have succeeded. Recently, State Space Models (SSMs) have exhibited superior performance across various tasks in both natural language processing and image processing due to their ability to model long-range dependencies. This study introduces SSM to rain removal and proposes a Deraining Frequency-Enhanced State Space Model (DFSSM). To effectively remove rain streaks, which produce high-intensity frequency components in specific directions, we employ frequency domain processing concurrently with SSM. Additionally, we develop a novel mixed-scale gated-convolutional block, which uses convolutions with multiple kernel sizes to capture various scale degradations effectively and integrates a gating mechanism to manage the flow of information. Finally, experiments on synthetic and real-world rainy image datasets show that our method surpasses state-of-the-art methods.
Authors: Qiguang Chen, Libo Qin, Jin Zhang, Zhi Chen, Xiao Xu, Wanxiang Che
Abstract: Multi-modal Chain-of-Thought (MCoT) requires models to leverage knowledge from both textual and visual modalities for step-by-step reasoning, which gains increasing attention. Nevertheless, the current MCoT benchmark still faces some challenges: (1) absence of visual modal reasoning, (2) single-step visual modal reasoning, and (3) Domain missing, thereby hindering the development of MCoT. Motivated by this, we introduce a novel benchmark (M$^3$CoT) to address the above challenges, advancing the multi-domain, multi-step, and multi-modal CoT. Additionally, we conduct a thorough evaluation involving abundant MCoT approaches on Vision Large Language Models (VLLMs). In addition, we highlight that the current VLLMs still struggle to correctly reason in M$^3$CoT and there remains a large gap between existing VLLMs and human performance in M$^3$CoT, despite their superior results on previous MCoT benchmarks. To our knowledge, we take the first meaningful step toward the multi-domain, multi-step, and multi-modal scenario in MCoT. We hope that M$^3$CoT can serve as a valuable resource, providing a pioneering foundation in multi-domain, multi-step, multi-modal chain-of-thought research.
Authors: Chathura Wimalasiri, Prasan Kumar Sahoo
Abstract: In response to the increasing demand for efficient and non-invasive methods to estimate food weight, this paper presents a vision-based approach utilizing 2D images. The study employs a dataset of 2380 images comprising fourteen different food types in various portions, orientations, and containers. The proposed methodology integrates deep learning and computer vision techniques, specifically employing Faster R-CNN for food detection and MobileNetV3 for weight estimation. The detection model achieved a mean average precision (mAP) of 83.41\%, an average Intersection over Union (IoU) of 91.82\%, and a classification accuracy of 100\%. For weight estimation, the model demonstrated a root mean squared error (RMSE) of 6.3204, a mean absolute percentage error (MAPE) of 0.0640\%, and an R-squared value of 98.65\%. The study underscores the potential applications of this technology in healthcare for nutrition counseling, fitness and wellness for dietary intake assessment, and smart food storage solutions to reduce waste. The results indicate that the combination of Faster R-CNN and MobileNetV3 provides a robust framework for accurate food weight estimation from 2D images, showcasing the synergy of computer vision and deep learning in practical applications.
Authors: Haoru Tan, Chuang Wang, Xu-Yao Zhang, Cheng-Lin Liu
Abstract: Graph matching is a fundamental tool in computer vision and pattern recognition. In this paper, we introduce an algorithm for graph matching based on the proximal operator, referred to as differentiable proximal graph matching (DPGM). Specifically, we relax and decompose the quadratic assignment problem for the graph matching into a sequence of convex optimization problems. The whole algorithm can be considered as a differentiable map from the graph affinity matrix to the prediction of node correspondence. Therefore, the proposed method can be organically integrated into an end-to-end deep learning framework to jointly learn both the deep feature representation and the graph affinity matrix. In addition, we provide a theoretical guarantee to ensure the proposed method converges to a stable point with a reasonable number of iterations. Numerical experiments show that PGM outperforms existing graph matching algorithms on diverse datasets such as synthetic data, and CMU House. Meanwhile, PGM can fully harness the capability of deep feature extractors and achieve state-of-art performance on PASCAL VOC keypoints.
Authors: Rongyu Zhang, Aosong Cheng, Yulin Luo, Gaole Dai, Huanrui Yang, Jiaming Liu, Ran Xu, Li Du, Yuan Du, Yanbing Jiang, Shanghang Zhang
Abstract: Continual Test-Time Adaptation (CTTA), which aims to adapt the pre-trained model to ever-evolving target domains, emerges as an important task for vision models. As current vision models appear to be heavily biased towards texture, continuously adapting the model from one domain distribution to another can result in serious catastrophic forgetting. Drawing inspiration from the human visual system's adeptness at processing both shape and texture according to the famous Trichromatic Theory, we explore the integration of a Mixture-of-Activation-Sparsity-Experts (MoASE) as an adapter for the CTTA task. Given the distinct reaction of neurons with low/high activation to domain-specific/agnostic features, MoASE decomposes the neural activation into high-activation and low-activation components with a non-differentiable Spatial Differentiate Dropout (SDD). Based on the decomposition, we devise a multi-gate structure comprising a Domain-Aware Gate (DAG) that utilizes domain information to adaptive combine experts that process the post-SDD sparse activations of different strengths, and the Activation Sparsity Gate (ASG) that adaptively assigned feature selection threshold of the SDD for different experts for more precise feature decomposition. Finally, we introduce a Homeostatic-Proximal (HP) loss to bypass the error accumulation problem when continuously adapting the model. Extensive experiments on four prominent benchmarks substantiate that our methodology achieves state-of-the-art performance in both classification and segmentation CTTA tasks. Our code is now available at https://github.com/RoyZry98/MoASE-Pytorch.
Authors: Yong Li, Han Gao
Abstract: Neural networks are widely known to be vulnerable to backdoor attacks, a method that poisons a portion of the training data to make the target model perform well on normal data sets, while outputting attacker-specified or random categories on the poisoned samples. Backdoor attacks are full of threats. Poisoned samples are becoming more and more similar to corresponding normal samples, and even the human eye cannot easily distinguish them. On the other hand, the accuracy of models carrying backdoors on normal samples is no different from that of clean models.In this article, by observing the characteristics of backdoor attacks, We provide a new model training method (PT) that freezes part of the model to train a model that can isolate suspicious samples. Then, on this basis, a clean model is fine-tuned to resist backdoor attacks.
Authors: Shuangpeng Han, Ziyu Wang, Mengmi Zhang
Abstract: Biological motion perception (BMP) refers to humans' ability to perceive and recognize the actions of living beings solely from their motion patterns, sometimes as minimal as those depicted on point-light displays. While humans excel at these tasks without any prior training, current AI models struggle with poor generalization performance. To close this research gap, we propose the Motion Perceiver (MP). MP solely relies on patch-level optical flows from video clips as inputs. During training, it learns prototypical flow snapshots through a competitive binding mechanism and integrates invariant motion representations to predict action labels for the given video. During inference, we evaluate the generalization ability of all AI models and humans on 62,656 video stimuli spanning 24 BMP conditions using point-light displays in neuroscience. Remarkably, MP outperforms all existing AI models with a maximum improvement of 29% in top-1 action recognition accuracy on these conditions. Moreover, we benchmark all AI models in point-light displays of two standard video datasets in computer vision. MP also demonstrates superior performance in these cases. More interestingly, via psychophysics experiments, we found that MP recognizes biological movements in a way that aligns with human behavioural data. All data and code will be made public.
Authors: Nicole Heng Yim Oo, Min Hun Lee, Jeong Hoon Lim
Abstract: Algorithmic detection of facial palsy offers the potential to improve current practices, which usually involve labor-intensive and subjective assessment by clinicians. In this paper, we present a multimodal fusion-based deep learning model that utilizes unstructured data (i.e. an image frame with facial line segments) and structured data (i.e. features of facial expressions) to detect facial palsy. We then contribute to a study to analyze the effect of different data modalities and the benefits of a multimodal fusion-based approach using videos of 21 facial palsy patients. Our experimental results show that among various data modalities (i.e. unstructured data - RGB images and images of facial line segments and structured data - coordinates of facial landmarks and features of facial expressions), the feed-forward neural network using features of facial expression achieved the highest precision of 76.22 while the ResNet-based model using images of facial line segments achieved the highest recall of 83.47. When we leveraged both images of facial line segments and features of facial expressions, our multimodal fusion-based deep learning model slightly improved the precision score to 77.05 at the expense of a decrease in the recall score.
Authors: Linhao Zhong, Yan Hong, Wentao Chen, Binglin Zhou, Yiyi Zhang, Jianfu Zhang, Liqing Zhang
Abstract: Text-to-image generation models have seen considerable advancement, catering to the increasing interest in personalized image creation. Current customization techniques often necessitate users to provide multiple images (typically 3-5) for each customized object, along with the classification of these objects and descriptive textual prompts for scenes. This paper questions whether the process can be made more user-friendly and the customization more intricate. We propose a method where users need only provide images along with text for each customization topic, and necessitates only a single image per visual concept. We introduce the concept of a ``multi-modal prompt'', a novel integration of text and images tailored to each customization concept, which simplifies user interaction and facilitates precise customization of both objects and scenes. Our proposed paradigm for customized text-to-image generation surpasses existing finetune-based methods in user-friendliness and the ability to customize complex objects with user-friendly inputs. Our code is available at $\href{https://github.com/zhongzero/Multi-Modal-Prompt}{https://github.com/zhongzero/Multi-Modal-Prompt}$.
URLs: https://github.com/zhongzero/Multi-Modal-Prompt, https://github.com/zhongzero/Multi-Modal-Prompt
Authors: Soumava Paul, Christopher Wewer, Bernt Schiele, Jan Eric Lenssen
Abstract: We aim to tackle sparse-view reconstruction of a 360 3D scene using priors from latent diffusion models (LDM). The sparse-view setting is ill-posed and underconstrained, especially for scenes where the camera rotates 360 degrees around a point, as no visual information is available beyond some frontal views focused on the central object(s) of interest. In this work, we show that pretrained 2D diffusion models can strongly improve the reconstruction of a scene with low-cost fine-tuning. Specifically, we present SparseSplat360 (Sp2360), a method that employs a cascade of in-painting and artifact removal models to fill in missing details and clean novel views. Due to superior training and rendering speeds, we use an explicit scene representation in the form of 3D Gaussians over NeRF-based implicit representations. We propose an iterative update strategy to fuse generated pseudo novel views with existing 3D Gaussians fitted to the initial sparse inputs. As a result, we obtain a multi-view consistent scene representation with details coherent with the observed inputs. Our evaluation on the challenging Mip-NeRF360 dataset shows that our proposed 2D to 3D distillation algorithm considerably improves the performance of a regularized version of 3DGS adapted to a sparse-view setting and outperforms existing sparse-view reconstruction methods in 360 scene reconstruction. Qualitatively, our method generates entire 360 scenes from as few as 9 input views, with a high degree of foreground and background detail.
Authors: Tianyun Yang, Juan Cao, Chang Xu
Abstract: Despite the impressive capabilities of generating images, text-to-image diffusion models are susceptible to producing undesirable outputs such as NSFW content and copyrighted artworks. To address this issue, recent studies have focused on fine-tuning model parameters to erase problematic concepts. However, existing methods exhibit a major flaw in robustness, as fine-tuned models often reproduce the undesirable outputs when faced with cleverly crafted prompts. This reveals a fundamental limitation in the current approaches and may raise risks for the deployment of diffusion models in the open world. To address this gap, we locate the concept-correlated neurons and find that these neurons show high sensitivity to adversarial prompts, thus could be deactivated when erasing and reactivated again under attacks. To improve the robustness, we introduce a new pruning-based strategy for concept erasing. Our method selectively prunes critical parameters associated with the concepts targeted for removal, thereby reducing the sensitivity of concept-related neurons. Our method can be easily integrated with existing concept-erasing techniques, offering a robust improvement against adversarial inputs. Experimental results show a significant enhancement in our model's ability to resist adversarial inputs, achieving nearly a 40% improvement in erasing the NSFW content and a 30% improvement in erasing artwork style.
Authors: Wenqi Ouyang, Yi Dong, Lei Yang, Jianlou Si, Xingang Pan
Abstract: The remarkable generative capabilities of diffusion models have motivated extensive research in both image and video editing. Compared to video editing which faces additional challenges in the time dimension, image editing has witnessed the development of more diverse, high-quality approaches and more capable software like Photoshop. In light of this gap, we introduce a novel and generic solution that extends the applicability of image editing tools to videos by propagating edits from a single frame to the entire video using a pre-trained image-to-video model. Our method, dubbed I2VEdit, adaptively preserves the visual and motion integrity of the source video depending on the extent of the edits, effectively handling global edits, local edits, and moderate shape changes, which existing methods cannot fully achieve. At the core of our method are two main processes: Coarse Motion Extraction to align basic motion patterns with the original video, and Appearance Refinement for precise adjustments using fine-grained attention matching. We also incorporate a skip-interval strategy to mitigate quality degradation from auto-regressive generation across multiple video clips. Experimental results demonstrate our framework's superior performance in fine-grained video editing, proving its capability to produce high-quality, temporally consistent outputs.
Authors: Paramita Kundu Maji, Soubhik Acharya, Priti Paul, Sanjay Chakraborty, Saikat Basu
Abstract: This paper aims to develop a new deep learning-inspired gaming approach for early detection of dementia. This research integrates a robust convolutional neural network (CNN)-based model for early dementia detection using health metrics data as well as facial image data through a cognitive assessment-based gaming application. We have collected 1000 data samples of health metrics dataset from Apollo Diagnostic Center Kolkata that is labeled as either demented or non-demented for the training of MOD-1D-CNN for the game level 1 and another dataset of facial images containing 1800 facial data that are labeled as either demented or non-demented is collected by our research team for the training of MOD-2D-CNN model in-game level 2. In our work, the loss for the proposed MOD-1D-CNN model is 0.2692 and the highest accuracy is 70.50% for identifying the dementia traits using real-life health metrics data. Similarly, the proposed MOD-2D-CNN model loss is 0.1755 and the highest accuracy is obtained here 95.72% for recognizing the dementia status using real-life face-based image data. Therefore, a rule-based weightage method is applied to combine both the proposed methods to achieve the final decision. The MOD-1D-CNN and MOD-2D-CNN models are more lightweight and computationally efficient alternatives because they have a significantly lower number of parameters when compared to the other state-of-the-art models. We have compared their accuracies and parameters with the other state-of-the-art deep learning models.
Authors: Erik Sandstr\"om, Keisuke Tateno, Michael Oechsle, Michael Niemeyer, Luc Van Gool, Martin R. Oswald, Federico Tombari
Abstract: 3D Gaussian Splatting has emerged as a powerful representation of geometry and appearance for RGB-only dense Simultaneous Localization and Mapping (SLAM), as it provides a compact dense map representation while enabling efficient and high-quality map rendering. However, existing methods show significantly worse reconstruction quality than competing methods using other 3D representations, e.g. neural points clouds, since they either do not employ global map and pose optimization or make use of monocular depth. In response, we propose the first RGB-only SLAM system with a dense 3D Gaussian map representation that utilizes all benefits of globally optimized tracking by adapting dynamically to keyframe pose and depth updates by actively deforming the 3D Gaussian map. Moreover, we find that refining the depth updates in inaccurate areas with a monocular depth estimator further improves the accuracy of the 3D reconstruction. Our experiments on the Replica, TUM-RGBD, and ScanNet datasets indicate the effectiveness of globally optimized 3D Gaussians, as the approach achieves superior or on par performance with existing RGB-only SLAM methods methods in tracking, mapping and rendering accuracy while yielding small map sizes and fast runtimes. The source code is available at https://github.com/eriksandstroem/Splat-SLAM.
Authors: Zhaozhi Wang, Yue Liu, Yunfan Liu, Hongtian Yu, Yaowei Wang, Qixiang Ye, Yunjie Tian
Abstract: A fundamental problem in learning robust and expressive visual representations lies in efficiently estimating the spatial relationships of visual semantics throughout the entire image. In this study, we propose vHeat, a novel vision backbone model that simultaneously achieves both high computational efficiency and global receptive field. The essential idea, inspired by the physical principle of heat conduction, is to conceptualize image patches as heat sources and model the calculation of their correlations as the diffusion of thermal energy. This mechanism is incorporated into deep models through the newly proposed module, the Heat Conduction Operator (HCO), which is physically plausible and can be efficiently implemented using DCT and IDCT operations with a complexity of $\mathcal{O}(N^{1.5})$. Extensive experiments demonstrate that vHeat surpasses Vision Transformers (ViTs) across various vision tasks, while also providing higher inference speeds, reduced FLOPs, and lower GPU memory usage for high-resolution images. The code will be released at https://github.com/MzeroMiko/vHeat.
Authors: Francesca Babiloni, Alexandros Lattas, Jiankang Deng, Stefanos Zafeiriou
Abstract: We propose ID-to-3D, a method to generate identity- and text-guided 3D human heads with disentangled expressions, starting from even a single casually captured in-the-wild image of a subject. The foundation of our approach is anchored in compositionality, alongside the use of task-specific 2D diffusion models as priors for optimization. First, we extend a foundational model with a lightweight expression-aware and ID-aware architecture, and create 2D priors for geometry and texture generation, via fine-tuning only 0.2% of its available training parameters. Then, we jointly leverage a neural parametric representation for the expressions of each subject and a multi-stage generation of highly detailed geometry and albedo texture. This combination of strong face identity embeddings and our neural representation enables accurate reconstruction of not only facial features but also accessories and hair and can be meshed to provide render-ready assets for gaming and telepresence. Our results achieve an unprecedented level of identity-consistent and high-quality texture and geometry generation, generalizing to a ``world'' of unseen 3D identities, without relying on large 3D captured datasets of human assets. Explore our 3D results at: https://https://idto3d.github.io.
Authors: Along He, Tao Li, Yanlin Wu, Ke Zou, Huazhu Fu
Abstract: Limited labeled data hinder the application of deep learning in medical domain. In clinical practice, there are sufficient unlabeled data that are not effectively used, and semi-supervised learning (SSL) is a promising way for leveraging these unlabeled data. However, existing SSL methods ignore frequency domain and region-level information and it is important for lesion regions located at low frequencies and with significant scale changes. In this paper, we introduce two consistency regularization strategies for semi-supervised medical image segmentation, including frequency domain consistency (FDC) to assist the feature learning in frequency domain and multi-granularity region similarity consistency (MRSC) to perform multi-scale region-level local context information feature learning. With the help of the proposed FDC and MRSC, we can leverage the powerful feature representation capability of them in an effective and efficient way. We perform comprehensive experiments on two datasets, and the results show that our method achieves large performance gains and exceeds other state-of-the-art methods.
Authors: Yusaku Ando, Miya Nakajima, Takahiro Saitoh, Tsuyoshi Kato
Abstract: In recent years, the deterioration of artificial materials used in structures has become a serious social issue, increasing the importance of inspections. Non-destructive testing is gaining increased demand due to its capability to inspect for defects and deterioration in structures while preserving their functionality. Among these, Laser Ultrasonic Visualization Testing (LUVT) stands out because it allows the visualization of ultrasonic propagation. This makes it visually straightforward to detect defects, thereby enhancing inspection efficiency. With the increasing number of the deterioration structures, challenges such as a shortage of inspectors and increased workload in non-destructive testing have become more apparent. Efforts to address these challenges include exploring automated inspection using machine learning. However, the lack of anomalous data with defects poses a barrier to improving the accuracy of automated inspection through machine learning. Therefore, in this study, we propose a method for automated LUVT inspection using an anomaly detection approach with a diffusion model that can be trained solely on negative examples (defect-free data). We experimentally confirmed that our proposed method improves defect detection and localization compared to general object detection algorithms used previously.
Authors: Qijie Wang, Guandu Liu, Bin Wang
Abstract: Recent advances in vision-language foundational models, such as CLIP, have demonstrated significant strides in zero-shot classification. However, the extensive parameterization of models like CLIP necessitates a resource-intensive fine-tuning process. In response, TIP-Adapter and SuS-X have introduced training-free methods aimed at bolstering the efficacy of downstream tasks. While these approaches incorporate support sets to maintain data distribution consistency between knowledge cache and test sets, they often fall short in terms of generalization on the test set, particularly when faced with test data exhibiting substantial distributional variations. In this work, we present CapS-Adapter, an innovative method that employs a caption-based support set, effectively harnessing both image and caption features to exceed existing state-of-the-art techniques in training-free scenarios. CapS-Adapter adeptly constructs support sets that closely mirror target distributions, utilizing instance-level distribution features extracted from multimodal large models. By leveraging CLIP's single and cross-modal strengths, CapS-Adapter enhances predictive accuracy through the use of multimodal support sets. Our method achieves outstanding zero-shot classification results across 19 benchmark datasets, improving accuracy by 2.19\% over the previous leading method. Our contributions are substantiated through extensive validation on multiple benchmark datasets, demonstrating superior performance and robust generalization capabilities. Our code is made publicly available at https://github.com/WLuLi/CapS-Adapter.
Authors: Runyi Li, Xuanyu Zhang, Zhipei Xu, Yongbing Zhang, Jian Zhang
Abstract: With the advent of personalized generation models, users can more readily create images resembling existing content, heightening the risk of violating portrait rights and intellectual property (IP). Traditional post-hoc detection and source-tracing methods for AI-generated content (AIGC) employ proactive watermark approaches; however, these are less effective against personalized generation models. Moreover, attribution techniques for AIGC rely on passive detection but often struggle to differentiate AIGC from authentic images, presenting a substantial challenge. Integrating these two processes into a cohesive framework not only meets the practical demands for protection and forensics but also improves the effectiveness of attribution tasks. Inspired by this insight, we propose a unified approach for image copyright source-tracing and attribution, introducing an innovative watermarking-attribution method that blends proactive and passive strategies. We embed copyright watermarks into protected images and train a watermark decoder to retrieve copyright information from the outputs of personalized models, using this watermark as an initial step for confirming if an image is AIGC-generated. To pinpoint specific generation techniques, we utilize powerful visual backbone networks for classification. Additionally, we implement an incremental learning strategy to adeptly attribute new personalized models without losing prior knowledge, thereby enhancing the model's adaptability to novel generation methods. We have conducted experiments using various celebrity portrait series sourced online, and the results affirm the efficacy of our method in source-tracing and attribution tasks, as well as its robustness against knowledge forgetting.
Authors: Qizao Wang, Xuelin Qian, Bin Li, Lifeng Chen, Yanwei Fu, Xiangyang Xue
Abstract: Cloth-changing person Re-IDentification (Re-ID) aims at recognizing the same person with clothing changes across non-overlapping cameras. Conventional person Re-ID methods usually bias the model's focus on cloth-related appearance features rather than identity-sensitive features associated with biological traits. Recently, advanced cloth-changing person Re-ID methods either resort to identity-related auxiliary modalities (e.g., sketches, silhouettes, keypoints and 3D shapes) or clothing labels to mitigate the impact of clothes. However, relying on unpractical and inflexible auxiliary modalities or annotations limits their real-world applicability. In this paper, we promote cloth-changing person Re-ID by effectively leveraging abundant semantics present within pedestrian images without the need for any auxiliaries. Specifically, we propose the Content and Salient Semantics Collaboration (CSSC) framework, facilitating cross-parallel semantics interaction and refinement. Our framework is simple yet effective, and the vital design is the Semantics Mining and Refinement (SMR) module. It extracts robust identity features about content and salient semantics, while mitigating interference from clothing appearances effectively. By capitalizing on the mined abundant semantic features, our proposed approach achieves state-of-the-art performance on three cloth-changing benchmarks as well as conventional benchmarks, demonstrating its superiority over advanced competitors.
Authors: Qizao Wang, Xuelin Qian, Bin Li, Yanwei Fu, Xiangyang Xue
Abstract: With the continuous expansion of intelligent surveillance networks, lifelong person re-identification (LReID) has received widespread attention, pursuing the need of self-evolution across different domains. However, existing LReID studies accumulate knowledge with the assumption that people would not change their clothes. In this paper, we propose a more practical task, namely lifelong person re-identification with hybrid clothing states (LReID-Hybrid), which takes a series of cloth-changing and cloth-consistent domains into account during lifelong learning. To tackle the challenges of knowledge granularity mismatch and knowledge presentation mismatch that occurred in LReID-Hybrid, we take advantage of the consistency and generalization of the text space, and propose a novel framework, dubbed $Teata$, to effectively align, transfer and accumulate knowledge in an "image-text-image" closed loop. Concretely, to achieve effective knowledge transfer, we design a Structured Semantic Prompt (SSP) learning to decompose the text prompt into several structured pairs to distill knowledge from the image space with a unified granularity of text description. Then, we introduce a Knowledge Adaptation and Projection strategy (KAP), which tunes text knowledge via a slow-paced learner to adapt to different tasks without catastrophic forgetting. Extensive experiments demonstrate the superiority of our proposed $Teata$ for LReID-Hybrid as well as on conventional LReID benchmarks over advanced methods.
Authors: Dongchen Han, Ziyi Wang, Zhuofan Xia, Yizeng Han, Yifan Pu, Chunjiang Ge, Jun Song, Shiji Song, Bo Zheng, Gao Huang
Abstract: Mamba is an effective state space model with linear computation complexity. It has recently shown impressive efficiency in dealing with high-resolution inputs across various vision tasks. In this paper, we reveal that the powerful Mamba model shares surprising similarities with linear attention Transformer, which typically underperform conventional Transformer in practice. By exploring the similarities and disparities between the effective Mamba and subpar linear attention Transformer, we provide comprehensive analyses to demystify the key factors behind Mamba's success. Specifically, we reformulate the selective state space model and linear attention within a unified formulation, rephrasing Mamba as a variant of linear attention Transformer with six major distinctions: input gate, forget gate, shortcut, no attention normalization, single-head, and modified block design. For each design, we meticulously analyze its pros and cons, and empirically evaluate its impact on model performance in vision tasks. Interestingly, the results highlight the forget gate and block design as the core contributors to Mamba's success, while the other four designs are less crucial. Based on these findings, we propose a Mamba-Like Linear Attention (MLLA) model by incorporating the merits of these two key designs into linear attention. The resulting model outperforms various vision Mamba models in both image classification and high-resolution dense prediction tasks, while enjoying parallelizable computation and fast inference speed. Code is available at https://github.com/LeapLabTHU/MLLA.
Authors: Konstanty Subbotko, Wojciech Jablonski, Piotr Bilinski
Abstract: Neural Architecture Search (NAS) has been widely adopted to design neural networks for various computer vision tasks. One of its most promising subdomains is differentiable NAS (DNAS), where the optimal architecture is found in a differentiable manner. However, gradient-based methods suffer from the discretization error, which can severely damage the process of obtaining the final architecture. In our work, we first study the risk of discretization error and show how it affects an unregularized supernet. Then, we present that penalizing high entropy, a common technique of architecture regularization, can hinder the supernet's performance. Therefore, to robustify the DNAS framework, we introduce a novel single-stage searching protocol, which is not reliant on decoding a continuous architecture. Our results demonstrate that this approach outperforms other DNAS methods by achieving 75.3% in the searching stage on the Cityscapes validation dataset and attains performance 1.1% higher than the optimal network of DCNAS on the non-dense search space comprising short connections. The entire training process takes only 5.5 GPU days due to the weight reuse, and yields a computationally efficient architecture. Additionally, we propose a new dataset split procedure, which substantially improves results and prevents architecture degeneration in DARTS.
Authors: Shuvendu Roy, Elham Dolatabadi, Arash Afkanpour, Ali Etemad
Abstract: For the first time, we explore few-shot tuning of vision foundation models for class-incremental learning. Unlike existing few-shot class incremental learning (FSCIL) methods, which train an encoder on a base session to ensure forward compatibility for future continual learning, foundation models are generally trained on large unlabelled data without such considerations. This renders prior methods from traditional FSCIL incompatible for FSCIL with the foundation model. To this end, we propose Consistency-guided Asynchronous Contrastive Tuning (CoACT), a new approach to continually tune foundation models for new classes in few-shot settings. CoACT comprises three components: (i) asynchronous contrastive tuning, which learns new classes by including LoRA modules in the pre-trained encoder, while enforcing consistency between two asynchronous encoders; (ii) controlled fine-tuning, which facilitates effective tuning of a subset of the foundation model; and (iii) consistency-guided incremental tuning, which enforces additional regularization during later sessions to reduce forgetting of the learned classes. We perform an extensive study on 16 diverse datasets and demonstrate the effectiveness of CoACT, outperforming the best baseline method by 2.47% on average and with up to 12.52% on individual datasets. Additionally, CoACT shows reduced forgetting and robustness in low-shot experiments. As an added bonus, CoACT shows up to 13.5% improvement in standard FSCIL over the current SOTA on benchmark evaluations. We make our code publicly available at https://github.com/ShuvenduRoy/CoACT-FSCIL.
Authors: Shaheer U. Saeed, Shiqi Huang, Jo\~ao Ramalhinho, Iani J. M. B. Gayo, Nina Monta\~na-Brown, Ester Bonmati, Stephen P. Pereira, Brian Davidson, Dean C. Barratt, Matthew J. Clarkson, Yipeng Hu
Abstract: Weakly-supervised segmentation (WSS) methods, reliant on image-level labels indicating object presence, lack explicit correspondence between labels and regions of interest (ROIs), posing a significant challenge. Despite this, WSS methods have attracted attention due to their much lower annotation costs compared to fully-supervised segmentation. Leveraging reinforcement learning (RL) self-play, we propose a novel WSS method that gamifies image segmentation of a ROI. We formulate segmentation as a competition between two agents that compete to select ROI-containing patches until exhaustion of all such patches. The score at each time-step, used to compute the reward for agent training, represents likelihood of object presence within the selection, determined by an object presence detector pre-trained using only image-level binary classification labels of object presence. Additionally, we propose a game termination condition that can be called by either side upon exhaustion of all ROI-containing patches, followed by the selection of a final patch from each. Upon termination, the agent is incentivised if ROI-containing patches are exhausted or disincentivised if an ROI-containing patch is found by the competitor. This competitive setup ensures minimisation of over- or under-segmentation, a common problem with WSS methods. Extensive experimentation across four datasets demonstrates significant performance improvements over recent state-of-the-art methods. Code: https://github.com/s-sd/spurl/tree/main/wss
Authors: Hanwen Liang, Yuyang Yin, Dejia Xu, Hanxue Liang, Zhangyang Wang, Konstantinos N. Plataniotis, Yao Zhao, Yunchao Wei
Abstract: The availability of large-scale multimodal datasets and advancements in diffusion models have significantly accelerated progress in 4D content generation. Most prior approaches rely on multiple image or video diffusion models, utilizing score distillation sampling for optimization or generating pseudo novel views for direct supervision. However, these methods are hindered by slow optimization speeds and multi-view inconsistency issues. Spatial and temporal consistency in 4D geometry has been extensively explored respectively in 3D-aware diffusion models and traditional monocular video diffusion models. Building on this foundation, we propose a strategy to migrate the temporal consistency in video diffusion models to the spatial-temporal consistency required for 4D generation. Specifically, we present a novel framework, \textbf{Diffusion4D}, for efficient and scalable 4D content generation. Leveraging a meticulously curated dynamic 3D dataset, we develop a 4D-aware video diffusion model capable of synthesizing orbital views of dynamic 3D assets. To control the dynamic strength of these assets, we introduce a 3D-to-4D motion magnitude metric as guidance. Additionally, we propose a novel motion magnitude reconstruction loss and 3D-aware classifier-free guidance to refine the learning and generation of motion dynamics. After obtaining orbital views of the 4D asset, we perform explicit 4D construction with Gaussian splatting in a coarse-to-fine manner. The synthesized multi-view consistent 4D image set enables us to swiftly generate high-fidelity and diverse 4D assets within just several minutes. Extensive experiments demonstrate that our method surpasses prior state-of-the-art techniques in terms of generation efficiency and 4D geometry consistency across various prompt modalities.
Authors: Abid Faisal Ayon, S M Maksudul Alam
Abstract: Facial Recognition is a technique, based on machine learning technology that can recognize a human being analyzing his facial profile, and is applied in solving various types of realworld problems nowadays. In this paper, a common real-world problem, finding a missing person has been solved in a secure and effective way with the help of facial recognition technology. Although there exist a few works on solving the problem, the proposed work is unique with respect to its security, design, and feasibility. Impeding intruders in participating in the processes and giving importance to both finders and family members of a missing person are two of the major features of this work. The proofs of the works of our system in finding a missing person have been described in the result section of the paper. The advantages that our system provides over the other existing systems can be realized from the comparisons, described in the result summary section of the paper. The work is capable of providing a worthy solution to find a missing person on the digital platform.
Authors: Mustafa Shukor, Matthieu Cord
Abstract: Large Language Models (LLMs) have demonstrated impressive performance on multimodal tasks, without any multimodal finetuning. They are the building block for Large Multimodal Models, yet, we still lack a proper understanding of their success. In this work, we expose frozen LLMs to image, video, audio and text inputs and analyse their internal representation aiming to understand their generalization beyond textual inputs. Findings. Perceptual tokens (1) are easily distinguishable from textual ones inside LLMs, with significantly different representations, and complete translation to textual tokens does not exist. Yet, (2) both perceptual and textual tokens activate similar LLM weights. Despite being different, (3) perceptual and textual tokens are implicitly aligned inside LLMs, we call this the implicit multimodal alignment (IMA), and argue that this is linked to architectural design, helping LLMs to generalize. This provide more evidence to believe that the generalization of LLMs to multimodal inputs is mainly due to their architecture. Implications. (1) We find a positive correlation between the implicit alignment score and the task performance, suggesting that this could act as a proxy metric for model evaluation and selection. (2) A negative correlation exists regarding hallucinations, revealing that this problem is mainly due to misalignment between the internal perceptual and textual representations. (3) Perceptual tokens change slightly throughout the model, thus, we propose different approaches to skip computations (e.g. in FFN layers), and significantly reduce the inference cost. (4) Due to the slowly changing embeddings across layers, and the high overlap between textual and multimodal activated weights, we compress LLMs by keeping only 1 subnetwork that works well across a wide range of multimodal tasks. Paper code: https://github.com/mshukor/ima-lmms.
Authors: Tong Shi, Xuri Ge, Joemon M. Jose, Nicolas Pugeault, Paul Henderson
Abstract: Capturing complex temporal relationships between video and audio modalities is vital for Audio-Visual Emotion Recognition (AVER). However, existing methods lack attention to local details, such as facial state changes between video frames, which can reduce the discriminability of features and thus lower recognition accuracy. In this paper, we propose a Detail-Enhanced Intra- and Inter-modal Interaction network(DE-III) for AVER, incorporating several novel aspects. We introduce optical flow information to enrich video representations with texture details that better capture facial state changes. A fusion module integrates the optical flow estimation with the corresponding video frames to enhance the representation of facial texture variations. We also design attentive intra- and inter-modal feature enhancement modules to further improve the richness and discriminability of video and audio representations. A detailed quantitative evaluation shows that our proposed model outperforms all existing methods on three benchmark datasets for both concrete and continuous emotion recognition. To encourage further research and ensure replicability, we will release our full code upon acceptance.
Authors: Lijun Yu
Abstract: Advancements in language foundation models have primarily fueled the recent surge in artificial intelligence. In contrast, generative learning of non-textual modalities, especially videos, significantly trails behind language modeling. This thesis chronicles our endeavor to build multi-task models for generating videos and other modalities under diverse conditions, as well as for understanding and compression applications. Given the high dimensionality of visual data, we pursue concise and accurate latent representations. Our video-native spatial-temporal tokenizers preserve high fidelity. We unveil a novel approach to mapping bidirectionally between visual observation and interpretable lexical terms. Furthermore, our scalable visual token representation proves beneficial across generation, compression, and understanding tasks. This achievement marks the first instances of language models surpassing diffusion models in visual synthesis and a video tokenizer outperforming industry-standard codecs. Within these multi-modal latent spaces, we study the design of multi-task generative models. Our masked multi-task transformer excels at the quality, efficiency, and flexibility of video generation. We enable a frozen language model, trained solely on text, to generate visual content. Finally, we build a scalable generative multi-modal transformer trained from scratch, enabling the generation of videos containing high-fidelity motion with the corresponding audio given diverse conditions. Throughout the course, we have shown the effectiveness of integrating multiple tasks, crafting high-fidelity latent representation, and generating multiple modalities. This work suggests intriguing potential for future exploration in generating non-textual data and enabling real-time, interactive experiences across various media forms.
Authors: Hastings Greer, Lin Tian, Francois-Xavier Vialard, Roland Kwitt, Raul San Jose Estepar, Marc Niethammer
Abstract: Image registration estimates spatial correspondences between a pair of images. These estimates are typically obtained via numerical optimization or regression by a deep network. A desirable property of such estimators is that a correspondence estimate (e.g., the true oracle correspondence) for an image pair is maintained under deformations of the input images. Formally, the estimator should be equivariant to a desired class of image transformations. In this work, we present careful analyses of the desired equivariance properties in the context of multi-step deep registration networks. Based on these analyses we 1) introduce the notions of $[U,U]$ equivariance (network equivariance to the same deformations of the input images) and $[W,U]$ equivariance (where input images can undergo different deformations); we 2) show that in a suitable multi-step registration setup it is sufficient for overall $[W,U]$ equivariance if the first step has $[W,U]$ equivariance and all others have $[U,U]$ equivariance; we 3) show that common displacement-predicting networks only exhibit $[U,U]$ equivariance to translations instead of the more powerful $[W,U]$ equivariance; and we 4) show how to achieve multi-step $[W,U]$ equivariance via a coordinate-attention mechanism combined with displacement-predicting refinement layers (CARL). Overall, our approach obtains excellent practical registration performance on several 3D medical image registration tasks and outperforms existing unsupervised approaches for the challenging problem of abdomen registration.
Authors: Md Mostafijur Rahman, Mustafa Munir, Debesh Jha, Ulas Bagci, Radu Marculescu
Abstract: The Segment Anything Model (SAM), originally designed for general-purpose segmentation tasks, has been used recently for polyp segmentation. Nonetheless, fine-tuning SAM with data from new imaging centers or clinics poses significant challenges. This is because this necessitates the creation of an expensive and time-intensive annotated dataset, along with the potential for variability in user prompts during inference. To address these issues, we propose a robust fine-tuning technique, PP-SAM, that allows SAM to adapt to the polyp segmentation task with limited images. To this end, we utilize variable perturbed bounding box prompts (BBP) to enrich the learning context and enhance the model's robustness to BBP perturbations during inference. Rigorous experiments on polyp segmentation benchmarks reveal that our variable BBP perturbation significantly improves model resilience. Notably, on Kvasir, 1-shot fine-tuning boosts the DICE score by 20% and 37% with 50 and 100-pixel BBP perturbations during inference, respectively. Moreover, our experiments show that 1-shot, 5-shot, and 10-shot PP-SAM with 50-pixel perturbations during inference outperform a recent state-of-the-art (SOTA) polyp segmentation method by 26%, 7%, and 5% DICE scores, respectively. Our results motivate the broader applicability of our PP-SAM for other medical imaging tasks with limited samples. Our implementation is available at https://github.com/SLDGroup/PP-SAM.
Authors: Loc Hoang Tran
Abstract: Face recognition is a very important topic in data science and biometric security research areas. It has multiple applications in military, finance, and retail, to name a few. In this paper, the novel hypergraph Laplacian Eigenmaps will be proposed and combine with the k nearest-neighbor method and/or with the kernel ridge regression method to solve the face recognition problem. Experimental results illustrate that the accuracy of the combination of the novel hypergraph Laplacian Eigenmaps and one specific classification system is similar to the accuracy of the combination of the old symmetric normalized hypergraph Laplacian Eigenmaps method and one specific classification system.
Authors: Cristina N. Vasconcelos, Abdullah Rashwan Austin Waters, Trevor Walker, Keyang Xu, Jimmy Yan, Rui Qian, Shixin Luo, Zarana Parekh, Andrew Bunner, Hongliang Fei, Roopal Garg, Mandy Guo, Ivana Kajic, Yeqing Li, Henna Nandwani, Jordi Pont-Tuset, Yasumasa Onoe, Sarah Rosston, Su Wang, Wenlei Zhou, Kevin Swersky, David J. Fleet, Jason M. Baldridge, Oliver Wang
Abstract: We address the long-standing problem of how to learn effective pixel-based image diffusion models at scale, introducing a remarkably simple greedy growing method for stable training of large-scale, high-resolution models. without the needs for cascaded super-resolution components. The key insight stems from careful pre-training of core components, namely, those responsible for text-to-image alignment {\it vs.} high-resolution rendering. We first demonstrate the benefits of scaling a {\it Shallow UNet}, with no down(up)-sampling enc(dec)oder. Scaling its deep core layers is shown to improve alignment, object structure, and composition. Building on this core model, we propose a greedy algorithm that grows the architecture into high-resolution end-to-end models, while preserving the integrity of the pre-trained representation, stabilizing training, and reducing the need for large high-resolution datasets. This enables a single stage model capable of generating high-resolution images without the need of a super-resolution cascade. Our key results rely on public datasets and show that we are able to train non-cascaded models up to 8B parameters with no further regularization schemes. Vermeer, our full pipeline model trained with internal datasets to produce 1024x1024 images, without cascades, is preferred by 44.0% vs. 21.4% human evaluators over SDXL.
Authors: Shiming Ge, Weijia Guo, Chenyu Li, Junzheng Zhang, Yong Li, Dan Zeng
Abstract: Masked face recognition is important for social good but challenged by diverse occlusions that cause insufficient or inaccurate representations. In this work, we propose a unified deep network to learn generative-to-discriminative representations for facilitating masked face recognition. To this end, we split the network into three modules and learn them on synthetic masked faces in a greedy module-wise pretraining manner. First, we leverage a generative encoder pretrained for face inpainting and finetune it to represent masked faces into category-aware descriptors. Attribute to the generative encoder's ability in recovering context information, the resulting descriptors can provide occlusion-robust representations for masked faces, mitigating the effect of diverse masks. Then, we incorporate a multi-layer convolutional network as a discriminative reformer and learn it to convert the category-aware descriptors into identity-aware vectors, where the learning is effectively supervised by distilling relation knowledge from off-the-shelf face recognition model. In this way, the discriminative reformer together with the generative encoder serves as the pretrained backbone, providing general and discriminative representations towards masked faces. Finally, we cascade one fully-connected layer following by one softmax layer into a feature classifier and finetune it to identify the reformed identity-aware vectors. Extensive experiments on synthetic and realistic datasets demonstrate the effectiveness of our approach in recognizing masked faces.
Authors: YuXiao Lee, Xiaofeng Cao
Abstract: The remarkable achievements of Large Language Models (LLMs) have captivated the attention of both academia and industry, transcending their initial role in dialogue generation. The utilization of LLMs as intermediary agents in various tasks has yielded promising results, sparking a wave of innovation in artificial intelligence. Building on these breakthroughs, we introduce a novel approach that integrates the agent paradigm into the Out-of-distribution (OOD) detection task, aiming to enhance its robustness and adaptability. Our proposed method, Concept Matching with Agent (CMA), employs neutral prompts as agents to augment the CLIP-based OOD detection process. These agents function as dynamic observers and communication hubs, interacting with both In-distribution (ID) labels and data inputs to form vector triangle relationships. This triangular framework offers a more nuanced approach than the traditional binary relationship, allowing for better separation and identification of ID and OOD inputs. Our extensive experimental results showcase the superior performance of CMA over both zero-shot and training-required methods in a diverse array of real-world scenarios.
Authors: Yongsheng Yu, Ziyun Zeng, Hang Hua, Jianlong Fu, Jiebo Luo
Abstract: Diffusion models equipped with language models demonstrate excellent controllability in image generation tasks, allowing image processing to adhere to human instructions. However, the lack of diverse instruction-following data hampers the development of models that effectively recognize and execute user-customized instructions, particularly in low-level tasks. Moreover, the stochastic nature of the diffusion process leads to deficiencies in image generation or editing tasks that require the detailed preservation of the generated images. To address these limitations, we propose PromptFix, a comprehensive framework that enables diffusion models to follow human instructions to perform a wide variety of image-processing tasks. First, we construct a large-scale instruction-following dataset that covers comprehensive image-processing tasks, including low-level tasks, image editing, and object creation. Next, we propose a high-frequency guidance sampling method to explicitly control the denoising process and preserve high-frequency details in unprocessed areas. Finally, we design an auxiliary prompting adapter, utilizing Vision-Language Models (VLMs) to enhance text prompts and improve the model's task generalization. Experimental results show that PromptFix outperforms previous methods in various image-processing tasks. Our proposed model also achieves comparable inference efficiency with these baseline models and exhibits superior zero-shot capabilities in blind restoration and combination tasks. The dataset and code will be aviliable at https://github.com/yeates/PromptFix.
Authors: Hanyu Chen, Bailey Miller, Ioannis Gkioulekas
Abstract: We introduce a technique for the reconstruction of high-fidelity surfaces from multi-view images. Our technique uses a new point-based representation, the dipole sum, which generalizes the winding number to allow for interpolation of arbitrary per-point attributes in point clouds with noisy or outlier points. Using dipole sums allows us to represent implicit geometry and radiance fields as per-point attributes of a point cloud, which we initialize directly from structure from motion. We additionally derive Barnes-Hut fast summation schemes for accelerated forward and reverse-mode dipole sum queries. These queries facilitate the use of ray tracing to efficiently and differentiably render images with our point-based representations, and thus update their point attributes to optimize scene geometry and appearance. We evaluate this inverse rendering framework against state-of-the-art alternatives, based on ray tracing of neural representations or rasterization of Gaussian point-based representations. Our technique significantly improves reconstruction quality at equal runtimes, while also supporting more general rendering techniques such as shadow rays for direct illumination. In the supplement, we provide interactive visualizations of our results.
Authors: Liwen Hu, Lei Ma, Yijia Guo, Tiejun Huang
Abstract: Spike cameras, with their exceptional temporal resolution, are revolutionizing high-speed visual applications. Large-scale synthetic datasets have significantly accelerated the development of these cameras, particularly in reconstruction and optical flow. However, current synthetic datasets for spike cameras lack sophistication. Addressing this gap, we introduce SCSim, a novel and more realistic spike camera simulator with a comprehensive noise model. SCSim is adept at autonomously generating driving scenarios and synthesizing corresponding spike streams. To enhance the fidelity of these streams, we've developed a comprehensive noise model tailored to the unique circuitry of spike cameras. Our evaluations demonstrate that SCSim outperforms existing simulation methods in generating authentic spike streams. Crucially, SCSim simplifies the creation of datasets, thereby greatly advancing spike-based visual tasks like reconstruction. Our project refers to https://github.com/Acnext/SCSim.
Authors: Mostofa Rafid Uddin, Min Xu
Abstract: Unsupervised disentanglement of content and transformation has recently drawn much research, given their efficacy in solving downstream unsupervised tasks like clustering, alignment, and shape analysis. This problem is particularly important for analyzing shape-focused real-world scientific image datasets, given their significant relevance to downstream tasks. The existing works address the problem by explicitly parameterizing the transformation factors, significantly reducing their expressiveness. Moreover, they are not applicable in cases where transformations can not be readily parametrized. An alternative to such explicit approaches is self-supervised methods with data augmentation, which implicitly disentangles transformations and content. We demonstrate that the existing self-supervised methods with data augmentation result in the poor disentanglement of content and transformations in real-world scenarios. Therefore, we developed a novel self-supervised method, DualContrast, specifically for unsupervised disentanglement of content and transformations in shape-focused image datasets. Our extensive experiments showcase the superiority of DualContrast over existing self-supervised and explicit parameterization approaches. We leveraged DualContrast to disentangle protein identities and protein conformations in cellular 3D protein images. Moreover, we also disentangled transformations in MNIST, viewpoint in the Linemod Object dataset, and human movement deformation in the Starmen dataset as transformations using DualContrast.
Authors: Xinyu Zhang, Mengxue Kang, Fei Wei, Shuang Xu, Yuhe Liu, Lin Ma
Abstract: As the field of image generation rapidly advances, traditional diffusion models and those integrated with multimodal large language models (LLMs) still encounter limitations in interpreting complex prompts and preserving image consistency pre and post-editing. To tackle these challenges, we present an innovative image editing framework that employs the robust Chain-of-Thought (CoT) reasoning and localizing capabilities of multimodal LLMs to aid diffusion models in generating more refined images. We first meticulously design a CoT process comprising instruction decomposition, region localization, and detailed description. Subsequently, we fine-tune the LISA model, a lightweight multimodal LLM, using the CoT process of Multimodal LLMs and the mask of the edited image. By providing the diffusion models with knowledge of the generated prompt and image mask, our models generate images with a superior understanding of instructions. Through extensive experiments, our model has demonstrated superior performance in image generation, surpassing existing state-of-the-art models. Notably, our model exhibits an enhanced ability to understand complex prompts and generate corresponding images, while maintaining high fidelity and consistency in images before and after generation.
Authors: Leo Hoshikawa, Marcos V. Conde, Takeshi Ohashi, Atsushi Irie
Abstract: Implicit Neural Representations (INRs) and Neural Fields are a novel paradigm for signal representation, from images and audio to 3D scenes and videos. The fundamental idea is to represent a signal as a continuous and differentiable neural network. This idea offers unprecedented benefits such as continuous resolution and memory efficiency, enabling new compression techniques. However, representing data as neural networks poses new challenges. For instance, given a 2D image as a neural network, how can we further compress such a neural image?. In this work, we present a novel analysis on compressing neural fields, with the focus on images. We also introduce Adaptive Neural Images (ANI), an efficient neural representation that enables adaptation to different inference or transmission requirements. Our proposed method allows to reduce the bits-per-pixel (bpp) of the neural image by 4x, without losing sensitive details or harming fidelity. We achieve this thanks to our successful implementation of 4-bit neural representations. Our work offers a new framework for developing compressed neural fields.
Authors: Trung Dang, Huy Hoang Nguyen, Aleksei Tiulpin
Abstract: One of the primary challenges in brain tumor segmentation arises from the uncertainty of voxels close to tumor boundaries. However, the conventional process of generating ground truth segmentation masks fails to treat such uncertainties properly. Those ``hard labels'' with 0s and 1s conceptually influenced the majority of prior studies on brain image segmentation. As a result, tumor segmentation is often solved through voxel classification. In this work, we instead view this problem as a voxel-level regression, where the ground truth represents a certainty mapping from any pixel based on the distance to tumor border. We propose a novel ground truth label transformation, which is based on a signed geodesic transform, to capture the uncertainty in brain tumors' vicinity, while maintaining a margin between positive and negative samples. We combine this idea with a Focal-like regression L1-loss that enables effective regression learning in high-dimensional output space by appropriately weighting voxels according to their difficulty. We thoroughly conduct an experimental evaluation to validate the components of our proposed method, compare it to a diverse array of state-of-the-art segmentation models, and show that it is architecture-agnostic. The code of our method is made publicly available (\url{https://github.com/Oulu-IMEDS/SiNGR/}).
Authors: Trung Dang, Huy Hoang Nguyen, Aleksei Tiulpin
Abstract: Accurate retinal vessel segmentation is a crucial step in the quantitative assessment of retinal vasculature, which is needed for the early detection of retinal diseases and other conditions. Numerous studies have been conducted to tackle the problem of segmenting vessels automatically using a pixel-wise classification approach. The common practice of creating ground truth labels is to categorize pixels as foreground and background. This approach is, however, biased, and it ignores the uncertainty of a human annotator when it comes to annotating e.g. thin vessels. In this work, we propose a simple and effective method that casts the retinal image segmentation task as an image-level regression. For this purpose, we first introduce a novel Segmentation Annotation Uncertainty-Aware (SAUNA) transform, which adds pixel uncertainty to the ground truth using the pixel's closeness to the annotation boundary and vessel thickness. To train our model with soft labels, we generalize the earlier proposed Jaccard metric loss to arbitrary hypercubes, which is a second contribution of this work. The proposed SAUNA transform and the new theoretical results allow us to directly train a standard U-Net-like architecture at the image level, outperforming all recently published methods. We conduct thorough experiments and compare our method to a diverse set of baselines across 5 retinal image datasets. Our implementation is available at \url{https://github.com/Oulu-IMEDS/SAUNA}.
Authors: Shoma Iwai, Tomo Miyazaki, Shinichiro Omachi
Abstract: In recent years, neural network-driven image compression (NIC) has gained significant attention. Some works adopt deep generative models such as GANs and diffusion models to enhance perceptual quality (realism). A critical obstacle of these generative NIC methods is that each model is optimized for a single bit rate. Consequently, multiple models are required to compress images to different bit rates, which is impractical for real-world applications. To tackle this issue, we propose a variable-rate generative NIC model. Specifically, we explore several discriminator designs tailored for the variable-rate approach and introduce a novel adversarial loss. Moreover, by incorporating the newly proposed multi-realism technique, our method allows the users to adjust the bit rate, distortion, and realism with a single model, achieving ultra-controllability. Unlike existing variable-rate generative NIC models, our method matches or surpasses the performance of state-of-the-art single-rate generative NIC models while covering a wide range of bit rates using just one model. Code will be available at https://github.com/iwa-shi/CRDR
Authors: Yikai Wang, Xinzhou Wang, Zilong Chen, Zhengyi Wang, Fuchun Sun, Jun Zhu
Abstract: Video generative models are receiving particular attention given their ability to generate realistic and imaginative frames. Besides, these models are also observed to exhibit strong 3D consistency, significantly enhancing their potential to act as world simulators. In this work, we present Vidu4D, a novel reconstruction model that excels in accurately reconstructing 4D (i.e., sequential 3D) representations from single generated videos, addressing challenges associated with non-rigidity and frame distortion. This capability is pivotal for creating high-fidelity virtual contents that maintain both spatial and temporal coherence. At the core of Vidu4D is our proposed Dynamic Gaussian Surfels (DGS) technique. DGS optimizes time-varying warping functions to transform Gaussian surfels (surface elements) from a static state to a dynamically warped state. This transformation enables a precise depiction of motion and deformation over time. To preserve the structural integrity of surface-aligned Gaussian surfels, we design the warped-state geometric regularization based on continuous warping fields for estimating normals. Additionally, we learn refinements on rotation and scaling parameters of Gaussian surfels, which greatly alleviates texture flickering during the warping process and enhances the capture of fine-grained appearance details. Vidu4D also contains a novel initialization state that provides a proper start for the warping fields in DGS. Equipping Vidu4D with an existing video generative model, the overall framework demonstrates high-fidelity text-to-4D generation in both appearance and geometry.
Authors: Gihyun Kwon, Jangho Park, Jong Chul Ye
Abstract: While text-to-image models have achieved impressive capabilities in image generation and editing, their application across various modalities often necessitates training separate models. Inspired by existing method of single image editing with self attention injection and video editing with shared attention, we propose a novel unified editing framework that combines the strengths of both approaches by utilizing only a basic 2D image text-to-image (T2I) diffusion model. Specifically, we design a sampling method that facilitates editing consecutive images while maintaining semantic consistency utilizing shared self-attention features during both reference and consecutive image sampling processes. Experimental results confirm that our method enables editing across diverse modalities including 3D scenes, videos, and panorama images.
Authors: Zipeng Wang, Dan Xu
Abstract: Neural Radiance Fields (NeRFs) have demonstrated remarkable proficiency in synthesizing photorealistic images of large-scale scenes. However, they are often plagued by a loss of fine details and long rendering durations. 3D Gaussian Splatting has recently been introduced as a potent alternative, achieving both high-fidelity visual results and accelerated rendering performance. Nonetheless, scaling 3D Gaussian Splatting is fraught with challenges. Specifically, large-scale scenes grapples with the integration of objects across multiple scales and disparate viewpoints, which often leads to compromised efficacy as the Gaussians need to balance between detail levels. Furthermore, the generation of initialization points via COLMAP from large-scale dataset is both computationally demanding and prone to incomplete reconstructions. To address these challenges, we present Pyramidal 3D Gaussian Splatting (PyGS) with NeRF Initialization. Our approach represent the scene with a hierarchical assembly of Gaussians arranged in a pyramidal fashion. The top level of the pyramid is composed of a few large Gaussians, while each subsequent layer accommodates a denser collection of smaller Gaussians. We effectively initialize these pyramidal Gaussians through sampling a rapidly trained grid-based NeRF at various frequencies. We group these pyramidal Gaussians into clusters and use a compact weighting network to dynamically determine the influence of each pyramid level of each cluster considering camera viewpoint during rendering. Our method achieves a significant performance leap across multiple large-scale datasets and attains a rendering time that is over 400 times faster than current state-of-the-art approaches.
Authors: Yinda Chen, Haoyuan Shi, Xiaoyu Liu, Te Shi, Ruobing Zhang, Dong Liu, Zhiwei Xiong, Feng Wu
Abstract: Autoregressive next-token prediction is a standard pretraining method for large-scale language models, but its application to vision tasks is hindered by the non-sequential nature of image data, leading to cumulative errors. Most vision models employ masked autoencoder (MAE) based pretraining, which faces scalability issues. To address these challenges, we introduce \textbf{TokenUnify}, a novel pretraining method that integrates random token prediction, next-token prediction, and next-all token prediction. We provide theoretical evidence demonstrating that TokenUnify mitigates cumulative errors in visual autoregression. Cooperated with TokenUnify, we have assembled a large-scale electron microscopy (EM) image dataset with ultra-high resolution, ideal for creating spatially correlated long sequences. This dataset includes over 120 million annotated voxels, making it the largest neuron segmentation dataset to date and providing a unified benchmark for experimental validation. Leveraging the Mamba network inherently suited for long-sequence modeling on this dataset, TokenUnify not only reduces the computational complexity but also leads to a significant 45\% improvement in segmentation performance on downstream EM neuron segmentation tasks compared to existing methods. Furthermore, TokenUnify demonstrates superior scalability over MAE and traditional autoregressive methods, effectively bridging the gap between pretraining strategies for language and vision models. Code is available at \url{https://github.com/ydchen0806/TokenUnify}.
Authors: Zhihang Song, Lihui Peng, Jianming Hu, Danya Yao, Yi Zhang
Abstract: Multi-modal object detection in autonomous driving has achieved great breakthroughs due to the usage of fusing complementary information from different sensors. The calibration in fusion between sensors such as LiDAR and camera is always supposed to be precise in previous work. However, in reality, calibration matrices are fixed when the vehicles leave the factory, but vibration, bumps, and data lags may cause calibration bias. As the research on the calibration influence on fusion detection performance is relatively few, flexible calibration dependency multi-sensor detection method has always been attractive. In this paper, we conducted experiments on SOTA detection method EPNet++ and proved slight bias on calibration can reduce the performance seriously. We also proposed a re-calibration model based on semantic segmentation which can be combined with a detection algorithm to improve the performance and robustness of multi-modal calibration bias.
Authors: Zhoujie Fu, Jiacheng Wei, Wenhao Shen, Chaoyue Song, Xiaofeng Yang, Fayao Liu, Xulei Yang, Guosheng Lin
Abstract: In this work, we introduce a novel approach for creating controllable dynamics in 3D-generated Gaussians using casually captured reference videos. Our method transfers the motion of objects from reference videos to a variety of generated 3D Gaussians across different categories, ensuring precise and customizable motion transfer. We achieve this by employing blend skinning-based non-parametric shape reconstruction to extract the shape and motion of reference objects. This process involves segmenting the reference objects into motion-related parts based on skinning weights and establishing shape correspondences with generated target shapes. To address shape and temporal inconsistencies prevalent in existing methods, we integrate physical simulation, driving the target shapes with matched motion. This integration is optimized through a displacement loss to ensure reliable and genuine dynamics. Our approach supports diverse reference inputs, including humans, quadrupeds, and articulated objects, and can generate dynamics of arbitrary length, providing enhanced fidelity and applicability. Unlike methods heavily reliant on diffusion video generation models, our technique offers specific and high-quality motion transfer, maintaining both shape integrity and temporal consistency.
Authors: Jingguo Liu, Yijun Xu, Shigang Li, Jianfeng Li
Abstract: Disconnectivity and distortion are the two problems which must be coped with when processing 360 degrees equirectangular images. In this paper, we propose a method of estimating the depth of monocular panoramic image with a teacher-student model fusing equirectangular and spherical representations. In contrast with the existing methods fusing an equirectangular representation with a cube map representation or tangent representation, a spherical representation is a better choice because a sampling on a sphere is more uniform and can also cope with distortion more effectively. In this processing, a novel spherical convolution kernel computing with sampling points on a sphere is developed to extract features from the spherical representation, and then, a Segmentation Feature Fusion(SFF) methodology is utilized to combine the features with ones extracted from the equirectangular representation. In contrast with the existing methods using a teacher-student model to obtain a lighter model of depth estimation, we use a teacher-student model to learn the latent features of depth images. This results in a trained model which estimates the depth map of an equirectangular image using not only the feature maps extracted from an input equirectangular image but also the distilled knowledge learnt from the ground truth of depth map of a training set. In experiments, the proposed method is tested on several well-known 360 monocular depth estimation benchmark datasets, and outperforms the existing methods for the most evaluation indexes.
Authors: Yunqi Zhang, Songda Li, Chunyuan Deng, Luyi Wang, Hui Zhao
Abstract: Gender bias in vision-language models (VLMs) can reinforce harmful stereotypes and discrimination. In this paper, we focus on mitigating gender bias towards vision-language tasks. We identify object hallucination as the essence of gender bias in VLMs. Existing VLMs tend to focus on salient or familiar attributes in images but ignore contextualized nuances. Moreover, most VLMs rely on the co-occurrence between specific objects and gender attributes to infer the ignored features, ultimately resulting in gender bias. We propose GAMA, a task-agnostic generation framework to mitigate gender bias. GAMA consists of two stages: narrative generation and answer inference. During narrative generation, GAMA yields all-sided but gender-obfuscated narratives, which prevents premature concentration on localized image features, especially gender attributes. During answer inference, GAMA integrates the image, generated narrative, and a task-specific question prompt to infer answers for different vision-language tasks. This approach allows the model to rethink gender attributes and answers. We conduct extensive experiments on GAMA, demonstrating its debiasing and generalization ability.
Authors: Tianhang Wang, Fan Lu, Zehan Zheng, Guang Chen, Changjun Jiang
Abstract: Collaborative perception is dedicated to tackling the constraints of single-agent perception, such as occlusions, based on the multiple agents' multi-view sensor inputs. However, most existing works assume an ideal condition that all agents' multi-view cameras are continuously available. In reality, cameras may be highly noisy, obscured or even failed during the collaboration. In this work, we introduce a new robust camera-insensitivity problem: how to overcome the issues caused by the failed camera perspectives, while stabilizing high collaborative performance with low calibration cost? To address above problems, we propose RCDN, a Robust Camera-insensitivity collaborative perception with a novel Dynamic feature-based 3D Neural modeling mechanism. The key intuition of RCDN is to construct collaborative neural rendering field representations to recover failed perceptual messages sent by multiple agents. To better model collaborative neural rendering field, RCDN first establishes a geometry BEV feature based time-invariant static field with other agents via fast hash grid modeling. Based on the static background field, the proposed time-varying dynamic field can model corresponding motion vectors for foregrounds with appropriate positions. To validate RCDN, we create OPV2V-N, a new large-scale dataset with manual labelling under different camera failed scenarios. Extensive experiments conducted on OPV2V-N show that RCDN can be ported to other baselines and improve their robustness in extreme camera-insensitivity settings. Our code and datasets will be available soon.
Authors: Ziying Song, Feiyang Jia, Hongyu Pan, Yadan Luo, Caiyan Jia, Guoxin Zhang, Lin Liu, Yang Ji, Lei Yang, Li Wang
Abstract: In the field of 3D object detection tasks, fusing heterogeneous features from LiDAR and camera sensors into a unified Bird's Eye View (BEV) representation is a widely adopted paradigm. However, existing methods are often compromised by imprecise sensor calibration, resulting in feature misalignment in LiDAR-camera BEV fusion. Moreover, such inaccuracies result in errors in depth estimation for the camera branch, ultimately causing misalignment between LiDAR and camera BEV features. In this work, we propose a novel ContrastAlign approach that utilizes contrastive learning to enhance the alignment of heterogeneous modalities, thereby improving the robustness of the fusion process. Specifically, our approach includes the L-Instance module, which directly outputs LiDAR instance features within LiDAR BEV features. Then, we introduce the C-Instance module, which predicts camera instance features through RoI (Region of Interest) pooling on the camera BEV features. We propose the InstanceFusion module, which utilizes contrastive learning to generate similar instance features across heterogeneous modalities. We then use graph matching to calculate the similarity between the neighboring camera instance features and the similarity instance features to complete the alignment of instance features. Our method achieves state-of-the-art performance, with an mAP of 70.3%, surpassing BEVFusion by 1.8% on the nuScenes validation set. Importantly, our method outperforms BEVFusion by 7.3% under conditions with misalignment noise.
Authors: Xingqun Qi, Hengyuan Zhang, Yatian Wang, Jiahao Pan, Chen Liu, Peng Li, Xiaowei Chi, Mengfei Li, Qixun Zhang, Wei Xue, Shanghang Zhang, Wenhan Luo, Qifeng Liu, Yike Guo
Abstract: Deriving co-speech 3D gestures has seen tremendous progress in virtual avatar animation. Yet, the existing methods often produce stiff and unreasonable gestures with unseen human speech inputs due to the limited 3D speech-gesture data. In this paper, we propose CoCoGesture, a novel framework enabling vivid and diverse gesture synthesis from unseen human speech prompts. Our key insight is built upon the custom-designed pretrain-fintune training paradigm. At the pretraining stage, we aim to formulate a large generalizable gesture diffusion model by learning the abundant postures manifold. Therefore, to alleviate the scarcity of 3D data, we first construct a large-scale co-speech 3D gesture dataset containing more than 40M meshed posture instances across 4.3K speakers, dubbed GES-X. Then, we scale up the large unconditional diffusion model to 1B parameters and pre-train it to be our gesture experts. At the finetune stage, we present the audio ControlNet that incorporates the human voice as condition prompts to guide the gesture generation. Here, we construct the audio ControlNet through a trainable copy of our pre-trained diffusion model. Moreover, we design a novel Mixture-of-Gesture-Experts (MoGE) block to adaptively fuse the audio embedding from the human speech and the gesture features from the pre-trained gesture experts with a routing mechanism. Such an effective manner ensures audio embedding is temporal coordinated with motion features while preserving the vivid and diverse gesture generation. Extensive experiments demonstrate that our proposed CoCoGesture outperforms the state-of-the-art methods on the zero-shot speech-to-gesture generation. The dataset will be publicly available at: https://mattie-e.github.io/GES-X/
Authors: Jiaqi Tang, Hao Lu, Ruizheng Wu, Xiaogang Xu, Ke Ma, Cheng Fang, Bin Guo, Jiangbo Lu, Qifeng Chen, Ying-Cong Chen
Abstract: Video Anomaly Detection (VAD) systems can autonomously monitor and identify disturbances, reducing the need for manual labor and associated costs. However, current VAD systems are often limited by their superficial semantic understanding of scenes and minimal user interaction. Additionally, the prevalent data scarcity in existing datasets restricts their applicability in open-world scenarios. In this paper, we introduce Hawk, a novel framework that leverages interactive large Visual Language Models (VLM) to interpret video anomalies precisely. Recognizing the difference in motion information between abnormal and normal videos, Hawk explicitly integrates motion modality to enhance anomaly identification. To reinforce motion attention, we construct an auxiliary consistency loss within the motion and video space, guiding the video branch to focus on the motion modality. Moreover, to improve the interpretation of motion-to-language, we establish a clear supervisory relationship between motion and its linguistic representation. Furthermore, we have annotated over 8,000 anomaly videos with language descriptions, enabling effective training across diverse open-world scenarios, and also created 8,000 question-answering pairs for users' open-world questions. The final results demonstrate that Hawk achieves SOTA performance, surpassing existing baselines in both video description generation and question-answering. Our codes/dataset/demo will be released at https://github.com/jqtangust/hawk.
Authors: Haohan Weng, Yikai Wang, Tong Zhang, C. L. Philip Chen, Jun Zhu
Abstract: Generating compact and sharply detailed 3D meshes poses a significant challenge for current 3D generative models. Different from extracting dense meshes from neural representation, some recent works try to model the native mesh distribution (i.e., a set of triangles), which generates more compact results as humans crafted. However, due to the complexity and variety of mesh topology, these methods are typically limited to small datasets with specific categories and are hard to extend. In this paper, we introduce a generic and scalable mesh generation framework PivotMesh, which makes an initial attempt to extend the native mesh generation to large-scale datasets. We employ a transformer-based auto-encoder to encode meshes into discrete tokens and decode them from face level to vertex level hierarchically. Subsequently, to model the complex typology, we first learn to generate pivot vertices as coarse mesh representation and then generate the complete mesh tokens with the same auto-regressive Transformer. This reduces the difficulty compared with directly modeling the mesh distribution and further improves the model controllability. PivotMesh demonstrates its versatility by effectively learning from both small datasets like Shapenet, and large-scale datasets like Objaverse and Objaverse-xl. Extensive experiments indicate that PivotMesh can generate compact and sharp 3D meshes across various categories, highlighting its great potential for native mesh modeling.
Authors: Liang Shi, Jie Zhang, Shiguang Shan
Abstract: Text-to-image diffusion models, such as Stable Diffusion, generate highly realistic images from text descriptions. However, the generation of certain content at such high quality raises concerns. A prominent issue is the accurate depiction of identifiable facial images, which could lead to malicious deepfake generation and privacy violations. In this paper, we propose Anonymization Prompt Learning (APL) to address this problem. Specifically, we train a learnable prompt prefix for text-to-image diffusion models, which forces the model to generate anonymized facial identities, even when prompted to produce images of specific individuals. Extensive quantitative and qualitative experiments demonstrate the successful anonymization performance of APL, which anonymizes any specific individuals without compromising the quality of non-identity-specific image generation. Furthermore, we reveal the plug-and-play property of the learned prompt prefix, enabling its effective application across different pretrained text-to-image models for transferrable privacy and security protection against the risks of deepfakes.
Authors: L\'eore Bensabath, Mathis Petrovich, G\"ul Varol
Abstract: We provide results of our study on text-based 3D human motion retrieval and particularly focus on cross-dataset generalization. Due to practical reasons such as dataset-specific human body representations, existing works typically benchmarkby training and testing on partitions from the same dataset. Here, we employ a unified SMPL body format for all datasets, which allows us to perform training on one dataset, testing on the other, as well as training on a combination of datasets. Our results suggest that there exist dataset biases in standard text-motion benchmarks such as HumanML3D, KIT Motion-Language, and BABEL. We show that text augmentations help close the domain gap to some extent, but the gap remains. We further provide the first zero-shot action recognition results on BABEL, without using categorical action labels during training, opening up a new avenue for future research.
Authors: Thao Nguyen, Matthew Wallingford, Sebastin Santy, Wei-Chiu Ma, Sewoong Oh, Ludwig Schmidt, Pang Wei Koh, Ranjay Krishna
Abstract: Massive web-crawled image-text datasets lay the foundation for recent progress in multimodal learning. These datasets are designed with the goal of training a model to do well on standard computer vision benchmarks, many of which, however, have been shown to be English-centric (e.g., ImageNet). Consequently, existing data curation techniques gravitate towards using predominantly English image-text pairs and discard many potentially useful non-English samples. Our work questions this practice. Multilingual data is inherently enriching not only because it provides a gateway to learn about culturally salient concepts, but also because it depicts common concepts differently from monolingual data. We thus conduct a systematic study to explore the performance benefits of using more samples of non-English origins with respect to English vision tasks. By translating all multilingual image-text pairs from a raw web crawl to English and re-filtering them, we increase the prevalence of (translated) multilingual data in the resulting training set. Pre-training on this dataset outperforms using English-only or English-dominated datasets on ImageNet, ImageNet distribution shifts, image-English-text retrieval and on average across 38 tasks from the DataComp benchmark. On a geographically diverse task like GeoDE, we also observe improvements across all regions, with the biggest gain coming from Africa. In addition, we quantitatively show that English and non-English data are significantly different in both image and (translated) text space. We hope that our findings motivate future work to be more intentional about including multicultural and multilingual data, not just when non-English or geographically diverse tasks are involved, but to enhance model capabilities at large.
Authors: Zejun Li, Ruipu Luo, Jiwen Zhang, Minghui Qiu, Zhongyu Wei
Abstract: While large multi-modal models (LMMs) have exhibited impressive capabilities across diverse tasks, their effectiveness in handling complex tasks has been limited by the prevailing single-step reasoning paradigm. To this end, this paper proposes VoCoT, a multi-step Visually grounded object-centric Chain-of-Thought reasoning framework tailored for inference with LMMs. VoCoT is characterized by two key features: (1) object-centric reasoning paths that revolve around cross-modal shared object-level information, and (2) visually grounded representation of object concepts in a multi-modal interleaved and aligned manner, which effectively bridges the modality gap within LMMs during long-term generation. Additionally, we construct an instruction dataset to facilitate LMMs in adapting to reasoning with VoCoT. By introducing VoCoT into the prevalent open-source LMM architecture, we introduce VolCano. With only 7B parameters and limited input resolution, VolCano demonstrates excellent performance across various scenarios, surpassing SOTA models, including GPT-4V, in tasks requiring complex reasoning. Our code, data and model will be available at https://github.com/RupertLuo/VoCoT.
Authors: Butian Xiong, Xiaoyu Ye, Tze Ho Elden Tse, Kai Han, Shuguang Cui, Zhen Li
Abstract: With the emergence of Gaussian Splats, recent efforts have focused on large-scale scene geometric reconstruction. However, most of these efforts either concentrate on memory reduction or spatial space division, neglecting information in the semantic space. In this paper, we propose a novel method, named SA-GS, for fine-grained 3D geometry reconstruction using semantic-aware 3D Gaussian Splats. Specifically, we leverage prior information stored in large vision models such as SAM and DINO to generate semantic masks. We then introduce a geometric complexity measurement function to serve as soft regularization, guiding the shape of each Gaussian Splat within specific semantic areas. Additionally, we present a method that estimates the expected number of Gaussian Splats in different semantic areas, effectively providing a lower bound for Gaussian Splats in these areas. Subsequently, we extract the point cloud using a novel probability density-based extraction method, transforming Gaussian Splats into a point cloud crucial for downstream tasks. Our method also offers the potential for detailed semantic inquiries while maintaining high image-based reconstruction results. We provide extensive experiments on publicly available large-scale scene reconstruction datasets with highly accurate point clouds as ground truth and our novel dataset. Our results demonstrate the superiority of our method over current state-of-the-art Gaussian Splats reconstruction methods by a significant margin in terms of geometric-based measurement metrics. Code and additional results will soon be available on our project page.
Authors: Guan Wang, Zhimin Li, Qingchao Chen, Yang Liu
Abstract: Dynamic Scene Graph Generation (DSGG) focuses on identifying visual relationships within the spatial-temporal domain of videos. Conventional approaches often employ multi-stage pipelines, which typically consist of object detection, temporal association, and multi-relation classification. However, these methods exhibit inherent limitations due to the separation of multiple stages, and independent optimization of these sub-problems may yield sub-optimal solutions. To remedy these limitations, we propose a one-stage end-to-end framework, termed OED, which streamlines the DSGG pipeline. This framework reformulates the task as a set prediction problem and leverages pair-wise features to represent each subject-object pair within the scene graph. Moreover, another challenge of DSGG is capturing temporal dependencies, we introduce a Progressively Refined Module (PRM) for aggregating temporal context without the constraints of additional trackers or handcrafted trajectories, enabling end-to-end optimization of the network. Extensive experiments conducted on the Action Genome benchmark demonstrate the effectiveness of our design. The code and models are available at \url{https://github.com/guanw-pku/OED}.
Authors: Zerun Wang, Jiafeng Mao, Liuyu Xiang, Toshihiko Yamasaki
Abstract: Semi-supervised learning (SSL) can utilize unlabeled data to enhance model performance. In recent years, with increasingly powerful generative models becoming available, a large number of synthetic images have been uploaded to public image sets. Therefore, when collecting unlabeled data from these sources, the inclusion of synthetic images is inevitable. This prompts us to consider the impact of unlabeled data mixed with real and synthetic images on SSL. In this paper, we set up a new task, Real and Synthetic hybrid SSL (RS-SSL), to investigate this problem. We discover that current SSL methods are unable to fully utilize synthetic data and are sometimes negatively affected. Then, by analyzing the issues caused by synthetic images, we propose a new SSL method, RSMatch, to tackle the RS-SSL problem. Extensive experimental results show that RSMatch can better utilize the synthetic data in unlabeled images to improve the SSL performance. The effectiveness is further verified through ablation studies and visualization.
Authors: Zhenyang Li, Yangyang Guo, Kejie Wang, Xiaolin Chen, Liqiang Nie, Mohan Kankanhalli
Abstract: Visual Commonsense Reasoning (VCR) calls for explanatory reasoning behind question answering over visual scenes. To achieve this goal, a model is required to provide an acceptable rationale as the reason for the predicted answers. Progress on the benchmark dataset stems largely from the recent advancement of Vision-Language Transformers (VL Transformers). These models are first pre-trained on some generic large-scale vision-text datasets, and then the learned representations are transferred to the downstream VCR task. Despite their attractive performance, this paper posits that the VL Transformers do not exhibit visual commonsense, which is the key to VCR. In particular, our empirical results pinpoint several shortcomings of existing VL Transformers: small gains from pre-training, unexpected language bias, limited model architecture for the two inseparable sub-tasks, and neglect of the important object-tag correlation. With these findings, we tentatively suggest some future directions from the aspect of dataset, evaluation metric, and training tricks. We believe this work could make researchers revisit the intuition and goals of VCR, and thus help tackle the remaining challenges in visual reasoning.
Authors: Fengfan Zhou, Qianyu Zhou, Xiangtai Li, Xuequan Lu, Lizhuang Ma, Hefei Ling
Abstract: Adversarial attacks on Face Recognition (FR) systems have proven highly effective in compromising pure FR models, yet adversarial examples may be ineffective to the complete FR systems as Face Anti-Spoofing (FAS) models are often incorporated and can detect a significant number of them. To address this under-explored and essential problem, we propose a novel setting of adversarially attacking both FR and FAS models simultaneously, aiming to enhance the practicability of adversarial attacks on FR systems. In particular, we introduce a new attack method, namely Style-aligned Distribution Biasing (SDB), to improve the capacity of black-box attacks on both FR and FAS models. Specifically, our SDB framework consists of three key components. Firstly, to enhance the transferability of FAS models, we design a Distribution-aware Score Biasing module to optimize adversarial face examples away from the distribution of spoof images utilizing scores. Secondly, to mitigate the substantial style differences between live images and adversarial examples initialized with spoof images, we introduce an Instance Style Alignment module that aligns the style of adversarial examples with live images. In addition, to alleviate the conflicts between the gradients of FR and FAS models, we propose a Gradient Consistency Maintenance module to minimize disparities between the gradients using Hessian approximation. Extensive experiments showcase the superiority of our proposed attack method to state-of-the-art adversarial attacks.
Authors: Qian Wang, Abdelrahman Eldesokey, Mohit Mendiratta, Fangneng Zhan, Adam Kortylewski, Christian Theobalt, Peter Wonka
Abstract: We introduce the first zero-shot approach for Video Semantic Segmentation (VSS) based on pre-trained diffusion models. A growing research direction attempts to employ diffusion models to perform downstream vision tasks by exploiting their deep understanding of image semantics. Yet, the majority of these approaches have focused on image-related tasks like semantic correspondence and segmentation, with less emphasis on video tasks such as VSS. Ideally, diffusion-based image semantic segmentation approaches can be applied to videos in a frame-by-frame manner. However, we find their performance on videos to be subpar due to the absence of any modeling of temporal information inherent in the video data. To this end, we tackle this problem and introduce a framework tailored for VSS based on pre-trained image and video diffusion models. We propose building a scene context model based on the diffusion features, where the model is autoregressively updated to adapt to scene changes. This context model predicts per-frame coarse segmentation maps that are temporally consistent. To refine these maps further, we propose a correspondence-based refinement strategy that aggregates predictions temporally, resulting in more confident predictions. Finally, we introduce a masked modulation approach to upsample the coarse maps to the full resolution at a high quality. Experiments show that our proposed approach outperforms existing zero-shot image semantic segmentation approaches significantly on various VSS benchmarks without any training or fine-tuning. Moreover, it rivals supervised VSS approaches on the VSPW dataset despite not being explicitly trained for VSS.
Authors: Simon Vellas, Bill Psomas, Kalliopi Karadima, Dimitrios Danopoulos, Alexandros Paterakis, George Lentaris, Dimitrios Soudris, Konstantinos Karantzalos
Abstract: Real-time analysis of Martian craters is crucial for mission-critical operations, including safe landings and geological exploration. This work leverages the latest breakthroughs for on-the-edge crater detection aboard spacecraft. We rigorously benchmark several YOLO networks using a Mars craters dataset, analyzing their performance on embedded systems with a focus on optimization for low-power devices. We optimize this process for a new wave of cost-effective, commercial-off-the-shelf-based smaller satellites. Implementations on diverse platforms, including Google Coral Edge TPU, AMD Versal SoC VCK190, Nvidia Jetson Nano and Jetson AGX Orin, undergo a detailed trade-off analysis. Our findings identify optimal network-device pairings, enhancing the feasibility of crater detection on resource-constrained hardware and setting a new precedent for efficient and resilient extraterrestrial imaging. Code at: https://github.com/billpsomas/mars_crater_detection.
Authors: Tiziana D'Alessandro, Cristina Carmona-Duarte, Claudio De Stefano, Moises Diaz, Miguel A. Ferrer, Francesco Fontanella
Abstract: Alzheimer's disease is one of the most incisive illnesses among the neurodegenerative ones, and it causes a progressive decline in cognitive abilities that, in the worst cases, becomes severe enough to interfere with daily life. Currently, there is no cure, so an early diagnosis is strongly needed to try and slow its progression through medical treatments. Handwriting analysis is considered a potential tool for detecting and understanding certain neurological conditions, including Alzheimer's disease. While handwriting analysis alone cannot provide a definitive diagnosis of Alzheimer's, it may offer some insights and be used for a comprehensive assessment. The Sigma-lognormal model is conceived for movement analysis and can also be applied to handwriting. This model returns a set of lognormal parameters as output, which forms the basis for the computation of novel and significant features. This paper presents a machine learning approach applied to handwriting features extracted through the sigma-lognormal model. The aim is to develop a support system to help doctors in the diagnosis and study of Alzheimer, evaluate the effectiveness of the extracted features and finally study the relation among them.
Authors: Mengtan Zhang, Yi Feng, Qijun Chen, Rui Fan
Abstract: There has been a recent surge of interest in learning to perceive depth from monocular videos in an unsupervised fashion. A key challenge in this field is achieving robust and accurate depth estimation in challenging scenarios, particularly in regions with weak textures or where dynamic objects are present. This study makes three major contributions by delving deeply into dense correspondence priors to provide existing frameworks with explicit geometric constraints. The first novelty is a contextual-geometric depth consistency loss, which employs depth maps triangulated from dense correspondences based on estimated ego-motion to guide the learning of depth perception from contextual information, since explicitly triangulated depth maps capture accurate relative distances among pixels. The second novelty arises from the observation that there exists an explicit, deducible relationship between optical flow divergence and depth gradient. A differential property correlation loss is, therefore, designed to refine depth estimation with a specific emphasis on local variations. The third novelty is a bidirectional stream co-adjustment strategy that enhances the interaction between rigid and optical flows, encouraging the former towards more accurate correspondence and making the latter more adaptable across various scenarios under the static scene hypotheses. DCPI-Depth, a framework that incorporates all these innovative components and couples two bidirectional and collaborative streams, achieves state-of-the-art performance and generalizability across multiple public datasets, outperforming all existing prior arts. Specifically, it demonstrates accurate depth estimation in texture-less and dynamic regions, and shows more reasonable smoothness.
Authors: Sven Teufel, J\"org Gamerdinger, Jan-Patrick Kirchner, Georg Volk, Oliver Bringmann
Abstract: To ensure safe operation of autonomous vehicles in complex urban environments, complete perception of the environment is necessary. However, due to environmental conditions, sensor limitations, and occlusions, this is not always possible from a single point of view. To address this issue, collective perception is an effective method. Realistic and large-scale datasets are essential for training and evaluating collective perception methods. This paper provides the first comprehensive technical review of collective perception datasets in the context of autonomous driving. The survey analyzes existing V2V and V2X datasets, categorizing them based on different criteria such as sensor modalities, environmental conditions, and scenario variety. The focus is on their applicability for the development of connected automated vehicles. This study aims to identify the key criteria of all datasets and to present their strengths, weaknesses, and anomalies. Finally, this survey concludes by making recommendations regarding which dataset is most suitable for collective 3D object detection, tracking, and semantic segmentation.
Authors: Hongtao Wang, Rongyu Feng, Liangyi Wu, Mutian Liu, Yinuo Cui, Chunxia Zhang, Zhenbo Guo
Abstract: In seismic exploration, identifying the first break (FB) is a critical component in establishing subsurface velocity models. Various automatic picking techniques based on deep neural networks have been developed to expedite this procedure. The most popular class is using semantic segmentation networks to pick on a shot gather called 2-dimensional (2-D) picking. Generally, 2-D segmentation-based picking methods input an image of a shot gather, and output a binary segmentation map, in which the maximum of each column is the location of FB. However, current designed segmentation networks is difficult to ensure the horizontal continuity of the segmentation. Additionally, FB jumps also exist in some areas, and it is not easy for current networks to detect such jumps. Therefore, it is important to pick as much as possible and ensure horizontal continuity. To alleviate this problem, we propose a novel semantic segmentation network for the 2-D seismic FB picking task, where we introduce the dynamic snake convolution into U-Net and call the new segmentation network dynamic-snake U-Net (DSU-Net). Specifically, we develop original dynamic-snake convolution (DSConv) in CV and propose a novel DSConv module, which can extract the horizontal continuous feature in the shallow feature of the shot gather. Many experiments have shown that DSU-Net demonstrates higher accuracy and robustness than the other 2-D segmentation-based models, achieving state-of-the-art (SOTA) performance in 2-D seismic field surveys. Particularly, it can effectively detect FB jumps and better ensure the horizontal continuity of FB. In addition, the ablation experiment and the anti-noise experiment, respectively, verify the optimal structure of the DSConv module and the robustness of the picking.
Authors: Zihua Zhao, Mengxi Chen, Tianjie Dai, Jiangchao Yao, Bo han, Ya Zhang, Yanfeng Wang
Abstract: Noisy correspondence that refers to mismatches in cross-modal data pairs, is prevalent on human-annotated or web-crawled datasets. Prior approaches to leverage such data mainly consider the application of uni-modal noisy label learning without amending the impact on both cross-modal and intra-modal geometrical structures in multimodal learning. Actually, we find that both structures are effective to discriminate noisy correspondence through structural differences when being well-established. Inspired by this observation, we introduce a Geometrical Structure Consistency (GSC) method to infer the true correspondence. Specifically, GSC ensures the preservation of geometrical structures within and between modalities, allowing for the accurate discrimination of noisy samples based on structural differences. Utilizing these inferred true correspondence labels, GSC refines the learning of geometrical structures by filtering out the noisy samples. Experiments across four cross-modal datasets confirm that GSC effectively identifies noisy samples and significantly outperforms the current leading methods.
Authors: Quan Van Nguyen, Quang Huy Pham, Dan Quang Tran, Thang Kien-Bao Nguyen, Nhat-Hao Nguyen-Dang, Bao-Thien Nguyen-Tat
Abstract: Purpose: This study focuses on the development of automated text generation from radiology images, termed diagnostic captioning, to assist medical professionals in reducing clinical errors and improving productivity. The aim is to provide tools that enhance report quality and efficiency, which can significantly impact both clinical practice and deep learning research in the biomedical field. Methods: In our participation in the ImageCLEFmedical2024 Caption evaluation campaign, we explored caption prediction tasks using advanced Transformer-based models. We developed methods incorporating Transformer encoder-decoder and Query Transformer architectures. These models were trained and evaluated to generate diagnostic captions from radiology images. Results: Experimental evaluations demonstrated the effectiveness of our models, with the VisionDiagnostor-BioBART model achieving the highest BERTScore of 0.6267. This performance contributed to our team, DarkCow, achieving third place on the leaderboard. Conclusion: Our diagnostic captioning models show great promise in aiding medical professionals by generating high-quality reports efficiently. This approach can facilitate better data processing and performance optimization in medical imaging departments, ultimately benefiting healthcare delivery.
Authors: Yang Zhang, Mingying Li, Huilin Pan, Moyun Liu, Yang Zhou
Abstract: Deep learning-based fault detection methods have achieved significant success. In visual fault detection of freight trains, there exists a large characteristic difference between inter-class components (scale variance) but intra-class on the contrary, which entails scale-awareness for detectors. Moreover, the design of task-specific networks heavily relies on human expertise. As a consequence, neural architecture search (NAS) that automates the model design process gains considerable attention because of its promising performance. However, NAS is computationally intensive due to the large search space and huge data volume. In this work, we propose an efficient NAS-based framework for visual fault detection of freight trains to search for the task-specific detection head with capacities of multi-scale representation. First, we design a scale-aware search space for discovering an effective receptive field in the head. Second, we explore the robustness of data volume to reduce search costs based on the specifically designed search space, and a novel sharing strategy is proposed to reduce memory and further improve search efficiency. Extensive experimental results demonstrate the effectiveness of our method with data volume robustness, which achieves 46.8 and 47.9 mAP on the Bottom View and Side View datasets, respectively. Our framework outperforms the state-of-the-art approaches and linearly decreases the search costs with reduced data volumes.
Authors: Qi Wu, Yubo Zhao, Yifan Wang, Yu-Wing Tai, Chi-Keung Tang
Abstract: Recent advancements in Multimodal Large Language Models (MM-LLMs) have demonstrated promising potential in terms of generalization and robustness when applied to different modalities. While previous works have already achieved 3D human motion generation using various approaches including language modeling, they mostly % are mostly carefully designed use specialized architecture and are restricted to single-human motion generation. Inspired by the success of MM-LLMs, we propose MotionLLM, a simple and general framework that can achieve single-human, multi-human motion generation, and motion captioning by fine-tuning pre-trained LLMs. Specifically, we encode and quantize motions into discrete LLM-understandable tokens, which results in a unified vocabulary consisting of both motion and text tokens. With only 1--3% parameters of the LLMs trained by using adapters, our single-human motion generation achieves comparable results to those diffusion models and other trained-from-scratch transformer-based models. Additionally, we show that our approach is scalable and flexible, allowing easy extension to multi-human motion generation through autoregressive generation of single-human motions. Project page: https://knoxzhao.github.io/MotionLLM
Authors: Weiquan Wang, Jun Xiao, Chunping Wang, Wei Liu, Zhao Wang, Long Chen
Abstract: Continuous diffusion models have demonstrated their effectiveness in addressing the inherent uncertainty and indeterminacy in monocular 3D human pose estimation (HPE). Despite their strengths, the need for large search spaces and the corresponding demand for substantial training data make these models prone to generating biomechanically unrealistic poses. This challenge is particularly noticeable in occlusion scenarios, where the complexity of inferring 3D structures from 2D images intensifies. In response to these limitations, we introduce the Discrete Diffusion Pose ($\text{Di}^2\text{Pose}$), a novel framework designed for occluded 3D HPE that capitalizes on the benefits of a discrete diffusion model. Specifically, $\text{Di}^2\text{Pose}$ employs a two-stage process: it first converts 3D poses into a discrete representation through a \emph{pose quantization step}, which is subsequently modeled in latent space through a \emph{discrete diffusion process}. This methodological innovation restrictively confines the search space towards physically viable configurations and enhances the model's capability to comprehend how occlusions affect human pose within the latent space. Extensive evaluations conducted on various benchmarks (e.g., Human3.6M, 3DPW, and 3DPW-Occ) have demonstrated its effectiveness.
Authors: Yixiong Zou, Shanghang Zhang, Haichen Zhou, Yuhua Li, Ruixuan Li
Abstract: Few-shot class-incremental learning (FSCIL) is proposed to continually learn from novel classes with only a few samples after the (pre-)training on base classes with sufficient data. However, this remains a challenge. In contrast, humans can easily recognize novel classes with a few samples. Cognitive science demonstrates that an important component of such human capability is compositional learning. This involves identifying visual primitives from learned knowledge and then composing new concepts using these transferred primitives, making incremental learning both effective and interpretable. To imitate human compositional learning, we propose a cognitive-inspired method for the FSCIL task. We define and build a compositional model based on set similarities, and then equip it with a primitive composition module and a primitive reuse module. In the primitive composition module, we propose to utilize the Centered Kernel Alignment (CKA) similarity to approximate the similarity between primitive sets, allowing the training and evaluation based on primitive compositions. In the primitive reuse module, we enhance primitive reusability by classifying inputs based on primitives replaced with the closest primitives from other classes. Experiments on three datasets validate our method, showing it outperforms current state-of-the-art methods with improved interpretability. Our code is available at https://github.com/Zoilsen/Comp-FSCIL.
Authors: Avinash Nittur Ramesh, Aitor Correas-Serrano, Mar\'ia Gonz\'alez-Huici
Abstract: We present a novel synthetically generated multi-modal dataset, SCaRL, to enable the training and validation of autonomous driving solutions. Multi-modal datasets are essential to attain the robustness and high accuracy required by autonomous systems in applications such as autonomous driving. As deep learning-based solutions are becoming more prevalent for object detection, classification, and tracking tasks, there is great demand for datasets combining camera, lidar, and radar sensors. Existing real/synthetic datasets for autonomous driving lack synchronized data collection from a complete sensor suite. SCaRL provides synchronized Synthetic data from RGB, semantic/instance, and depth Cameras; Range-Doppler-Azimuth/Elevation maps and raw data from Radar; and 3D point clouds/2D maps of semantic, depth and Doppler data from coherent Lidar. SCaRL is a large dataset based on the CARLA Simulator, which provides data for diverse, dynamic scenarios and traffic conditions. SCaRL is the first dataset to include synthetic synchronized data from coherent Lidar and MIMO radar sensors. The dataset can be accessed here: https://fhr-ihs-sva.pages.fraunhofer.de/asp/scarl/
Authors: Zongkai Zhang, Zidong Xu, Wenming Yang, Qingmin Liao, Jing-Hao Xue
Abstract: Existing 3D occupancy networks demand significant hardware resources, hindering the deployment of edge devices. Binarized Neural Networks (BNN) offer substantially reduced computational and memory requirements. However, their performance decreases notably compared to full-precision networks. Moreover, it is challenging to enhance the performance of binarized models by increasing the number of binarized convolutional layers, which limits their practicability for 3D occupancy prediction. To bridge these gaps, we propose a novel binarized deep convolution (BDC) unit that effectively enhances performance while increasing the number of binarized convolutional layers. Firstly, through theoretical analysis, we demonstrate that 1 \times 1 binarized convolutions introduce minimal binarization errors. Therefore, additional binarized convolutional layers are constrained to 1 \times 1 in the BDC unit. Secondly, we introduce the per-channel weight branch to mitigate the impact of binarization errors from unimportant channel features on the performance of binarized models, thereby improving performance while increasing the number of binarized convolutional layers. Furthermore, we decompose the 3D occupancy network into four convolutional modules and utilize the proposed BDC unit to binarize these modules. Our BDC-Occ model is created by applying the proposed BDC unit to binarize the existing 3D occupancy networks. Comprehensive quantitative and qualitative experiments demonstrate that the proposed BDC-Occ is the state-of-the-art binarized 3D occupancy network algorithm.
Authors: Jinqi Wang, Yunfei Fu, Zhangcan Ding, Bailin Deng, Yu-Kun Lai, Yipeng Qin
Abstract: Inspired by the software industry's practice of offering different editions or versions of a product tailored to specific user groups or use cases, we propose a novel task, namely, training-free editioning, for text-to-image models. Specifically, we aim to create variations of a base text-to-image model without retraining, enabling the model to cater to the diverse needs of different user groups or to offer distinct features and functionalities. To achieve this, we propose that different editions of a given text-to-image model can be formulated as concept subspaces in the latent space of its text encoder (e.g., CLIP). In such a concept subspace, all points satisfy a specific user need (e.g., generating images of a cat lying on the grass/ground/falling leaves). Technically, we apply Principal Component Analysis (PCA) to obtain the desired concept subspaces from representative text embedding that correspond to a specific user need or requirement. Projecting the text embedding of a given prompt into these low-dimensional subspaces enables efficient model editioning without retraining. Intuitively, our proposed editioning paradigm enables a service provider to customize the base model into its "cat edition" (or other editions) that restricts image generation to cats, regardless of the user's prompt (e.g., dogs, people, etc.). This introduces a new dimension for product differentiation, targeted functionality, and pricing strategies, unlocking novel business models for text-to-image generators. Extensive experimental results demonstrate the validity of our approach and its potential to enable a wide range of customized text-to-image model editions across various domains and applications.
Authors: Hongming Chen, Xiang Chen, Chen Wu, Zhuoran Zheng, Jinshan Pan, Xianping Fu
Abstract: Despite significant progress has been made in image deraining, existing approaches are mostly carried out on low-resolution images. The effectiveness of these methods on high-resolution images is still unknown, especially for ultra-high-definition (UHD) images, given the continuous advancement of imaging devices. In this paper, we focus on the task of UHD image deraining, and contribute the first large-scale UHD image deraining dataset, 4K-Rain13k, that contains 13,000 image pairs at 4K resolution. Based on this dataset, we conduct a benchmark study on existing methods for processing UHD images. Furthermore, we develop an effective and efficient vision MLP-based architecture (UDR-Mixer) to better solve this task. Specifically, our method contains two building components: a spatial feature rearrangement layer that captures long-range information of UHD images, and a frequency feature modulation layer that facilitates high-quality UHD image reconstruction. Extensive experimental results demonstrate that our method performs favorably against the state-of-the-art approaches while maintaining a lower model complexity. The code and dataset will be available at https://github.com/cschenxiang/UDR-Mixer.
Authors: Cong Wang, Kuan Tian, Yonghang Guan, Jun Zhang, Zhiwei Jiang, Fei Shen, Xiao Han, Qing Gu, Wei Yang
Abstract: The success of the text-guided diffusion model has inspired the development and release of numerous powerful diffusion models within the open-source community. These models are typically fine-tuned on various expert datasets, showcasing diverse denoising capabilities. Leveraging multiple high-quality models to produce stronger generation ability is valuable, but has not been extensively studied. Existing methods primarily adopt parameter merging strategies to produce a new static model. However, they overlook the fact that the divergent denoising capabilities of the models may dynamically change across different states, such as when experiencing different prompts, initial noises, denoising steps, and spatial locations. In this paper, we propose a novel ensembling method, Adaptive Feature Aggregation (AFA), which dynamically adjusts the contributions of multiple models at the feature level according to various states (i.e., prompts, initial noises, denoising steps, and spatial locations), thereby keeping the advantages of multiple diffusion models, while suppressing their disadvantages. Specifically, we design a lightweight Spatial-Aware Block-Wise (SABW) feature aggregator that adaptive aggregates the block-wise intermediate features from multiple U-Net denoisers into a unified one. The core idea lies in dynamically producing an individual attention map for each model's features by comprehensively considering various states. It is worth noting that only SABW is trainable with about 50 million parameters, while other models are frozen. Both the quantitative and qualitative experiments demonstrate the effectiveness of our proposed Adaptive Feature Aggregation method. The code is available at https://github.com/tenvence/afa/.
Authors: Xiangyu Sun, Joo Chan Lee, Daniel Rho, Jong Hwan Ko, Usman Ali, Eunbyung Park
Abstract: The neural radiance field (NeRF) has made significant strides in representing 3D scenes and synthesizing novel views. Despite its advancements, the high computational costs of NeRF have posed challenges for its deployment in resource-constrained environments and real-time applications. As an alternative to NeRF-like neural rendering methods, 3D Gaussian Splatting (3DGS) offers rapid rendering speeds while maintaining excellent image quality. However, as it represents objects and scenes using a myriad of Gaussians, it requires substantial storage to achieve high-quality representation. To mitigate the storage overhead, we propose Factorized 3D Gaussian Splatting (F-3DGS), a novel approach that drastically reduces storage requirements while preserving image quality. Inspired by classical matrix and tensor factorization techniques, our method represents and approximates dense clusters of Gaussians with significantly fewer Gaussians through efficient factorization. We aim to efficiently represent dense 3D Gaussians by approximating them with a limited amount of information for each axis and their combinations. This method allows us to encode a substantially large number of Gaussians along with their essential attributes -- such as color, scale, and rotation -- necessary for rendering using a relatively small number of elements. Extensive experimental results demonstrate that F-3DGS achieves a significant reduction in storage costs while maintaining comparable quality in rendered images.
Authors: Steven Landgraf, Markus Hillemann, Theodor Kapler, Markus Ulrich
Abstract: While a number of promising uncertainty quantification methods have been proposed to address the prevailing shortcomings of deep neural networks like overconfidence and lack of explainability, quantifying predictive uncertainties in the context of joint semantic segmentation and monocular depth estimation has not been explored yet. Since many real-world applications are multi-modal in nature and, hence, have the potential to benefit from multi-task learning, this is a substantial gap in current literature. To this end, we conduct a comprehensive series of experiments to study how multi-task learning influences the quality of uncertainty estimates in comparison to solving both tasks separately.
Authors: Yifan Mao, Ming Li, Jian Liu, Jiayang Liu, Zihan Qin, Chunxi Chu, Jialei Xu, Wenbo Zhao, Junjun Jiang, Xianming Liu
Abstract: Surround-view depth estimation is a crucial task aims to acquire the depth maps of the surrounding views. It has many applications in real world scenarios such as autonomous driving, AR/VR and 3D reconstruction, etc. However, given that most of the data in the autonomous driving dataset is collected in daytime scenarios, this leads to poor depth model performance in the face of out-of-distribution(OoD) data. While some works try to improve the robustness of depth model under OoD data, these methods either require additional training data or lake generalizability. In this report, we introduce the DINO-SD, a novel surround-view depth estimation model. Our DINO-SD does not need additional data and has strong robustness. Our DINO-SD get the best performance in the track4 of ICRA 2024 RoboDepth Challenge.
Authors: Haoyu Zhao, Wenhang Ge, Ying-cong Chen
Abstract: Visual grounding is an essential tool that links user-provided text queries with query-specific regions within an image. Despite advancements in visual grounding models, their ability to comprehend complex queries remains limited. To overcome this limitation, we introduce LLM-Optic, an innovative method that utilizes Large Language Models (LLMs) as an optical lens to enhance existing visual grounding models in comprehending complex text queries involving intricate text structures, multiple objects, or object spatial relationships, situations that current models struggle with. LLM-Optic first employs an LLM as a Text Grounder to interpret complex text queries and accurately identify objects the user intends to locate. Then a pre-trained visual grounding model is used to generate candidate bounding boxes given the refined query by the Text Grounder. After that, LLM-Optic annotates the candidate bounding boxes with numerical marks to establish a connection between text and specific image regions, thereby linking two distinct modalities. Finally, it employs a Large Multimodal Model (LMM) as a Visual Grounder to select the marked candidate objects that best correspond to the original text query. Through LLM-Optic, we have achieved universal visual grounding, which allows for the detection of arbitrary objects specified by arbitrary human language input. Importantly, our method achieves this enhancement without requiring additional training or fine-tuning. Extensive experiments across various challenging benchmarks demonstrate that LLM-Optic achieves state-of-the-art zero-shot visual grounding capabilities.
Authors: Shujun Yang, Yu Zhang, Yao Ding, Danfeng Hong
Abstract: Insufficient prior knowledge of a captured hyperspectral image (HSI) scene may lead the experts or the automatic labeling systems to offer incorrect labels or ambiguous labels (i.e., assigning each training sample to a group of candidate labels, among which only one of them is valid; this is also known as partial label learning) during the labeling process. Accordingly, how to learn from such data with ambiguous labels is a problem of great practical importance. In this paper, we propose a novel superpixelwise low-rank approximation (LRA)-based partial label learning method, namely SLAP, which is the first to take into account partial label learning in HSI classification. SLAP is mainly composed of two phases: disambiguating the training labels and acquiring the predictive model. Specifically, in the first phase, we propose a superpixelwise LRA-based model, preparing the affinity graph for the subsequent label propagation process while extracting the discriminative representation to enhance the following classification task of the second phase. Then to disambiguate the training labels, label propagation propagates the labeling information via the affinity graph of training pixels. In the second phase, we take advantage of the resulting disambiguated training labels and the discriminative representations to enhance the classification performance. The extensive experiments validate the advantage of the proposed SLAP method over state-of-the-art methods.
Authors: Tomohiro Hayase, Braun Sacha, Hikari Yanagawa, Itsuki Orito, Yuichi Hiroi
Abstract: Social VR platforms enable social, economic, and creative activities by allowing users to create and share their own virtual spaces. In social VR, photography within a VR scene is an important indicator of visitors' activities. Although automatic identification of photo spots within a VR scene can facilitate the process of creating a VR scene and enhance the visitor experience, there are challenges in quantitatively evaluating photos taken in the VR scene and efficiently exploring the large VR scene. We propose PanoTree, an automated photo-spot explorer in VR scenes. To assess the aesthetics of images captured in VR scenes, a deep scoring network is trained on a large dataset of photos collected by a social VR platform to determine whether humans are likely to take similar photos. Furthermore, we propose a Hierarchical Optimistic Optimization (HOO)-based search algorithm to efficiently explore 3D VR spaces with the reward from the scoring network. Our user study shows that the scoring network achieves human-level performance in distinguishing randomly taken images from those taken by humans. In addition, we show applications using the explored photo spots, such as automatic thumbnail generation, support for VR world creation, and visitor flow planning within a VR scene.
Authors: Kangye Ji, Fei Cheng, Zeqing Wang, Bohu Huang
Abstract: Sample selection is the most straightforward technique to combat label noise, aiming to distinguish mislabeled samples during training and avoid the degradation of the robustness of the model. In the workflow, $\textit{selecting possibly clean data}$ and $\textit{model update}$ are iterative. However, their interplay and intrinsic characteristics hinder the robustness and efficiency of learning with noisy labels: 1)~The model chooses clean data with selection bias, leading to the accumulated error in the model update. 2) Most selection strategies leverage partner networks or supplementary information to mitigate label corruption, albeit with increased computation resources and lower throughput speed. Therefore, we employ only one network with the jump manner update to decouple the interplay and mine more semantic information from the loss for a more precise selection. Specifically, the selection of clean data for each model update is based on one of the prior models, excluding the last iteration. The strategy of model update exhibits a jump behavior in the form. Moreover, we map the outputs of the network and labels into the same semantic feature space, respectively. In this space, a detailed and simple loss distribution is generated to distinguish clean samples more effectively. Our proposed approach achieves almost up to $2.53\times$ speedup, $0.46\times$ peak memory footprint, and superior robustness over state-of-the-art works with various noise settings.
Authors: Cristian Rodriguez-Opazo, Ehsan Abbasnejad, Damien Teney, Edison Marrese-Taylor, Hamed Damirchi, Anton van den Hengel
Abstract: Contrastive Language-Image Pretraining (CLIP) stands out as a prominent method for image representation learning. Various architectures, from vision transformers (ViTs) to convolutional networks (ResNets) have been trained with CLIP to serve as general solutions to diverse vision tasks. This paper explores the differences across various CLIP-trained vision backbones. Despite using the same data and training objective, we find that these architectures have notably different representations, different classification performance across datasets, and different robustness properties to certain types of image perturbations. Our findings indicate a remarkable possible synergy across backbones by leveraging their respective strengths. In principle, classification accuracy could be improved by over 40 percentage with an informed selection of the optimal backbone per test example.Using this insight, we develop a straightforward yet powerful approach to adaptively ensemble multiple backbones. The approach uses as few as one labeled example per class to tune the adaptive combination of backbones. On a large collection of datasets, the method achieves a remarkable increase in accuracy of up to 39.1% over the best single backbone, well beyond traditional ensembles
Authors: Yong-Qiang Mao, Hanbo Bi, Liangyu Xu, Kaiqiang Chen, Zhirui Wang, Xian Sun, Kun Fu
Abstract: Research on multi-view stereo based on remote sensing images has promoted the development of large-scale urban 3D reconstruction. However, remote sensing multi-view image data suffers from the problems of occlusion and uneven brightness between views during acquisition, which leads to the problem of blurred details in depth estimation. To solve the above problem, we re-examine the deformable learning method in the Multi-View Stereo task and propose a novel paradigm based on view Space and Depth deformable Learning (SDL-MVS), aiming to learn deformable interactions of features in different view spaces and deformably model the depth ranges and intervals to enable high accurate depth estimation. Specifically, to solve the problem of view noise caused by occlusion and uneven brightness, we propose a Progressive Space deformable Sampling (PSS) mechanism, which performs deformable learning of sampling points in the 3D frustum space and the 2D image space in a progressive manner to embed source features to the reference feature adaptively. To further optimize the depth, we introduce Depth Hypothesis deformable Discretization (DHD), which achieves precise positioning of the depth prior by adaptively adjusting the depth range hypothesis and performing deformable discretization of the depth interval hypothesis. Finally, our SDL-MVS achieves explicit modeling of occlusion and uneven brightness faced in multi-view stereo through the deformable learning paradigm of view space and depth, achieving accurate multi-view depth estimation. Extensive experiments on LuoJia-MVS and WHU datasets show that our SDL-MVS reaches state-of-the-art performance. It is worth noting that our SDL-MVS achieves an MAE error of 0.086, an accuracy of 98.9% for <0.6m, and 98.9% for <3-interval on the LuoJia-MVS dataset under the premise of three views as input.
Authors: Juan C. P\'erez, Alejandro Pardo, Mattia Soldan, Hani Itani, Juan Leon-Alcazar, Bernard Ghanem
Abstract: This study investigates whether Compressed-Language Models (CLMs), i.e. language models operating on raw byte streams from Compressed File Formats~(CFFs), can understand files compressed by CFFs. We focus on the JPEG format as a representative CFF, given its commonality and its representativeness of key concepts in compression, such as entropy coding and run-length encoding. We test if CLMs understand the JPEG format by probing their capabilities to perform along three axes: recognition of inherent file properties, handling of files with anomalies, and generation of new files. Our findings demonstrate that CLMs can effectively perform these tasks. These results suggest that CLMs can understand the semantics of compressed data when directly operating on the byte streams of files produced by CFFs. The possibility to directly operate on raw compressed files offers the promise to leverage some of their remarkable characteristics, such as their ubiquity, compactness, multi-modality and segment-nature.
Authors: Yaohua Zha, Naiqi Li, Yanzi Wang, Tao Dai, Hang Guo, Bin Chen, Zhi Wang, Zhihao Ouyang, Shu-Tao Xia
Abstract: The pre-trained point cloud model based on Masked Point Modeling (MPM) has exhibited substantial improvements across various tasks. However, these models heavily rely on the Transformer, leading to quadratic complexity and limited decoder, hindering their practice application. To address this limitation, we first conduct a comprehensive analysis of existing Transformer-based MPM, emphasizing the idea that redundancy reduction is crucial for point cloud analysis. To this end, we propose a Locally constrained Compact point cloud Model (LCM) consisting of a locally constrained compact encoder and a locally constrained Mamba-based decoder. Our encoder replaces self-attention with our local aggregation layers to achieve an elegant balance between performance and efficiency. Considering the varying information density between masked and unmasked patches in the decoder inputs of MPM, we introduce a locally constrained Mamba-based decoder. This decoder ensures linear complexity while maximizing the perception of point cloud geometry information from unmasked patches with higher information density. Extensive experimental results show that our compact model significantly surpasses existing Transformer-based models in both performance and efficiency, especially our LCM-based Point-MAE model, compared to the Transformer-based model, achieved an improvement of 2.24%, 0.87%, and 0.94% in performance on the three variants of ScanObjectNN while reducing parameters by 88% and computation by 73%.
Authors: Yong Liu, Hang Dong, Jinshan Pan, Qingji Dong, Kai Chen, Rongxiang Zhang, Xing Mei, Lean Fu, Fei Wang
Abstract: Diffusion models significantly improve the quality of super-resolved images with their impressive content generation capabilities. However, the huge computational costs limit the applications of these methods.Recent efforts have explored reasonable inference acceleration to reduce the number of sampling steps, but the computational cost remains high as each step is performed on the entire image.This paper introduces PatchScaler, a patch-independent diffusion-based single image super-resolution (SR) method, designed to enhance the efficiency of the inference process.The proposed method is motivated by the observation that not all the image patches within an image need the same sampling steps for reconstructing high-resolution images.Based on this observation, we thus develop a Patch-adaptive Group Sampling (PGS) to divide feature patches into different groups according to the patch-level reconstruction difficulty and dynamically assign an appropriate sampling configuration for each group so that the inference speed can be better accelerated.In addition, to improve the denoising ability at each step of the sampling, we develop a texture prompt to guide the estimations of the diffusion model by retrieving high-quality texture priors from a patch-independent reference texture memory.Experiments show that our PatchScaler achieves favorable performance in both quantitative and qualitative evaluations with fast inference speed.Our code and model are available at \url{https://github.com/yongliuy/PatchScaler}.
Authors: Yiming Li, Zehong Wang, Yue Wang, Zhiding Yu, Zan Gojcic, Marco Pavone, Chen Feng, Jose M. Alvarez
Abstract: Humans naturally retain memories of permanent elements, while ephemeral moments often slip through the cracks of memory. This selective retention is crucial for robotic perception, localization, and mapping. To endow robots with this capability, we introduce 3D Gaussian Mapping (3DGM), a self-supervised, camera-only offline mapping framework grounded in 3D Gaussian Splatting. 3DGM converts multitraverse RGB videos from the same region into a Gaussian-based environmental map while concurrently performing 2D ephemeral object segmentation. Our key observation is that the environment remains consistent across traversals, while objects frequently change. This allows us to exploit self-supervision from repeated traversals to achieve environment-object decomposition. More specifically, 3DGM formulates multitraverse environmental mapping as a robust differentiable rendering problem, treating pixels of the environment and objects as inliers and outliers, respectively. Using robust feature distillation, feature residuals mining, and robust optimization, 3DGM jointly performs 3D mapping and 2D segmentation without human intervention. We build the Mapverse benchmark, sourced from the Ithaca365 and nuPlan datasets, to evaluate our method in unsupervised 2D segmentation, 3D reconstruction, and neural rendering. Extensive results verify the effectiveness and potential of our method for self-driving and robotics.
Authors: Jian Zhao, Lei Jin, Jianshu Li, Zheng Zhu, Yinglei Teng, Jiaojiao Zhao, Sadaf Gulshad, Zheng Wang, Bo Zhao, Xiangbo Shu, Yunchao Wei, Xuecheng Nie, Xiaojie Jin, Xiaodan Liang, Shin'ichi Satoh, Yandong Guo, Cewu Lu, Junliang Xing, Jane Shen Shengmei
Abstract: The SkatingVerse Workshop & Challenge aims to encourage research in developing novel and accurate methods for human action understanding. The SkatingVerse dataset used for the SkatingVerse Challenge has been publicly released. There are two subsets in the dataset, i.e., the training subset and testing subset. The training subsets consists of 19,993 RGB video sequences, and the testing subsets consists of 8,586 RGB video sequences. Around 10 participating teams from the globe competed in the SkatingVerse Challenge. In this paper, we provide a brief summary of the SkatingVerse Workshop & Challenge including brief introductions to the top three methods. The submission leaderboard will be reopened for researchers that are interested in the human action understanding challenge. The benchmark dataset and other information can be found at: https://skatingverse.github.io/.
Authors: Baoren Xiao, Hao Ni, Weixin Yang
Abstract: Generative adversarial networks (GANs) have emerged as a powerful tool for generating high-fidelity data. However, the main bottleneck of existing approaches is the lack of supervision on the generator training, which often results in undamped oscillation and unsatisfactory performance. To address this issue, we propose an algorithm called Monte Carlo GAN (MCGAN). This approach, utilizing an innovative generative loss function, termly the regression loss, reformulates the generator training as a regression task and enables the generator training by minimizing the mean squared error between the discriminator's output of real data and the expected discriminator of fake data. We demonstrate the desirable analytic properties of the regression loss, including discriminability and optimality, and show that our method requires a weaker condition on the discriminator for effective generator training. These properties justify the strength of this approach to improve the training stability while retaining the optimality of GAN by leveraging strong supervision of the regression loss. Numerical results on CIFAR-10 and CIFAR-100 datasets demonstrate that the proposed MCGAN significantly and consistently improves the existing state-of-the-art GAN models in terms of quality, accuracy, training stability, and learned latent space. Furthermore, the proposed algorithm exhibits great flexibility for integrating with a variety of backbone models to generate spatial images, temporal time-series, and spatio-temporal video data.
Authors: Jin Wang, Shichao Dong, Yapeng Zhu, Kelu Yao, Weidong Zhao, Chao Li, Ping Luo
Abstract: Compositional reasoning capabilities are usually considered as fundamental skills to characterize human perception. Recent studies show that current Vision Language Models (VLMs) surprisingly lack sufficient knowledge with respect to such capabilities. To this end, we propose to thoroughly diagnose the composition representations encoded by VLMs, systematically revealing the potential cause for this weakness. Specifically, we propose evaluation methods from a novel game-theoretic view to assess the vulnerability of VLMs on different aspects of compositional understanding, e.g., relations and attributes. Extensive experimental results demonstrate and validate several insights to understand the incapabilities of VLMs on compositional reasoning, which provide useful and reliable guidance for future studies. The deliverables will be updated at https://vlms-compositionality-gametheory.github.io/.
Authors: Zhaoyang Sun, Shengwu Xiong, Yaxiong Chen, Yi Rong
Abstract: The absence of real targets to guide the model training is one of the main problems with the makeup transfer task. Most existing methods tackle this problem by synthesizing pseudo ground truths (PGTs). However, the generated PGTs are often sub-optimal and their imprecision will eventually lead to performance degradation. To alleviate this issue, in this paper, we propose a novel Content-Style Decoupled Makeup Transfer (CSD-MT) method, which works in a purely unsupervised manner and thus eliminates the negative effects of generating PGTs. Specifically, based on the frequency characteristics analysis, we assume that the low-frequency (LF) component of a face image is more associated with its makeup style information, while the high-frequency (HF) component is more related to its content details. This assumption allows CSD-MT to decouple the content and makeup style information in each face image through the frequency decomposition. After that, CSD-MT realizes makeup transfer by maximizing the consistency of these two types of information between the transferred result and input images, respectively. Two newly designed loss functions are also introduced to further improve the transfer performance. Extensive quantitative and qualitative analyses show the effectiveness of our CSD-MT method. Our code is available at https://github.com/Snowfallingplum/CSD-MT.
Authors: Yisi Luo, Xile Zhao, Kai Ye, Deyu Meng
Abstract: Recently, we have witnessed the success of total variation (TV) for many imaging applications. However, traditional TV is defined on the original pixel domain, which limits its potential. In this work, we suggest a new TV regularization defined on the neural domain. Concretely, the discrete data is continuously and implicitly represented by a deep neural network (DNN), and we use the derivatives of DNN outputs w.r.t. input coordinates to capture local correlations of data. As compared with classical TV on the original domain, the proposed TV on the neural domain (termed NeurTV) enjoys two advantages. First, NeurTV is not limited to meshgrid but is suitable for both meshgrid and non-meshgrid data. Second, NeurTV can more exactly capture local correlations across data for any direction and any order of derivatives attributed to the implicit and continuous nature of neural domain. We theoretically reinterpret NeurTV under the variational approximation framework, which allows us to build the connection between classical TV and NeurTV and inspires us to develop variants (e.g., NeurTV with arbitrary resolution and space-variant NeurTV). Extensive numerical experiments with meshgrid data (e.g., color and hyperspectral images) and non-meshgrid data (e.g., point clouds and spatial transcriptomics) showcase the effectiveness of the proposed methods.
Authors: Junyoung Seo, Kazumi Fukuda, Takashi Shibuya, Takuya Narihira, Naoki Murata, Shoukang Hu, Chieh-Hsin Lai, Seungryong Kim, Yuki Mitsufuji
Abstract: Generating novel views from a single image remains a challenging task due to the complexity of 3D scenes and the limited diversity in the existing multi-view datasets to train a model on. Recent research combining large-scale text-to-image (T2I) models with monocular depth estimation (MDE) has shown promise in handling in-the-wild images. In these methods, an input view is geometrically warped to novel views with estimated depth maps, then the warped image is inpainted by T2I models. However, they struggle with noisy depth maps and loss of semantic details when warping an input view to novel viewpoints. In this paper, we propose a novel approach for single-shot novel view synthesis, a semantic-preserving generative warping framework that enables T2I generative models to learn where to warp and where to generate, through augmenting cross-view attention with self-attention. Our approach addresses the limitations of existing methods by conditioning the generative model on source view images and incorporating geometric warping signals. Qualitative and quantitative evaluations demonstrate that our model outperforms existing methods in both in-domain and out-of-domain scenarios. Project page is available at https://GenWarp-NVS.github.io/.
Authors: Shengjie Liu, Lu Zhang
Abstract: Remote sensing data provide a low-cost solution for large-scale monitoring of air pollution via the retrieval of aerosol optical depth (AOD), but is often limited by cloud contamination. Existing methods for AOD reconstruction rely on temporal information. However, for remote sensing data at high spatial resolution, multi-temporal observations are often unavailable. In this letter, we take advantage of deep representation learning from convolutional neural networks and propose Deep Feature Gaussian Processes (DFGP) for single-scene AOD reconstruction. By using deep learning, we transform the variables to a feature space with better explainable power. By using Gaussian processes, we explicitly consider the correlation between observed AOD and missing AOD in spatial and feature domains. Experiments on two AOD datasets with real-world cloud patterns showed that the proposed method outperformed deep CNN and random forest, achieving R$^2$ of 0.7431 on MODIS AOD and R$^2$ of 0.9211 on EMIT AOD, compared to deep CNN's R$^2$ of 0.6507 and R$^2$ of 0.8619. The proposed methods increased R$^2$ by over 0.35 compared to the popular random forest in AOD reconstruction. The data and code used in this study are available at \url{https://skrisliu.com/dfgp}.
Authors: Qiang Wang, Minghua Liu, Junjun Hu, Fan Jiang, Mu Xu
Abstract: Generating realistic animated videos from static images is an important area of research in computer vision. Methods based on physical simulation and motion prediction have achieved notable advances, but they are often limited to specific object textures and motion trajectories, failing to exhibit highly complex environments and physical dynamics. In this paper, we introduce an open-domain controllable image animation method using motion priors with video diffusion models. Our method achieves precise control over the direction and speed of motion in the movable region by extracting the motion field information from videos and learning moving trajectories and strengths. Current pretrained video generation models are typically limited to producing very short videos, typically less than 30 frames. In contrast, we propose an efficient long-duration video generation method based on noise reschedule specifically tailored for image animation tasks, facilitating the creation of videos over 100 frames in length while maintaining consistency in content scenery and motion coordination. Specifically, we decompose the denoise process into two distinct phases: the shaping of scene contours and the refining of motion details. Then we reschedule the noise to control the generated frame sequences maintaining long-distance noise correlation. We conducted extensive experiments with 10 baselines, encompassing both commercial tools and academic methodologies, which demonstrate the superiority of our method. Our project page: \url{https://wangqiang9.github.io/Controllable.github.io/}
Authors: Vadim Ezhov, Hyoungseob Park, Zhaoyang Zhang, Rishi Upadhyay, Howard Zhang, Chethan Chinder Chandrappa, Achuta Kadambi, Yunhao Ba, Julie Dorsey, Alex Wong
Abstract: We propose a method for depth estimation under different illumination conditions, i.e., day and night time. As photometry is uninformative in regions under low-illumination, we tackle the problem through a multi-sensor fusion approach, where we take as input an additional synchronized sparse point cloud (i.e., from a LiDAR) projected onto the image plane as a sparse depth map, along with a camera image. The crux of our method lies in the use of the abundantly available synthetic data to first approximate the 3D scene structure by learning a mapping from sparse to (coarse) dense depth maps along with their predictive uncertainty - we term this, SpaDe. In poorly illuminated regions where photometric intensities do not afford the inference of local shape, the coarse approximation of scene depth serves as a prior; the uncertainty map is then used with the image to guide refinement through an uncertainty-driven residual learning (URL) scheme. The resulting depth completion network leverages complementary strengths from both modalities - depth is sparse but insensitive to illumination and in metric scale, and image is dense but sensitive with scale ambiguity. SpaDe can be used in a plug-and-play fashion, which allows for 25% improvement when augmented onto existing methods to preprocess sparse depth. We demonstrate URL on the nuScenes dataset where we improve over all baselines by an average 11.65% in all-day scenarios, 11.23% when tested specifically for daytime, and 13.12% for nighttime scenes.
Authors: Tingwei Liu, Yasutomo Kawanishi, Takahiro Komamizu, Ichiro Ide
Abstract: This paper focuses on tracking birds that appear small in a panoramic video. When the size of the tracked object is small in the image (small object tracking) and move quickly, object detection and association suffers. To address these problems, we propose Adaptive Slicing Aided Hyper Inference (Adaptive SAHI), which reduces the candidate regions to apply detection, and Detection History-aware Similarity Criterion (DHSC), which accurately associates objects in consecutive frames based on the detection history. Experiments on the NUBird2022 dataset verifies the effectiveness of the proposed method by showing improvements in both accuracy and speed.
Authors: Yujie Wang, Praneeth Chakravarthula, Baoquan Chen
Abstract: 3D Gaussian Splatting-based techniques have recently advanced 3D scene reconstruction and novel view synthesis, achieving high-quality real-time rendering. However, these approaches are inherently limited by the underlying pinhole camera assumption in modeling the images and hence only work for All-in-Focus (AiF) sharp image inputs. This severely affects their applicability in real-world scenarios where images often exhibit defocus blur due to the limited depth-of-field (DOF) of imaging devices. Additionally, existing 3D Gaussian Splatting (3DGS) methods also do not support rendering of DOF effects. To address these challenges, we introduce DOF-GS that allows for rendering adjustable DOF effects, removing defocus blur as well as refocusing of 3D scenes, all from multi-view images degraded by defocus blur. To this end, we re-imagine the traditional Gaussian Splatting pipeline by employing a finite aperture camera model coupled with explicit, differentiable defocus rendering guided by the Circle-of-Confusion (CoC). The proposed framework provides for dynamic adjustment of DOF effects by changing the aperture and focal distance of the underlying camera model on-demand. It also enables rendering varying DOF effects of 3D scenes post-optimization, and generating AiF images from defocused training images. Furthermore, we devise a joint optimization strategy to further enhance details in the reconstructed scenes by jointly optimizing rendered defocused and AiF images. Our experimental results indicate that DOF-GS produces high-quality sharp all-in-focus renderings conditioned on inputs compromised by defocus blur, with the training process incurring only a modest increase in GPU memory consumption. We further demonstrate the applications of the proposed method for adjustable defocus rendering and refocusing of the 3D scene from input images degraded by defocus blur.
Authors: J. D. Peiffer, Kunal Shah, Shawana Anarwala, Kayan Abdou, R. James Cotton
Abstract: Video and wearable sensor data provide complementary information about human movement. Video provides a holistic understanding of the entire body in the world while wearable sensors provide high-resolution measurements of specific body segments. A robust method to fuse these modalities and obtain biomechanically accurate kinematics would have substantial utility for clinical assessment and monitoring. While multiple video-sensor fusion methods exist, most assume that a time-intensive, and often brittle, sensor-body calibration process has already been performed. In this work, we present a method to combine handheld smartphone video and uncalibrated wearable sensor data at their full temporal resolution. Our monocular, video-only, biomechanical reconstruction already performs well, with only several degrees of error at the knee during walking compared to markerless motion capture. Reconstructing from a fusion of video and wearable sensor data further reduces this error. We validate this in a mixture of people with no gait impairments, lower limb prosthesis users, and individuals with a history of stroke. We also show that sensor data allows tracking through periods of visual occlusion.
Authors: Amin Ahmadi Kasani, Hedieh Sajedi
Abstract: The way organs are positioned and moved in the workplace can cause pain and physical harm. Therefore, ergonomists use ergonomic risk assessments based on visual observation of the workplace, or review pictures and videos taken in the workplace. Sometimes the workers in the photos are not in perfect condition. Some parts of the workers' bodies may not be in the camera's field of view, could be obscured by objects, or by self-occlusion, this is the main problem in 2D human posture recognition. It is difficult to predict the position of body parts when they are not visible in the image, and geometric mathematical methods are not entirely suitable for this purpose. Therefore, we created a dataset with artificial images of a 3D human model, specifically for painful postures, and real human photos from different viewpoints. Each image we captured was based on a predefined joint angle for each 3D model or human model. We created various images, including images where some body parts are not visible. Nevertheless, the joint angle is estimated beforehand, so we could study the case by converting the input images into the sequence of joint connections between predefined body parts and extracting the desired joint angle with a convolutional neural network. In the end, we obtained root mean square error (RMSE) of 12.89 and mean absolute error (MAE) of 4.7 on the test dataset.
Authors: Sai Raj Kishore Perla, Yizhi Wang, Ali Mahdavi-Amiri, Hao Zhang
Abstract: We present a novel approach for single-image mesh texturing, which employs a diffusion model with judicious conditioning to seamlessly transfer an object's texture from a single RGB image to a given 3D mesh object. We do not assume that the two objects belong to the same category, and even if they do, there can be significant discrepancies in their geometry and part proportions. Our method aims to rectify the discrepancies by conditioning a pre-trained Stable Diffusion generator with edges describing the mesh through ControlNet, and features extracted from the input image using IP-Adapter to generate textures that respect the underlying geometry of the mesh and the input texture without any optimization or training. We also introduce Image Inversion, a novel technique to quickly personalize the diffusion model for a single concept using a single image, for cases where the pre-trained IP-Adapter falls short in capturing all the details from the input image faithfully. Experimental results demonstrate the efficiency and effectiveness of our edge-aware single-image mesh texturing approach, coined EASI-Tex, in preserving the details of the input texture on diverse 3D objects, while respecting their geometry.
Authors: Niloofar Azizi, Mohsen Fayyaz, Horst Bischof
Abstract: Understanding human behavior fundamentally relies on accurate 3D human pose estimation. Graph Convolutional Networks (GCNs) have recently shown promising advancements, delivering state-of-the-art performance with rather lightweight architectures. In the context of graph-structured data, leveraging the eigenvectors of the graph Laplacian matrix for positional encoding is effective. Yet, the approach does not specify how to handle scenarios where edges in the input graph are missing. To this end, we propose a novel positional encoding technique, PerturbPE, that extracts consistent and regular components from the eigenbasis. Our method involves applying multiple perturbations and taking their average to extract the consistent and regular component from the eigenbasis. PerturbPE leverages the Rayleigh-Schrodinger Perturbation Theorem (RSPT) for calculating the perturbed eigenvectors. Employing this labeling technique enhances the robustness and generalizability of the model. Our results support our theoretical findings, e.g. our experimental analysis observed a performance enhancement of up to $12\%$ on the Human3.6M dataset in instances where occlusion resulted in the absence of one edge. Furthermore, our novel approach significantly enhances performance in scenarios where two edges are missing, setting a new benchmark for state-of-the-art.
Authors: Shenyuan Gao, Jiazhi Yang, Li Chen, Kashyap Chitta, Yihang Qiu, Andreas Geiger, Jun Zhang, Hongyang Li
Abstract: World models can foresee the outcomes of different actions, which is of paramount importance for autonomous driving. Nevertheless, existing driving world models still have limitations in generalization to unseen environments, prediction fidelity of critical details, and action controllability for flexible application. In this paper, we present Vista, a generalizable driving world model with high fidelity and versatile controllability. Based on a systematic diagnosis of existing methods, we introduce several key ingredients to address these limitations. To accurately predict real-world dynamics at high resolution, we propose two novel losses to promote the learning of moving instances and structural information. We also devise an effective latent replacement approach to inject historical frames as priors for coherent long-horizon rollouts. For action controllability, we incorporate a versatile set of controls from high-level intentions (command, goal point) to low-level maneuvers (trajectory, angle, and speed) through an efficient learning strategy. After large-scale training, the capabilities of Vista can seamlessly generalize to different scenarios. Extensive experiments on multiple datasets show that Vista outperforms the most advanced general-purpose video generator in over 70% of comparisons and surpasses the best-performing driving world model by 55% in FID and 27% in FVD. Moreover, for the first time, we utilize the capacity of Vista itself to establish a generalizable reward for real-world action evaluation without accessing the ground truth actions.
Authors: Ruizhi Shao, Youxin Pang, Zerong Zheng, Jingxiang Sun, Yebin Liu
Abstract: We present a novel approach for generating high-quality, spatio-temporally coherent human videos from a single image under arbitrary viewpoints. Our framework combines the strengths of U-Nets for accurate condition injection and diffusion transformers for capturing global correlations across viewpoints and time. The core is a cascaded 4D transformer architecture that factorizes attention across views, time, and spatial dimensions, enabling efficient modeling of the 4D space. Precise conditioning is achieved by injecting human identity, camera parameters, and temporal signals into the respective transformers. To train this model, we curate a multi-dimensional dataset spanning images, videos, multi-view data and 3D/4D scans, along with a multi-dimensional training strategy. Our approach overcomes the limitations of previous methods based on GAN or UNet-based diffusion models, which struggle with complex motions and viewpoint changes. Through extensive experiments, we demonstrate our method's ability to synthesize realistic, coherent and free-view human videos, paving the way for advanced multimedia applications in areas such as virtual reality and animation. Our project website is https://human4dit.github.io.
Authors: Zhengfei Kuang, Shengqu Cai, Hao He, Yinghao Xu, Hongsheng Li, Leonidas Guibas, Gordon Wetzstein
Abstract: Research on video generation has recently made tremendous progress, enabling high-quality videos to be generated from text prompts or images. Adding control to the video generation process is an important goal moving forward and recent approaches that condition video generation models on camera trajectories make strides towards it. Yet, it remains challenging to generate a video of the same scene from multiple different camera trajectories. Solutions to this multi-video generation problem could enable large-scale 3D scene generation with editable camera trajectories, among other applications. We introduce collaborative video diffusion (CVD) as an important step towards this vision. The CVD framework includes a novel cross-video synchronization module that promotes consistency between corresponding frames of the same video rendered from different camera poses using an epipolar attention mechanism. Trained on top of a state-of-the-art camera-control module for video generation, CVD generates multiple videos rendered from different camera trajectories with significantly better consistency than baselines, as shown in extensive experiments. Project page: https://collaborativevideodiffusion.github.io/.
Authors: Jiaming Liu, Chenxuan Li, Guanqun Wang, Lily Lee, Kaichen Zhou, Sixiang Chen, Chuyan Xiong, Jiaxin Ge, Renrui Zhang, Shanghang Zhang
Abstract: Robot manipulation policies have shown unsatisfactory action performance when confronted with novel task or object instances. Hence, the capability to automatically detect and self-correct failure action is essential for a practical robotic system. Recently, Multimodal Large Language Models (MLLMs) have shown promise in visual instruction following and demonstrated strong reasoning abilities in various tasks. To unleash general MLLMs as an end-to-end robotic agent, we introduce a Self-Corrected (SC)-MLLM, equipping our model not only to predict end-effector poses but also to autonomously recognize and correct failure actions. Specifically, we first conduct parameter-efficient fine-tuning to empower MLLM with pose prediction ability, which is reframed as a language modeling problem. When facing execution failures, our model learns to identify low-level action error causes (i.e., position and rotation errors) and adaptively seeks prompt feedback from experts. Based on the feedback, SC-MLLM rethinks the current failure scene and generates the corrected actions. Furthermore, we design a continuous policy learning method for successfully corrected samples, enhancing the model's adaptability to the current scene configuration and reducing the frequency of expert intervention. To evaluate our SC-MLLM, we conduct extensive experiments in both simulation and real-world settings. SC-MLLM agent significantly improve manipulation accuracy compared to previous state-of-the-art robotic MLLM (ManipLLM), increasing from 57\% to 79\% on seen object categories and from 47\% to 69\% on unseen novel categories.
Authors: Hao Dong, Yue Zhao, Eleni Chatzi, Olga Fink
Abstract: Detecting out-of-distribution (OOD) samples is important for deploying machine learning models in safety-critical applications such as autonomous driving and robot-assisted surgery. Existing research has mainly focused on unimodal scenarios on image data. However, real-world applications are inherently multimodal, which makes it essential to leverage information from multiple modalities to enhance the efficacy of OOD detection. To establish a foundation for more realistic Multimodal OOD Detection, we introduce the first-of-its-kind benchmark, MultiOOD, characterized by diverse dataset sizes and varying modality combinations. We first evaluate existing unimodal OOD detection algorithms on MultiOOD, observing that the mere inclusion of additional modalities yields substantial improvements. This underscores the importance of utilizing multiple modalities for OOD detection. Based on the observation of Modality Prediction Discrepancy between in-distribution (ID) and OOD data, and its strong correlation with OOD performance, we propose the Agree-to-Disagree (A2D) algorithm to encourage such discrepancy during training. Moreover, we introduce a novel outlier synthesis method, NP-Mix, which explores broader feature spaces by leveraging the information from nearest neighbor classes and complements A2D to strengthen OOD detection performance. Extensive experiments on MultiOOD demonstrate that training with A2D and NP-Mix improves existing OOD detection algorithms by a large margin. Our source code and MultiOOD benchmark are available at https://github.com/donghao51/MultiOOD.
Authors: Jiahui Lei, Yijia Weng, Adam Harley, Leonidas Guibas, Kostas Daniilidis
Abstract: We introduce 4D Motion Scaffolds (MoSca), a neural information processing system designed to reconstruct and synthesize novel views of dynamic scenes from monocular videos captured casually in the wild. To address such a challenging and ill-posed inverse problem, we leverage prior knowledge from foundational vision models, lift the video data to a novel Motion Scaffold (MoSca) representation, which compactly and smoothly encodes the underlying motions / deformations. The scene geometry and appearance are then disentangled from the deformation field, and are encoded by globally fusing the Gaussians anchored onto the MoSca and optimized via Gaussian Splatting. Additionally, camera poses can be seamlessly initialized and refined during the dynamic rendering process, without the need for other pose estimation tools. Experiments demonstrate state-of-the-art performance on dynamic rendering benchmarks.
Authors: Shuai Zeng, Wenzhao Zheng, Jiwen Lu, Haibin Yan
Abstract: 3D object detection aims to recover the 3D information of concerning objects and serves as the fundamental task of autonomous driving perception. Its performance greatly depends on the scale of labeled training data, yet it is costly to obtain high-quality annotations for point cloud data. While conventional methods focus on generating pseudo-labels for unlabeled samples as supplements for training, the structural nature of 3D point cloud data facilitates the composition of objects and backgrounds to synthesize realistic scenes. Motivated by this, we propose a hardness-aware scene synthesis (HASS) method to generate adaptive synthetic scenes to improve the generalization of the detection models. We obtain pseudo-labels for unlabeled objects and generate diverse scenes with different compositions of objects and backgrounds. As the scene synthesis is sensitive to the quality of pseudo-labels, we further propose a hardness-aware strategy to reduce the effect of low-quality pseudo-labels and maintain a dynamic pseudo-database to ensure the diversity and quality of synthetic scenes. Extensive experimental results on the widely used KITTI and Waymo datasets demonstrate the superiority of the proposed HASS method, which outperforms existing semi-supervised learning methods on 3D object detection. Code: https://github.com/wzzheng/HASS.
Authors: Laurens Samson, Nimrod Barazani, Sennay Ghebreab, Yuki M. Asano
Abstract: This paper aims to advance our understanding of how Visual Language Models (VLMs) handle privacy-sensitive information, a crucial concern as these technologies become integral to everyday life. To this end, we introduce a new benchmark PrivBench, which contains images from 8 sensitive categories such as passports, or fingerprints. We evaluate 10 state-of-the-art VLMs on this benchmark and observe a generally limited understanding of privacy, highlighting a significant area for model improvement. Based on this we introduce PrivTune, a new instruction-tuning dataset aimed at equipping VLMs with knowledge about visual privacy. By tuning two pretrained VLMs, TinyLLaVa and MiniGPT-v2, on this small dataset, we achieve strong gains in their ability to recognize sensitive content, outperforming even GPT4-V. At the same time, we show that privacy-tuning only minimally affects the VLMs performance on standard benchmarks such as VQA. Overall, this paper lays out a crucial challenge for making VLMs effective in handling real-world data safely and provides a simple recipe that takes the first step towards building privacy-aware VLMs.
Authors: Zhuoling Li, Xiaogang Xu, Zhenhua Xu, SerNam Lim, Hengshuang Zhao
Abstract: Due to the need to interact with the real world, embodied agents are required to possess comprehensive prior knowledge, long-horizon planning capability, and a swift response speed. Despite recent large language model (LLM) based agents achieving promising performance, they still exhibit several limitations. For instance, the output of LLMs is a descriptive sentence, which is ambiguous when determining specific actions. To address these limitations, we introduce the large auto-regressive model (LARM). LARM leverages both text and multi-view images as input and predicts subsequent actions in an auto-regressive manner. To train LARM, we develop a novel data format named auto-regressive node transmission structure and assemble a corresponding dataset. Adopting a two-phase training regimen, LARM successfully harvests enchanted equipment in Minecraft, which demands significantly more complex decision-making chains than the highest achievements of prior best methods. Besides, the speed of LARM is 6.8x faster.
Authors: Shaoyuan Xie, Lingdong Kong, Wenwei Zhang, Jiawei Ren, Liang Pan, Kai Chen, Ziwei Liu
Abstract: Recent advancements in bird's eye view (BEV) representations have shown remarkable promise for in-vehicle 3D perception. However, while these methods have achieved impressive results on standard benchmarks, their robustness in varied conditions remains insufficiently assessed. In this study, we present RoboBEV, an extensive benchmark suite designed to evaluate the resilience of BEV algorithms. This suite incorporates a diverse set of camera corruption types, each examined over three severity levels. Our benchmarks also consider the impact of complete sensor failures that occur when using multi-modal models. Through RoboBEV, we assess 33 state-of-the-art BEV-based perception models spanning tasks like detection, map segmentation, depth estimation, and occupancy prediction. Our analyses reveal a noticeable correlation between the model's performance on in-distribution datasets and its resilience to out-of-distribution challenges. Our experimental results also underline the efficacy of strategies like pre-training and depth-free BEV transformations in enhancing robustness against out-of-distribution data. Furthermore, we observe that leveraging extensive temporal information significantly improves the model's robustness. Based on our observations, we design an effective robustness enhancement strategy based on the CLIP model. The insights from this study pave the way for the development of future BEV models that seamlessly combine accuracy with real-world robustness.
Authors: Kuan-Chih Huang, Xiangtai Li, Lu Qi, Shuicheng Yan, Ming-Hsuan Yang
Abstract: Recent advancements in multimodal large language models (LLMs) have shown their potential in various domains, especially concept reasoning. Despite these developments, applications in understanding 3D environments remain limited. This paper introduces Reason3D, a novel LLM designed for comprehensive 3D understanding. Reason3D takes point cloud data and text prompts as input to produce textual responses and segmentation masks, facilitating advanced tasks like 3D reasoning segmentation, hierarchical searching, express referring, and question answering with detailed mask outputs. Specifically, we propose a hierarchical mask decoder to locate small objects within expansive scenes. This decoder initially generates a coarse location estimate covering the object's general area. This foundational estimation facilitates a detailed, coarse-to-fine segmentation strategy that significantly enhances the precision of object identification and segmentation. Experiments validate that Reason3D achieves remarkable results on large-scale ScanNet and Matterport3D datasets for 3D express referring, 3D question answering, and 3D reasoning segmentation tasks. Code and models are available at: https://github.com/KuanchihHuang/Reason3D.
Authors: Yuanhui Huang, Wenzhao Zheng, Yunpeng Zhang, Jie Zhou, Jiwen Lu
Abstract: 3D semantic occupancy prediction aims to obtain 3D fine-grained geometry and semantics of the surrounding scene and is an important task for the robustness of vision-centric autonomous driving. Most existing methods employ dense grids such as voxels as scene representations, which ignore the sparsity of occupancy and the diversity of object scales and thus lead to unbalanced allocation of resources. To address this, we propose an object-centric representation to describe 3D scenes with sparse 3D semantic Gaussians where each Gaussian represents a flexible region of interest and its semantic features. We aggregate information from images through the attention mechanism and iteratively refine the properties of 3D Gaussians including position, covariance, and semantics. We then propose an efficient Gaussian-to-voxel splatting method to generate 3D occupancy predictions, which only aggregates the neighboring Gaussians for a certain position. We conduct extensive experiments on the widely adopted nuScenes and KITTI-360 datasets. Experimental results demonstrate that GaussianFormer achieves comparable performance with state-of-the-art methods with only 17.8% - 24.8% of their memory consumption. Code is available at: https://github.com/huang-yh/GaussianFormer.
Authors: Mu Cai, Jianwei Yang, Jianfeng Gao, Yong Jae Lee
Abstract: Large Multimodal Models (LMMs) such as LLaVA have shown strong performance in visual-linguistic reasoning. These models first embed images into a fixed large number of visual tokens and then feed them into a Large Language Model (LLM). However, this design causes an excessive number of tokens for dense visual scenarios such as high-resolution images and videos, leading to great inefficiency. While token pruning/merging methods do exist, they produce a single length output for each image and do not afford flexibility in trading off information density v.s. efficiency. Inspired by the concept of Matryoshka Dolls, we propose M3: Matryoshka Multimodal Models, which learns to represent visual content as nested sets of visual tokens that capture information across multiple coarse-to-fine granularities. Our approach offers several unique benefits for LMMs: (1) One can explicitly control the visual granularity per test instance during inference, e.g. , adjusting the number of tokens used to represent an image based on the anticipated complexity or simplicity of the content; (2) M3 provides a framework for analyzing the granularity needed for existing datasets, where we find that COCO-style benchmarks only need around ~9 visual tokens to obtain accuracy similar to that of using all 576 tokens; (3) Our approach provides a foundation to explore the best trade-off between performance and visual token length at sample level, where our investigation reveals that a large gap exists between the oracle upper bound and current fixed-scale representations.
Authors: Szymon Mazurek, Monika Pytlarz, Sylwia Malec, Alessandro Crimi
Abstract: Artificial intelligence have contributed to advancements across various industries. However, the rapid growth of artificial intelligence technologies also raises concerns about their environmental impact, due to associated carbon footprints to train computational models. Fetal brain segmentation in medical imaging is challenging due to the small size of the fetal brain and the limited image quality of fast 2D sequences. Deep neural networks are a promising method to overcome this challenge. In this context, the construction of larger models requires extensive data and computing power, leading to high energy consumption. Our study aims to explore model architectures and compression techniques that promote energy efficiency by optimizing the trade-off between accuracy and energy consumption through various strategies such as lightweight network design, architecture search, and optimized distributed training tools. We have identified several effective strategies including optimization of data loading, modern optimizers, distributed training strategy implementation, and reduced floating point operations precision usage with light model architectures while tuning parameters according to available computer resources. Our findings demonstrate that these methods lead to satisfactory model performance with low energy consumption during deep neural network training for medical image segmentation.
Authors: Ngoc-Du Tran, Thi-Thao Tran, Quang-Huy Nguyen, Manh-Hung Vu, Van-Truong Pham
Abstract: The emergence of deep learning techniques has advanced the image segmentation task, especially for medical images. Many neural network models have been introduced in the last decade bringing the automated segmentation accuracy close to manual segmentation. However, cutting-edge models like Transformer-based architectures rely on large scale annotated training data, and are generally designed with densely consecutive layers in the encoder, decoder, and skip connections resulting in large number of parameters. Additionally, for better performance, they often be pretrained on a larger data, thus requiring large memory size and increasing resource expenses. In this study, we propose a new lightweight but efficient model, namely LiteNeXt, based on convolutions and mixing modules with simplified decoder, for medical image segmentation. The model is trained from scratch with small amount of parameters (0.71M) and Giga Floating Point Operations Per Second (0.42). To handle boundary fuzzy as well as occlusion or clutter in objects especially in medical image regions, we propose the Marginal Weight Loss that can help effectively determine the marginal boundary between object and background. Furthermore, we propose the Self-embedding Representation Parallel technique, that can help augment the data in a self-learning manner. Experiments on public datasets including Data Science Bowls, GlaS, ISIC2018, PH2, and Sunnybrook data show promising results compared to other state-of-the-art CNN-based and Transformer-based architectures. Our code will be published at: https://github.com/tranngocduvnvp/LiteNeXt.
Authors: Chunyu Yuan, Dongfang Zhao, Sos S. Agaian
Abstract: Skin lesion segmentation is key for early skin cancer detection. Challenges in automatic segmentation from dermoscopic images include variations in color, texture, and artifacts of indistinct lesion boundaries. Deep learning methods like CNNs and U-Net have shown promise in addressing these issues. To further aid early diagnosis, especially on mobile devices with limited computing power, we present MUCM-Net. This efficient model combines Mamba State-Space Models with our UCM-Net architecture for improved feature learning and segmentation. MUCM-Net's Mamba-UCM Layer is optimized for mobile deployment, offering high accuracy with low computational needs. Tested on ISIC datasets, it outperforms other methods in accuracy and computational efficiency, making it a scalable tool for early detection in settings with limited resources. Our MUCM-Net source code is available for research and collaboration, supporting advances in mobile health diagnostics and the fight against skin cancer. In order to facilitate accessibility and further research in the field, the MUCM-Net source code is https://github.com/chunyuyuan/MUCM-Net
Authors: Jonathan Peck, Bart Goossens
Abstract: Deep neural networks are vulnerable to so-called adversarial examples: inputs which are intentionally constructed to cause the model to make incorrect predictions or classifications. Adversarial examples are often visually indistinguishable from natural data samples, making them hard to detect. As such, they pose significant threats to the reliability of deep learning systems. In this work, we study an adversarial defense based on the robust width property (RWP), which was recently introduced for compressed sensing. We show that a specific input purification scheme based on the RWP gives theoretical robustness guarantees for images that are approximately sparse. The defense is easy to implement and can be applied to any existing model without additional training or finetuning. We empirically validate the defense on ImageNet against $L^\infty$ perturbations at perturbation budgets ranging from $4/255$ to $32/255$. In the black-box setting, our method significantly outperforms the state-of-the-art, especially for large perturbations. In the white-box setting, depending on the choice of base classifier, we closely match the state of the art in robust ImageNet classification while avoiding the need for additional data, larger models or expensive adversarial training routines. Our code is available at https://github.com/peck94/robust-width-defense.
Authors: Jieren Deng, Hanbin Hong, Aaron Palmer, Xin Zhou, Jinbo Bi, Kaleel Mahmood, Yuan Hong, Derek Aguiar
Abstract: Randomized smoothing has become a leading method for achieving certified robustness in deep classifiers against l_{p}-norm adversarial perturbations. Current approaches for achieving certified robustness, such as data augmentation with Gaussian noise and adversarial training, require expensive training procedures that tune large models for different Gaussian noise levels and thus cannot leverage high-performance pre-trained neural networks. In this work, we introduce a novel certifying adapters framework (CAF) that enables and enhances the certification of classifier adversarial robustness. Our approach makes few assumptions about the underlying training algorithm or feature extractor and is thus broadly applicable to different feature extractor architectures (e.g., convolutional neural networks or vision transformers) and smoothing algorithms. We show that CAF (a) enables certification in uncertified models pre-trained on clean datasets and (b) substantially improves the performance of certified classifiers via randomized smoothing and SmoothAdv at multiple radii in CIFAR-10 and ImageNet. We demonstrate that CAF achieves improved certified accuracies when compared to methods based on random or denoised smoothing, and that CAF is insensitive to certifying adapter hyperparameters. Finally, we show that an ensemble of adapters enables a single pre-trained feature extractor to defend against a range of noise perturbation scales.
Authors: Hongye Zeng, Ke Zou, Zhihao Chen, Rui Zheng, Huazhu Fu
Abstract: Source-Free Unsupervised Domain Adaptation (SFUDA) has recently become a focus in the medical image domain adaptation, as it only utilizes the source model and does not require annotated target data. However, current SFUDA approaches cannot tackle the complex segmentation task across different MRI sequences, such as the vestibular schwannoma segmentation. To address this problem, we proposed Reliable Source Approximation (RSA), which can generate source-like and structure-preserved images from the target domain for updating model parameters and adapting domain shifts. Specifically, RSA deploys a conditional diffusion model to generate multiple source-like images under the guidance of varying edges of one target image. An uncertainty estimation module is then introduced to predict and refine reliable pseudo labels of generated images, and the prediction consistency is developed to select the most reliable generations. Subsequently, all reliable generated images and their pseudo labels are utilized to update the model. Our RSA is validated on vestibular schwannoma segmentation across multi-modality MRI. The experimental results demonstrate that RSA consistently improves domain adaptation performance over other state-of-the-art SFUDA methods. Code is available at https://github.com/zenghy96/Reliable-Source-Approximation.
URLs: https://github.com/zenghy96/Reliable-Source-Approximation.
Authors: Shaokui Wei, Hongyuan Zha, Baoyuan Wu
Abstract: Data-poisoning backdoor attacks are serious security threats to machine learning models, where an adversary can manipulate the training dataset to inject backdoors into models. In this paper, we focus on in-training backdoor defense, aiming to train a clean model even when the dataset may be potentially poisoned. Unlike most existing methods that primarily detect and remove/unlearn suspicious samples to mitigate malicious backdoor attacks, we propose a novel defense approach called PDB (Proactive Defensive Backdoor). Specifically, PDB leverages the "home field" advantage of defenders by proactively injecting a defensive backdoor into the model during training. Taking advantage of controlling the training process, the defensive backdoor is designed to suppress the malicious backdoor effectively while remaining secret to attackers. In addition, we introduce a reversible mapping to determine the defensive target label. During inference, PDB embeds a defensive trigger in the inputs and reverses the model's prediction, suppressing malicious backdoor and ensuring the model's utility on the original task. Experimental results across various datasets and models demonstrate that our approach achieves state-of-the-art defense performance against a wide range of backdoor attacks.
Authors: Huanbai Liu, Fanlong Zhang, Yin Tan, Lian Huang, Yan Li, Guoheng Huang, Shenghong Luo, An Zeng
Abstract: In recent years, deep learning has led to significant advances in bearing fault diagnosis (FD). Most techniques aim to achieve greater accuracy. However, they are sensitive to noise and lack robustness, resulting in insufficient domain adaptation and anti-noise ability. The comparison of studies reveals that giving equal attention to all features does not differentiate their significance. In this work, we propose a novel FD model by integrating multi-scale quaternion convolutional neural network (MQCNN), bidirectional gated recurrent unit (BiGRU), and cross self-attention feature fusion (CSAFF). We have developed innovative designs in two modules, namely MQCNN and CSAFF. Firstly, MQCNN applies quaternion convolution to multi-scale architecture for the first time, aiming to extract the rich hidden features of the original signal from multiple scales. Then, the extracted multi-scale information is input into CSAFF for feature fusion, where CSAFF innovatively incorporates cross self-attention mechanism to enhance discriminative interaction representation within features. Finally, BiGRU captures temporal dependencies while a softmax layer is employed for fault classification, achieving accurate FD. To assess the efficacy of our approach, we experiment on three public datasets (CWRU, MFPT, and Ottawa) and compare it with other excellent methods. The results confirm its state-of-the-art, which the average accuracies can achieve up to 99.99%, 100%, and 99.21% on CWRU, MFPT, and Ottawa datasets. Moreover, we perform practical tests and ablation experiments to validate the efficacy and robustness of the proposed approach. Code is available at https://github.com/mubai011/MQCCAF.
Authors: Yuzhuo Chen, Zetong Chen, Yuanyuan Liu
Abstract: Abnormalities in retinal fundus images may indicate certain pathologies such as diabetic retinopathy, hypertension, stroke, glaucoma, retinal macular edema, venous occlusion, and atherosclerosis, making the study and analysis of retinal images of great significance. In conventional medicine, the diagnosis of retina-related diseases relies on a physician's subjective assessment of the retinal fundus images, which is a time-consuming process and the accuracy is highly dependent on the physician's subjective experience. To this end, this paper proposes a fast, objective, and accurate method for the diagnosis of diseases related to retinal fundus images. This method is a multiclassification study of normal samples and 13 categories of disease samples on the STARE database, with a test set accuracy of 99.96%. Compared with other studies, our method achieved the highest accuracy. This study innovatively propose Segmentation-based Vascular Enhancement(SVE). After comparing the classification performances of the deep learning models of SVE images, original images and Smooth Grad-CAM ++ images, we extracted the deep learning features and traditional features of the SVE images and input them into nine meta learners for classification. The results shows that our proposed UNet-SVE-VGG-MLP model has the optimal performance for classifying diseases related to retinal fundus images on the STARE database, with a overall accuracy of 99.96% and a weighted AUC of 99.98% for the 14 categories on test dataset. This method can be used to realize rapid, objective, and accurate classification and diagnosis of retinal fundus image related diseases.
Authors: Junlin Song, Yuzhuo Chen, Yuan Yao, Zetong Chen, Renhao Guo, Lida Yang, Xinyi Sui, Qihang Wang, Xijiao Li, Aihua Cao, Wei Li
Abstract: Autism Spectrum Disorder is a condition characterized by a typical brain development leading to impairments in social skills, communication abilities, repetitive behaviors, and sensory processing. There have been many studies combining brain MRI images with machine learning algorithms to achieve objective diagnosis of autism, but the correlation between white matter and autism has not been fully utilized. To address this gap, we develop a computer-aided diagnostic model focusing on white matter regions in brain MRI by employing radiomics and machine learning methods. This study introduced a MultiUNet model for segmenting white matter, leveraging the UNet architecture and utilizing manually segmented MRI images as the training data. Subsequently, we extracted white matter features using the Pyradiomics toolkit and applied different machine learning models such as Support Vector Machine, Random Forest, Logistic Regression, and K-Nearest Neighbors to predict autism. The prediction sets all exceeded 80% accuracy. Additionally, we employed Convolutional Neural Network to analyze segmented white matter images, achieving a prediction accuracy of 86.84%. Notably, Support Vector Machine demonstrated the highest prediction accuracy at 89.47%. These findings not only underscore the efficacy of the models but also establish a link between white matter abnormalities and autism. Our study contributes to a comprehensive evaluation of various diagnostic models for autism and introduces a computer-aided diagnostic algorithm for early and objective autism diagnosis based on MRI white matter regions.
Authors: Brendan Park, Madeline Janecek, Naser Ezzati-Jivan, Yifeng Li, Ali Emami
Abstract: Large Language Models (LLMs) have demonstrated remarkable success in tasks like the Winograd Schema Challenge (WSC), showcasing advanced textual common-sense reasoning. However, applying this reasoning to multimodal domains, where understanding text and images together is essential, remains a substantial challenge. To address this, we introduce WinoVis, a novel dataset specifically designed to probe text-to-image models on pronoun disambiguation within multimodal contexts. Utilizing GPT-4 for prompt generation and Diffusion Attentive Attribution Maps (DAAM) for heatmap analysis, we propose a novel evaluation framework that isolates the models' ability in pronoun disambiguation from other visual processing challenges. Evaluation of successive model versions reveals that, despite incremental advancements, Stable Diffusion 2.0 achieves a precision of 56.7% on WinoVis, only marginally surpassing random guessing. Further error analysis identifies important areas for future research aimed at advancing text-to-image models in their ability to interpret and interact with the complex visual world.
Authors: Romario Gualdr\'on-Hurtado, Roman Jacome, Sergio Urrea, Henry Arguello, Luis Gonzalez
Abstract: Deep-learning (DL)-based image deconvolution (ID) has exhibited remarkable recovery performance, surpassing traditional linear methods. However, unlike traditional ID approaches that rely on analytical properties of the point spread function (PSF) to achieve high recovery performance - such as specific spectrum properties or small conditional numbers in the convolution matrix - DL techniques lack quantifiable metrics for evaluating PSF suitability for DL-assisted recovery. Aiming to enhance deconvolution quality, we propose a metric that employs a non-linear approach to learn the invertibility of an arbitrary PSF using a neural network by mapping it to a unit impulse. A lower discrepancy between the mapped PSF and a unit impulse indicates a higher likelihood of successful inversion by a DL network. Our findings reveal that this metric correlates with high recovery performance in DL and traditional methods, thereby serving as an effective regularizer in deconvolution tasks. This approach reduces the computational complexity over conventional condition number assessments and is a differentiable process. These useful properties allow its application in designing diffractive optical elements through end-to-end (E2E) optimization, achieving invertible PSFs, and outperforming the E2E baseline framework.
Authors: Zechun Liu, Changsheng Zhao, Igor Fedorov, Bilge Soran, Dhruv Choudhary, Raghuraman Krishnamoorthi, Vikas Chandra, Yuandong Tian, Tijmen Blankevoort
Abstract: Post-training quantization (PTQ) techniques applied to weights, activations, and the KV cache greatly reduce memory usage, latency, and power consumption of Large Language Models (LLMs), but may lead to large quantization errors when outliers are present. Recent findings suggest that rotating activation or weight matrices helps remove outliers and benefits quantization. In this work, we identify a collection of applicable rotation parameterizations that lead to identical outputs in full-precision Transformer architectures, and find that some random rotations lead to much better quantization than others, with an up to 13 points difference in downstream zero-shot reasoning performance. As a result, we propose SpinQuant that optimizes (or learns) the rotation matrices with Cayley optimization on a small validation set. With 4-bit quantization of weight, activation, and KV-cache, SpinQuant narrows the accuracy gap on zero-shot reasoning tasks with full precision to merely 2.9 points on the LLaMA-2 7B model, surpassing LLM-QAT by 19.1 points and SmoothQuant by 25.0 points. SpinQuant also outperforms concurrent work QuaRot, which applies random rotations to remove outliers. In particular, for LLaMA-2 7B/LLaMA-3 8B models that are hard to quantize, SpinQuant reduces the gap to full precision by 30.2%/34.1% relative to QuaRot.
Authors: Jiuxiang Gu, Yingyu Liang, Zhenmei Shi, Zhao Song, Yufa Zhou
Abstract: Diffusion models have made rapid progress in generating high-quality samples across various domains. However, a theoretical understanding of the Lipschitz continuity and second momentum properties of the diffusion process is still lacking. In this paper, we bridge this gap by providing a detailed examination of these smoothness properties for the case where the target data distribution is a mixture of Gaussians, which serves as a universal approximator for smooth densities such as image data. We prove that if the target distribution is a $k$-mixture of Gaussians, the density of the entire diffusion process will also be a $k$-mixture of Gaussians. We then derive tight upper bounds on the Lipschitz constant and second momentum that are independent of the number of mixture components $k$. Finally, we apply our analysis to various diffusion solvers, both SDE and ODE based, to establish concrete error guarantees in terms of the total variation distance and KL divergence between the target and learned distributions. Our results provide deeper theoretical insights into the dynamics of the diffusion process under common data distributions.
Authors: Hongwei Bran Li, Cheng Ouyang, Tamaz Amiranashvili, Matthew S. Rosen, Bjoern Menze, Juan Eugenio Iglesias
Abstract: Self-supervised contrastive learning has predominantly adopted deterministic methods, which are not suited for environments characterized by uncertainty and noise. This paper introduces a new perspective on incorporating uncertainty into contrastive learning by embedding representations within a spherical space, inspired by the von Mises-Fisher distribution (vMF). We introduce an unnormalized form of vMF and leverage the concentration parameter, kappa, as a direct, interpretable measure to quantify uncertainty explicitly. This approach not only provides a probabilistic interpretation of the embedding space but also offers a method to calibrate model confidence against varying levels of data corruption and characteristics. Our empirical results demonstrate that the estimated concentration parameter correlates strongly with the degree of unforeseen data corruption encountered at test time, enables failure analysis, and enhances existing out-of-distribution detection methods.
Authors: Tianchen Deng, Yi Zhou, Wenhua Wu, Mingrui Li, Jingwei Huang, Shuhong Liu, Yanzeng Song, Hao Zuo, Yanbo Wang, Yutao Yue, Hesheng Wang, Weidong Chen
Abstract: This technical report presents the 1st winning model for UG2+, a task in CVPR 2024 UAV Tracking and Pose-Estimation Challenge. This challenge faces difficulties in drone detection, UAV-type classification and 2D/3D trajectory estimation in extreme weather conditions with multi-modal sensor information, including stereo vision, various Lidars, Radars, and audio arrays. Leveraging this information, we propose a multi-modal UAV detection, classification, and 3D tracking method for accurate UAV classification and tracking. A novel classification pipeline which incorporates sequence fusion, region of interest (ROI) cropping, and keyframe selection is proposed. Our system integrates cutting-edge classification techniques and sophisticated post-processing steps to boost accuracy and robustness. The designed pose estimation pipeline incorporates three modules: dynamic points analysis, a multi-object tracker, and trajectory completion techniques. Extensive experiments have validated the effectiveness and precision of our approach. In addition, we also propose a novel dataset pre-processing method and conduct a comprehensive ablation study for our design. We finally achieved the best performance in the classification and tracking of the MMUAD dataset. The code and configuration of our method are available at https://github.com/dtc111111/Multi-Modal-UAV.
Authors: Regev Cohen, Idan Kligvasser, Ehud Rivlin, Daniel Freedman
Abstract: The pursuit of high perceptual quality in image restoration has driven the development of revolutionary generative models, capable of producing results often visually indistinguishable from real data. However, as their perceptual quality continues to improve, these models also exhibit a growing tendency to generate hallucinations - realistic-looking details that do not exist in the ground truth images. The presence of hallucinations introduces uncertainty regarding the reliability of the models' predictions, raising major concerns about their practical application. In this paper, we employ information-theory tools to investigate this phenomenon, revealing a fundamental tradeoff between uncertainty and perception. We rigorously analyze the relationship between these two factors, proving that the global minimal uncertainty in generative models grows in tandem with perception. In particular, we define the inherent uncertainty of the restoration problem and show that attaining perfect perceptual quality entails at least twice this uncertainty. Additionally, we establish a relation between mean squared-error distortion, uncertainty and perception, through which we prove the aforementioned uncertainly-perception tradeoff induces the well-known perception-distortion tradeoff. This work uncovers fundamental limitations of generative models in achieving both high perceptual quality and reliable predictions for image restoration. We demonstrate our theoretical findings through an analysis of single image super-resolution algorithms. Our work aims to raise awareness among practitioners about this inherent tradeoff, empowering them to make informed decisions and potentially prioritize safety over perceptual performance.
Authors: Kun Huang, Xiao Ma, Yuhan Zhang, Na Su, Songtao Yuan, Yong Liu, Qiang Chen, Huazhu Fu
Abstract: Optical coherence tomography (OCT) image analysis plays an important role in the field of ophthalmology. Current successful analysis models rely on available large datasets, which can be challenging to be obtained for certain tasks. The use of deep generative models to create realistic data emerges as a promising approach. However, due to limitations in hardware resources, it is still difficulty to synthesize high-resolution OCT volumes. In this paper, we introduce a cascaded amortized latent diffusion model (CA-LDM) that can synthesis high-resolution OCT volumes in a memory-efficient way. First, we propose non-holistic autoencoders to efficiently build a bidirectional mapping between high-resolution volume space and low-resolution latent space. In tandem with autoencoders, we propose cascaded diffusion processes to synthesize high-resolution OCT volumes with a global-to-local refinement process, amortizing the memory and computational demands. Experiments on a public high-resolution OCT dataset show that our synthetic data have realistic high-resolution and global features, surpassing the capabilities of existing methods. Moreover, performance gains on two down-stream fine-grained segmentation tasks demonstrate the benefit of the proposed method in training deep learning models for medical imaging tasks. The code is public available at: https://github.com/nicetomeetu21/CA-LDM.
Authors: Koya Sakamoto, Daichi Azuma, Taiki Miyanishi, Shuhei Kurita, Motoaki Kawanabe
Abstract: Building robots capable of interacting with humans through natural language in the visual world presents a significant challenge in the field of robotics. To overcome this challenge, Embodied Question Answering (EQA) has been proposed as a benchmark task to measure the ability to identify an object navigating through a previously unseen environment in response to human-posed questions. Although some methods have been proposed, their evaluations have been limited to simulations, without experiments in real-world scenarios. Furthermore, all of these methods are constrained by a limited vocabulary for question-and-answer interactions, making them unsuitable for practical applications. In this work, we propose a map-based modular EQA method that enables real robots to navigate unknown environments through frontier-based map creation and address unknown QA pairs using foundation models that support open vocabulary. Unlike the questions of the previous EQA dataset on Matterport 3D (MP3D), questions in our real-world experiments contain various question formats and vocabularies not included in the training data. We conduct comprehensive experiments on virtual environments (MP3D-EQA) and two real-world house environments and demonstrate that our method can perform EQA even in the real world.
Authors: Tianyi Bai, Hao Liang, Binwang Wan, Ling Yang, Bozhou Li, Yifan Wang, Bin Cui, Conghui He, Binhang Yuan, Wentao Zhang
Abstract: Human beings perceive the world through diverse senses such as sight, smell, hearing, and touch. Similarly, multimodal large language models (MLLMs) enhance the capabilities of traditional large language models by integrating and processing data from multiple modalities including text, vision, audio, video, and 3D environments. Data plays a pivotal role in the development and refinement of these models. In this survey, we comprehensively review the literature on MLLMs from a data-centric perspective. Specifically, we explore methods for preparing multimodal data during the pretraining and adaptation phases of MLLMs. Additionally, we analyze the evaluation methods for datasets and review benchmarks for evaluating MLLMs. Our survey also outlines potential future research directions. This work aims to provide researchers with a detailed understanding of the data-driven aspects of MLLMs, fostering further exploration and innovation in this field.
Authors: Manish Saini, Melvin Paul Jacob, Minh Nguyen, Nico Hochgeschwender
Abstract: When performing manipulation-based activities such as picking objects, a mobile robot needs to position its base at a location that supports successful execution. To address this problem, prominent approaches typically rely on costly grasp planners to provide grasp poses for a target object, which are then are then analysed to identify the best robot placements for achieving each grasp pose. In this paper, we propose instead to first find robot placements that would not result in collision with the environment and from where picking up the object is feasible, then evaluate them to find the best placement candidate. Our approach takes into account the robot's reachability, as well as RGB-D images and occupancy grid maps of the environment for identifying suitable robot poses. The proposed algorithm is embedded in a service robotic workflow, in which a person points to select the target object for grasping. We evaluate our approach with a series of grasping experiments, against an existing baseline implementation that sends the robot to a fixed navigation goal. The experimental results show how the approach allows the robot to grasp the target object from locations that are very challenging to the baseline implementation.
Authors: Hengkang Wang, Xu Zhang, Taihui Li, Yuxiang Wan, Tiancong Chen, Ju Sun
Abstract: Pretrained diffusion models (DMs) have recently been popularly used in solving inverse problems (IPs). The existing methods mostly interleave iterative steps in the reverse diffusion process and iterative steps to bring the iterates closer to satisfying the measurement constraint. However, such interleaving methods struggle to produce final results that look like natural objects of interest (i.e., manifold feasibility) and fit the measurement (i.e., measurement feasibility), especially for nonlinear IPs. Moreover, their capabilities to deal with noisy IPs with unknown types and levels of measurement noise are unknown. In this paper, we advocate viewing the reverse process in DMs as a function and propose a novel plug-in method for solving IPs using pretrained DMs, dubbed DMPlug. DMPlug addresses the issues of manifold feasibility and measurement feasibility in a principled manner, and also shows great potential for being robust to unknown types and levels of noise. Through extensive experiments across various IP tasks, including two linear and three nonlinear IPs, we demonstrate that DMPlug consistently outperforms state-of-the-art methods, often by large margins especially for nonlinear IPs. The code is available at https://github.com/sun-umn/DMPlug.
Authors: SeungWon Seo, Junhyeok Lee, SeongRae Noh, HyeongYeop Kang
Abstract: We address the challenge of multi-agent cooperation, where agents achieve a common goal by interacting with a 3D scene and cooperating with decentralized agents under complex partial observations. This involves managing communication costs and optimizing interaction trajectories in dynamic environments. Our research focuses on three primary limitations of existing cooperative agent systems. Firstly, current systems demonstrate inefficiency in managing acquired information through observation, resulting in declining planning performance as the environment becomes more complex with additional objects or goals. Secondly, the neglect of false plans in partially observable settings leads to suboptimal cooperative performance, as agents struggle to adapt to environmental changes influenced by the unseen actions of other agents. Lastly, the failure to incorporate spatial data into decision-making processes restricts the agent's ability to construct optimized trajectories. To overcome these limitations, we propose the RElevance and Validation-Enhanced Cooperative Language Agent (REVECA), a novel cognitive architecture powered by GPT-3.5. REVECA leverages relevance assessment, plan validation, and spatial information to enhance the efficiency and robustness of agent cooperation in dynamic and partially observable environments while minimizing continuous communication costs and effectively managing irrelevant dummy objects. Our extensive experiments demonstrate the superiority of REVECA over previous approaches, including those driven by GPT-4.0. Additionally, a user study highlights REVECA's potential for achieving trustworthy human-AI cooperation. We expect that REVECA will have significant applications in gaming, XR applications, educational tools, and humanoid robots, contributing to substantial economic, commercial, and academic advancements.
Authors: Runzhao Yang, Yinda Chen, Zhihong Zhang, Xiaoyu Liu, Zongren Li, Kunlun He, Zhiwei Xiong, Jinli Suo, Qionghai Dai
Abstract: In the field of medical image compression, Implicit Neural Representation (INR) networks have shown remarkable versatility due to their flexible compression ratios, yet they are constrained by a one-to-one fitting approach that results in lengthy encoding times. Our novel method, ``\textbf{UniCompress}'', innovatively extends the compression capabilities of INR by being the first to compress multiple medical data blocks using a single INR network. By employing wavelet transforms and quantization, we introduce a codebook containing frequency domain information as a prior input to the INR network. This enhances the representational power of INR and provides distinctive conditioning for different image blocks. Furthermore, our research introduces a new technique for the knowledge distillation of implicit representations, simplifying complex model knowledge into more manageable formats to improve compression ratios. Extensive testing on CT and electron microscopy (EM) datasets has demonstrated that UniCompress outperforms traditional INR methods and commercial compression solutions like HEVC, especially in complex and high compression scenarios. Notably, compared to existing INR techniques, UniCompress achieves a 4$\sim$5 times increase in compression speed, marking a significant advancement in the field of medical image compression. Codes will be publicly available.
Authors: Anran Liu, Cheng Lin, Yuan Liu, Xiaoxiao Long, Zhiyang Dou, Hao-Xiang Guo, Ping Luo, Wenping Wang
Abstract: Recently, the emergence of diffusion models has opened up new opportunities for single-view reconstruction. However, all the existing methods represent the target object as a closed mesh devoid of any structural information, thus neglecting the part-based structure, which is crucial for many downstream applications, of the reconstructed shape. Moreover, the generated meshes usually suffer from large noises, unsmooth surfaces, and blurry textures, making it challenging to obtain satisfactory part segments using 3D segmentation techniques. In this paper, we present Part123, a novel framework for part-aware 3D reconstruction from a single-view image. We first use diffusion models to generate multiview-consistent images from a given image, and then leverage Segment Anything Model (SAM), which demonstrates powerful generalization ability on arbitrary objects, to generate multiview segmentation masks. To effectively incorporate 2D part-based information into 3D reconstruction and handle inconsistency, we introduce contrastive learning into a neural rendering framework to learn a part-aware feature space based on the multiview segmentation masks. A clustering-based algorithm is also developed to automatically derive 3D part segmentation results from the reconstructed models. Experiments show that our method can generate 3D models with high-quality segmented parts on various objects. Compared to existing unstructured reconstruction methods, the part-aware 3D models from our method benefit some important applications, including feature-preserving reconstruction, primitive fitting, and 3D shape editing.
Authors: Richard Elvira, Juan D. Tard\'os, Jos\'e M. M. Montiel
Abstract: Monocular visual simultaneous localization and mapping (V-SLAM) is nowadays an irreplaceable tool in mobile robotics and augmented reality, where it performs robustly. However, human colonoscopies pose formidable challenges like occlusions, blur, light changes, lack of texture, deformation, water jets or tool interaction, which result in very frequent tracking losses. ORB-SLAM3, the top performing multiple-map V-SLAM, is unable to recover from them by merging sub-maps or relocalizing the camera, due to the poor performance of its place recognition algorithm based on ORB features and DBoW2 bag-of-words. We present CudaSIFT-SLAM, the first V-SLAM system able to process complete human colonoscopies in real-time. To overcome the limitations of ORB-SLAM3, we use SIFT instead of ORB features and replace the DBoW2 direct index with the more computationally demanding brute-force matching, being able to successfully match images separated in time for relocation and map merging. Real-time performance is achieved thanks to CudaSIFT, a GPU implementation for SIFT extraction and brute-force matching. We benchmark our system in the C3VD phantom colon dataset, and in a full real colonoscopy from the Endomapper dataset, demonstrating the capabilities to merge sub-maps and relocate in them, obtaining significantly longer sub-maps. Our system successfully maps in real-time 88 % of the frames in the C3VD dataset. In a real screening colonoscopy, despite the much higher prevalence of occluded and blurred frames, the mapping coverage is 53 % in carefully explored areas and 38 % in the full sequence, a 70 % improvement over ORB-SLAM3.
Authors: Yitong Li, Igor Yakushev, Dennis M. Hedderich, Christian Wachinger
Abstract: Positron emission tomography (PET) is a well-established functional imaging technique for diagnosing brain disorders. However, PET's high costs and radiation exposure limit its widespread use. In contrast, magnetic resonance imaging (MRI) does not have these limitations. Although it also captures neurodegenerative changes, MRI is a less sensitive diagnostic tool than PET. To close this gap, we aim to generate synthetic PET from MRI. Herewith, we introduce PASTA, a novel pathology-aware image translation framework based on conditional diffusion models. Compared to the state-of-the-art methods, PASTA excels in preserving both structural and pathological details in the target modality, which is achieved through its highly interactive dual-arm architecture and multi-modal condition integration. A cycle exchange consistency and volumetric generation strategy elevate PASTA's capability to produce high-quality 3D PET scans. Our qualitative and quantitative results confirm that the synthesized PET scans from PASTA not only reach the best quantitative scores but also preserve the pathology correctly. For Alzheimer's classification, the performance of synthesized scans improves over MRI by 4%, almost reaching the performance of actual PET. Code is available at https://github.com/ai-med/PASTA.
Authors: Wen Hanlin
Abstract: In the Vision-and-Language Navigation (VLN) field, agents are tasked with navigating real-world scenes guided by linguistic instructions. Enabling the agent to adhere to instructions throughout the process of navigation represents a significant challenge within the domain of VLN. To address this challenge, common approaches often rely on encoders to explicitly record past locations and actions, increasing model complexity and resource consumption. Our proposal, the Vision-and-Language Navigation Generative Pretrained Transformer (VLN-GPT), adopts a transformer decoder model (GPT2) to model trajectory sequence dependencies, bypassing the need for historical encoding modules. This method allows for direct historical information access through trajectory sequence, enhancing efficiency. Furthermore, our model separates the training process into offline pre-training with imitation learning and online fine-tuning with reinforcement learning. This distinction allows for more focused training objectives and improved performance. Performance assessments on the VLN dataset reveal that VLN-GPT surpasses complex state-of-the-art encoder-based models.
Authors: James L. Gray, Aous T. Naman, David S. Taubman
Abstract: Variational approaches to disparity estimation typically use a linearised brightness constancy constraint, which only applies in smooth regions and over small distances. Accordingly, current variational approaches rely on a schedule to progressively include image data. This paper proposes the use of Gradient Consistency information to assess the validity of the linearisation; this information is used to determine the weights applied to the data term as part of an analytically inspired Gradient Consistency Model. The Gradient Consistency Model penalises the data term for view pairs that have a mismatch between the spatial gradients in the source view and the spatial gradients in the target view. Instead of relying on a tuned or learned schedule, the Gradient Consistency Model is self-scheduling, since the weights evolve as the algorithm progresses. We show that the Gradient Consistency Model outperforms standard coarse-to-fine schemes and the recently proposed progressive inclusion of views approach in both rate of convergence and accuracy.
Authors: Peratham Wiriyathammabhum
Abstract: This paper presents a mixture version of the method-of-moment unsupervised lexicon classification by an incorporation of a Dirichlet process.
Authors: Xiaohong Fan, Ke Chen, Huaming Yi, Yin Yang, Jianping Zhang
Abstract: X-ray Computed Tomography (CT) is one of the most important diagnostic imaging techniques in clinical applications. Sparse-view CT imaging reduces the number of projection views to a lower radiation dose and alleviates the potential risk of radiation exposure. Most existing deep learning (DL) and deep unfolding sparse-view CT reconstruction methods: 1) do not fully use the projection data; 2) do not always link their architecture designs to a mathematical theory; 3) do not flexibly deal with multi-sparse-view reconstruction assignments. This paper aims to use mathematical ideas and design optimal DL imaging algorithms for sparse-view tomography reconstructions. We propose a novel dual-domain deep unfolding unified framework that offers a great deal of flexibility for multi-sparse-view CT reconstruction with different sampling views through a single model. This framework combines the theoretical advantages of model-based methods with the superior reconstruction performance of DL-based methods, resulting in the expected generalizability of DL. We propose a refinement module that utilizes unfolding projection domain to refine full-sparse-view projection errors, as well as an image domain correction module that distills multi-scale geometric error corrections to reconstruct sparse-view CT. This provides us with a new way to explore the potential of projection information and a new perspective on designing network architectures. All parameters of our proposed framework are learnable end to end, and our method possesses the potential to be applied to plug-and-play reconstruction. Extensive experiments demonstrate that our framework is superior to other existing state-of-the-art methods. Our source codes are available at https://github.com/fanxiaohong/MVMS-RCN.
Authors: Wenhao Zhang, Bin Huang, Shuyue Chen, Xiaoling Xu, Weiwen Wu, Qiegen Liu
Abstract: Low-dose computed tomography (LDCT) plays a vital role in clinical applications by mitigating radiation risks. Nevertheless, reducing radiation doses significantly degrades image quality. Concurrently, common deep learning methods demand extensive data, posing concerns about privacy, cost, and time constraints. Consequently, we propose a few-shot low-dose CT reconstruction method using Partitioned Hankel-based Diffusion (PHD) models. During the prior learning stage, the projection data is first transformed into multiple partitioned Hankel matrices. Structured tensors are then extracted from these matrices to facilitate prior learning through multiple diffusion models. In the iterative reconstruction stage, an iterative stochastic differential equation solver is employed along with data consistency constraints to update the acquired projection data. Furthermore, penalized weighted least-squares and total variation techniques are introduced to enhance the resulting image quality. The results approximate those of normal-dose counterparts, validating PHD model as an effective and practical model for reducing artifacts and noise while preserving image quality.
Authors: Sheng Yang, Jacob A. Zavatone-Veth, Cengiz Pehlevan
Abstract: The vulnerability of neural network classifiers to adversarial attacks is a major obstacle to their deployment in safety-critical applications. Regularization of network parameters during training can be used to improve adversarial robustness and generalization performance. Usually, the network is regularized end-to-end, with parameters at all layers affected by regularization. However, in settings where learning representations is key, such as self-supervised learning (SSL), layers after the feature representation will be discarded when performing inference. For these models, regularizing up to the feature space is more suitable. To this end, we propose a new spectral regularizer for representation learning that encourages black-box adversarial robustness in downstream classification tasks. In supervised classification settings, we show empirically that this method is more effective in boosting test accuracy and robustness than previously-proposed methods that regularize all layers of the network. We then show that this method improves the adversarial robustness of classifiers using representations learned with self-supervised training or transferred from another classification task. In all, our work begins to unveil how representational structure affects adversarial robustness.
Authors: Franco Coltraro, Jaume Amor\'os, Maria Alberich-Carrami\~nana, Carme Torras
Abstract: In this work, we study the perception problem for garments using tools from computational topology: the identification of their geometry and position in space from point-cloud samples, as obtained e.g. with 3D scanners. We present a reconstruction algorithm based on a direct topological study of the sampled textile surface that allows us to obtain a cellular decomposition of it via a Morse function. No intermediate triangulation or local implicit equations are used, avoiding reconstruction-induced artifices. No a priori knowledge of the surface topology, density or regularity of the point-sample is required to run the algorithm. The results are a piecewise decomposition of the surface as a union of Morse cells (i.e. topological disks), suitable for tasks such as noise-filtering or mesh-independent reparametrization, and a cell complex of small rank determining the surface topology. This algorithm can be applied to smooth surfaces with or without boundary, embedded in an ambient space of any dimension.
Authors: Yoeri Poels, Koen Minartz, Harshit Bansal, Vlado Menkovski
Abstract: Simulation is a powerful tool to better understand physical systems, but generally requires computationally expensive numerical methods. Downstream applications of such simulations can become computationally infeasible if they require many forward solves, for example in the case of inverse design with many degrees of freedom. In this work, we investigate and extend neural PDE solvers as a tool to aid in scaling simulations for two-phase flow problems, and simulations of oil expulsion from a pore specifically. We extend existing numerical methods for this problem to a more complex setting involving varying geometries of the domain to generate a challenging dataset. Further, we investigate three prominent neural PDE solver methods, namely the UNet, DRN and U-FNO, and extend them for characteristics of the oil-expulsion problem: (1) spatial conditioning on the geometry; (2) periodicity in the boundary; (3) approximate mass conservation. We scale all methods and benchmark their speed-accuracy trade-off, evaluate qualitative properties, and perform an ablation study. We find that the investigated methods can accurately model the droplet dynamics with up to three orders of magnitude speed-up, that our extensions improve performance over the baselines, and that the introduced varying geometries constitute a significantly more challenging setting over the previously considered oil expulsion problem.
Authors: Denis Kuznedelev, Valerii Startsev, Daniil Shlenskii, Sergey Kastryulin
Abstract: There is a prevalent opinion in the recent literature that Diffusion-based models outperform GAN-based counterparts on the Image Super Resolution (ISR) problem. However, in most studies, Diffusion-based ISR models were trained longer and utilized larger networks than the GAN baselines. This raises the question of whether the superiority of Diffusion models is due to the Diffusion paradigm being better suited for the ISR task or if it is a consequence of the increased scale and computational resources used in contemporary studies. In our work, we compare Diffusion-based and GAN-based Super Resolution under controlled settings, where both approaches are matched in terms of architecture, model and dataset size, and computational budget. We show that a GAN-based model can achieve results comparable to a Diffusion-based model. Additionally, we explore the impact of design choices such as text conditioning and augmentation on the performance of ISR models, showcasing their effect on several downstream tasks. We will release the inference code and weights of our scaled GAN.
Authors: Yuting Ma, Lechao Cheng, Yaxiong Wang, Zhun Zhong, Xiaohua Xu, Meng Wang
Abstract: Federated learning (FL) is a popular privacy-preserving paradigm that enables distributed clients to collaboratively train models with a central server while keeping raw data locally. In practice, distinct model architectures, varying data distributions, and limited resources across local clients inevitably cause model performance degradation and a slowdown in convergence speed. However, existing FL methods can only solve some of the above heterogeneous challenges and have obvious performance limitations. Notably, a unified framework has not yet been explored to overcome these challenges. Accordingly, we propose FedHPL, a parameter-efficient unified $\textbf{Fed}$erated learning framework for $\textbf{H}$eterogeneous settings based on $\textbf{P}$rompt tuning and $\textbf{L}$ogit distillation. Specifically, we employ a local prompt tuning scheme that leverages a few learnable visual prompts to efficiently fine-tune the frozen pre-trained foundation model for downstream tasks, thereby accelerating training and improving model performance under limited local resources and data heterogeneity. Moreover, we design a global logit distillation scheme to handle the model heterogeneity and guide the local training. In detail, we leverage logits to implicitly capture local knowledge and design a weighted knowledge aggregation mechanism to generate global client-specific logits. We provide a theoretical guarantee on the generalization error bound for FedHPL. The experiments on various benchmark datasets under diverse settings of models and data demonstrate that our framework outperforms state-of-the-art FL approaches, with less computation overhead and training rounds.
Authors: Shaoan Wang, Zhanhua Xin, Yaoqing Hu, Dongyue Li, Mingzhu Zhu, Junzhi Yu
Abstract: Event camera, a bio-inspired asynchronous triggered camera, offers promising prospects for fusion with frame-based cameras owing to its low latency and high dynamic range. However, calibrating stereo vision systems that incorporate both event and frame-based cameras remains a significant challenge. In this letter, we present EF-Calib, a spatiotemporal calibration framework for event- and frame-based cameras using continuous-time trajectories. A novel calibration pattern applicable to both camera types and the corresponding event recognition algorithm is proposed. Leveraging the asynchronous nature of events, a derivable piece-wise B-spline to represent camera pose continuously is introduced, enabling calibration for intrinsic parameters, extrinsic parameters, and time offset, with analytical Jacobians provided. Various experiments are carried out to evaluate the calibration performance of EF-Calib, including calibration experiments for intrinsic parameters, extrinsic parameters, and time offset. Experimental results show that EF-Calib achieves the most accurate intrinsic parameters compared to current SOTA, the close accuracy of the extrinsic parameters compared to the frame-based results, and accurate time offset estimation. EF-Calib provides a convenient and accurate toolbox for calibrating the system that fuses events and frames. The code of this paper will also be open-sourced at: https://github.com/wsakobe/EF-Calib.
Authors: Litu Rout, Yujia Chen, Nataniel Ruiz, Abhishek Kumar, Constantine Caramanis, Sanjay Shakkottai, Wen-Sheng Chu
Abstract: We propose Reference-Based Modulation (RB-Modulation), a new plug-and-play solution for training-free personalization of diffusion models. Existing training-free approaches exhibit difficulties in (a) style extraction from reference images in the absence of additional style or content text descriptions, (b) unwanted content leakage from reference style images, and (c) effective composition of style and content. RB-Modulation is built on a novel stochastic optimal controller where a style descriptor encodes the desired attributes through a terminal cost. The resulting drift not only overcomes the difficulties above, but also ensures high fidelity to the reference style and adheres to the given text prompt. We also introduce a cross-attention-based feature aggregation scheme that allows RB-Modulation to decouple content and style from the reference image. With theoretical justification and empirical evidence, our framework demonstrates precise extraction and control of content and style in a training-free manner. Further, our method allows a seamless composition of content and style, which marks a departure from the dependency on external adapters or ControlNets.
Authors: Abdulaziz Almuzairee, Nicklas Hansen, Henrik I. Christensen
Abstract: $Q$-learning algorithms are appealing for real-world applications due to their data-efficiency, but they are very prone to overfitting and training instabilities when trained from visual observations. Prior work, namely SVEA, finds that selective application of data augmentation can improve the visual generalization of RL agents without destabilizing training. We revisit its recipe for data augmentation, and find an assumption that limits its effectiveness to augmentations of a photometric nature. Addressing these limitations, we propose a generalized recipe, SADA, that works with wider varieties of augmentations. We benchmark its effectiveness on DMC-GB2 -- our proposed extension of the popular DMControl Generalization Benchmark -- as well as tasks from Meta-World and the Distracting Control Suite, and find that our method, SADA, greatly improves training stability and generalization of RL agents across a diverse set of augmentations. Visualizations, code, and benchmark: see https://aalmuzairee.github.io/SADA/
Authors: Gal Lifshitz, Dan Raviv
Abstract: Steepest descent algorithms, which are commonly used in deep learning, use the gradient as the descent direction, either as-is or after a direction shift using preconditioning. In many scenarios calculating the gradient is numerically hard due to complex or non-differentiable cost functions, specifically next to singular points. In this work we focus on the derivation of the Total Variation semi-norm commonly used in unsupervised cost functions. Specifically, we derive a differentiable proxy to the hard L1 smoothness constraint in a novel iterative scheme which we refer to as Cost Unrolling. Producing more accurate gradients during training, our method enables finer predictions of a given DNN model through improved convergence, without modifying its architecture or increasing computational complexity. We demonstrate our method in the unsupervised optical flow task. Replacing the L1 smoothness constraint with our unrolled cost during the training of a well known baseline, we report improved results on both MPI Sintel and KITTI 2015 unsupervised optical flow benchmarks. Particularly, we report EPE reduced by up to 15.82% on occluded pixels, where the smoothness constraint is dominant, enabling the detection of much sharper motion edges.
Authors: Puneet Kumar, Sarthak Malik, Balasubramanian Raman, Xiaobai Li
Abstract: This paper proposes a multimodal emotion recognition system, VIsual Spoken Textual Additive Net (VISTANet), to classify emotions reflected by input containing image, speech, and text into discrete classes. A new interpretability technique, K-Average Additive exPlanation (KAAP), has been developed that identifies important visual, spoken, and textual features leading to predicting a particular emotion class. The VISTANet fuses information from image, speech, and text modalities using a hybrid of early and late fusion. It automatically adjusts the weights of their intermediate outputs while computing the weighted average. The KAAP technique computes the contribution of each modality and corresponding features toward predicting a particular emotion class. To mitigate the insufficiency of multimodal emotion datasets labeled with discrete emotion classes, we have constructed a large-scale IIT-R MMEmoRec dataset consisting of images, corresponding speech and text, and emotion labels ('angry,' 'happy,' 'hate,' and 'sad'). The VISTANet has resulted in 95.99% emotion recognition accuracy on the IIT-R MMEmoRec dataset using visual, audio, and textual modalities, outperforming when using any one or two modalities. The IIT-R MMEmoRec dataset can be accessed at https://github.com/MIntelligence-Group/MMEmoRec.
Authors: Mat\'ias Tailanian, \'Alvaro Pardo, Pablo Mus\'e
Abstract: In this work we propose a one-class self-supervised method for anomaly segmentation in images that benefits both from a modern machine learning approach and a more classic statistical detection theory. The method consists of four phases. First, features are extracted using a multi-scale image Transformer architecture. Then, these features are fed into a U-shaped Normalizing Flow (NF) that lays the theoretical foundations for the subsequent phases. The third phase computes a pixel-level anomaly map from the NF embedding, and the last phase performs a segmentation based on the a contrario framework. This multiple hypothesis testing strategy permits the derivation of robust unsupervised detection thresholds, which are crucial in real-world applications where an operational point is needed. The segmentation results are evaluated using the Mean Intersection over Union (mIoU) metric, and for assessing the generated anomaly maps we report the area under the Receiver Operating Characteristic curve (AUROC), as well as the Area Under the Per-Region-Overlap curve (AUPRO). Extensive experimentation in various datasets shows that the proposed approach produces state-of-the-art results for all metrics and all datasets, ranking first in most MVTec-AD categories, with a mean pixel-level AUROC of 98.74%. Code and trained models are available at https:// github.com/mtailanian/uflow.
Authors: Xiwei Xuan, Ziquan Deng, Hsuan-Tien Lin, Zhaodan Kong, Kwan-Liu Ma
Abstract: Researchers have proposed various methods for visually interpreting the Convolutional Neural Network (CNN) via saliency maps, which include Class-Activation-Map (CAM) based approaches as a leading family. However, in terms of the internal design logic, existing CAM-based approaches often overlook the causal perspective that answers the core "why" question to help humans understand the explanation. Additionally, current CNN explanations lack the consideration of both necessity and sufficiency, two complementary sides of a desirable explanation. This paper presents a causality-driven framework, SUNY, designed to rationalize the explanations toward better human understanding. Using the CNN model's input features or internal filters as hypothetical causes, SUNY generates explanations by bi-directional quantifications on both the necessary and sufficient perspectives. Extensive evaluations justify that SUNY not only produces more informative and convincing explanations from the angles of necessity and sufficiency, but also achieves performances competitive to other approaches across different CNN architectures over large-scale datasets, including ILSVRC2012 and CUB-200-2011.
Authors: Sangmin Yoo, Eric Yeu-Jer Lee, Ziyu Wang, Xinxin Wang, Wei D. Lu
Abstract: Event-based cameras are inspired by the sparse and asynchronous spike representation of the biological visual system. However, processing the event data requires either using expensive feature descriptors to transform spikes into frames, or using spiking neural networks that are expensive to train. In this work, we propose a neural network architecture, Reservoir Nodes-enabled neuromorphic vision sensing Network (RN-Net), based on simple convolution layers integrated with dynamic temporal encoding reservoirs for local and global spatiotemporal feature detection with low hardware and training costs. The RN-Net allows efficient processing of asynchronous temporal features, and achieves the highest accuracy of 99.2% for DVS128 Gesture reported to date, and one of the highest accuracy of 67.5% for DVS Lip dataset at a much smaller network size. By leveraging the internal device and circuit dynamics, asynchronous temporal feature encoding can be implemented at very low hardware cost without preprocessing and dedicated memory and arithmetic units. The use of simple DNN blocks and standard backpropagation-based training rules further reduces implementation costs.
Authors: Xiaoxuan Ma, Jiajun Su, Yuan Xu, Wentao Zhu, Chunyu Wang, Yizhou Wang
Abstract: Monocular 3D human mesh estimation faces challenges due to depth ambiguity and the complexity of mapping images to complex parameter spaces. Recent methods propose to use 3D poses as a proxy representation, which often lose crucial body shape information, leading to mediocre performance. Conversely, advanced motion capture systems, though accurate, are impractical for markerless wild images. Addressing these limitations, we introduce an innovative intermediate representation as virtual markers, which are learned from large-scale mocap data, mimicking the effects of physical markers. Building upon virtual markers, we propose VMarker, which detects virtual markers from wild images, and the intact mesh with realistic shapes can be obtained by simply interpolation from these markers. To address occlusions that obscure 3D virtual marker estimation, we further enhance our method with VMarker-Pro, a probabilistic framework that generates multiple plausible meshes aligned with images. This framework models the 3D virtual marker estimation as a conditional denoising process, enabling robust 3D mesh estimation. Our approaches surpass existing methods on three benchmark datasets, particularly demonstrating significant improvements on the SURREAL dataset, which features diverse body shapes. Additionally, VMarker-Pro excels in accurately modeling data distributions, significantly enhancing performance in occluded scenarios. Code and models are available at https://github.com/ShirleyMaxx/VMarker-Pro.
Authors: Yixin Zhang, Maciej A. Mazurowski
Abstract: Shape learning, or the ability to leverage shape information, could be a desirable property of convolutional neural networks (CNNs) when target objects have specific shapes. While some research on the topic is emerging, there is no systematic study to conclusively determine whether and under what circumstances CNNs learn shape. Here, we present such a study in the context of segmentation networks where shapes are particularly important. We define shape and propose a new behavioral metric to measure the extent to which a CNN utilizes shape information. We then execute a set of experiments with synthetic and real-world data to progressively uncover under which circumstances CNNs learn shape and what can be done to encourage such behavior. We conclude that (i) CNNs do not learn shape in typical settings but rather rely on other features available to identify the objects of interest, (ii) CNNs can learn shape, but only if the shape is the only feature available to identify the object, (iii) sufficiently large receptive field size relative to the size of target objects is necessary for shape learning; (iv) a limited set of augmentations can encourage shape learning; (v) learning shape is indeed useful in the presence of out-of-distribution data.
Authors: De Cheng, Lingfeng He, Nannan Wang, Shizhou Zhang, Zhen Wang, Xinbo Gao
Abstract: Unsupervised visible-infrared person re-identification (USL-VI-ReID) aims to match pedestrian images of the same identity from different modalities without annotations. Existing works mainly focus on alleviating the modality gap by aligning instance-level features of the unlabeled samples. However, the relationships between cross-modality clusters are not well explored. To this end, we propose a novel bilateral cluster matching-based learning framework to reduce the modality gap by matching cross-modality clusters. Specifically, we design a Many-to-many Bilateral Cross-Modality Cluster Matching (MBCCM) algorithm through optimizing the maximum matching problem in a bipartite graph. Then, the matched pairwise clusters utilize shared visible and infrared pseudo-labels during the model training. Under such a supervisory signal, a Modality-Specific and Modality-Agnostic (MSMA) contrastive learning framework is proposed to align features jointly at a cluster-level. Meanwhile, the cross-modality Consistency Constraint (CC) is proposed to explicitly reduce the large modality discrepancy. Extensive experiments on the public SYSU-MM01 and RegDB datasets demonstrate the effectiveness of the proposed method, surpassing state-of-the-art approaches by a large margin of 8.76% mAP on average.
Authors: Royden Wagner, Omer Sahin Tas, Marvin Klemp, Carlos Fernandez, Christoph Stiller
Abstract: We introduce RedMotion, a transformer model for motion prediction in self-driving vehicles that learns environment representations via redundancy reduction. Our first type of redundancy reduction is induced by an internal transformer decoder and reduces a variable-sized set of local road environment tokens, representing road graphs and agent data, to a fixed-sized global embedding. The second type of redundancy reduction is obtained by self-supervised learning and applies the redundancy reduction principle to embeddings generated from augmented views of road environments. Our experiments reveal that our representation learning approach outperforms PreTraM, Traj-MAE, and GraphDINO in a semi-supervised setting. Moreover, RedMotion achieves competitive results compared to HPTR or MTR++ in the Waymo Motion Prediction Challenge. Our open-source implementation is available at: https://github.com/kit-mrt/future-motion
Authors: Weikang Yu, Yonghao Xu, Pedram Ghamisi
Abstract: Deep neural networks (DNNs) have risen to prominence as key solutions in numerous AI applications for earth observation (AI4EO). However, their susceptibility to adversarial examples poses a critical challenge, compromising the reliability of AI4EO algorithms. This paper presents a novel Universal Adversarial Defense approach in Remote Sensing Imagery (UAD-RS), leveraging pre-trained diffusion models to protect DNNs against universal adversarial examples exhibiting heterogeneous patterns. Specifically, a universal adversarial purification framework is developed utilizing pre-trained diffusion models to mitigate adversarial perturbations through the introduction of Gaussian noise and subsequent purification of the perturbations from adversarial examples. Additionally, an Adaptive Noise Level Selection (ANLS) mechanism is introduced to determine the optimal noise level for the purification framework with a task-guided Frechet Inception Distance (FID) ranking strategy, thereby enhancing purification performance. Consequently, only a single pre-trained diffusion model is required for purifying universal adversarial samples with heterogeneous patterns across each dataset, significantly reducing training efforts for multiple attack settings while maintaining high performance without prior knowledge of adversarial perturbations. Experimental results on four heterogeneous RS datasets, focusing on scene classification and semantic segmentation, demonstrate that UAD-RS outperforms state-of-the-art adversarial purification approaches, providing universal defense against seven commonly encountered adversarial perturbations. Codes and the pre-trained models are available online (https://github.com/EricYu97/UAD-RS).
Authors: Juncheng Li, Kaihang Pan, Zhiqi Ge, Minghe Gao, Wei Ji, Wenqiao Zhang, Tat-Seng Chua, Siliang Tang, Hanwang Zhang, Yueting Zhuang
Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have been utilizing Visual Prompt Generators (VPGs) to convert visual features into tokens that LLMs can recognize. This is achieved by training the VPGs on millions of image-caption pairs, where the VPG-generated tokens of images are fed into a frozen LLM to generate the corresponding captions. However, this image-captioning based training objective inherently biases the VPG to concentrate solely on the primary visual contents sufficient for caption generation, often neglecting other visual details. This shortcoming results in MLLMs' underperformance in comprehending demonstrative instructions consisting of multiple, interleaved, and multimodal instructions that demonstrate the required context to complete a task. To address this issue, we introduce a generic and lightweight Visual Prompt Generator Complete module (VPG-C), which can infer and complete the missing details essential for comprehending demonstrative instructions. Further, we propose a synthetic discriminative training strategy to fine-tune VPG-C, eliminating the need for supervised demonstrative instructions. As for evaluation, we build DEMON, a comprehensive benchmark for demonstrative instruction understanding. Synthetically trained with the proposed strategy, VPG-C achieves significantly stronger zero-shot performance across all tasks of DEMON. Further evaluation on the MME and OwlEval benchmarks also demonstrate the superiority of VPG-C. Our benchmark, code, and pre-trained models are available at https://github.com/DCDmllm/Cheetah.
Authors: Yu Liang, Yufeng Zhang, Shiliang Zhang, Yaowei Wang, Sheng Xiao, Rong Xiao, Xiaoyu Wang
Abstract: Backward-compatible training circumvents the need for expensive updates to the old gallery database when deploying an advanced new model in the retrieval system. Previous methods achieved backward compatibility by aligning prototypes of the new model with the old one, yet they often overlooked the distribution of old features, limiting their effectiveness when the low quality of the old model results in a weakly feature discriminability. Instance-based methods like L2 regression take into account the distribution of old features but impose strong constraints on the performance of the new model itself. In this paper, we propose MixBCT, a simple yet highly effective backward-compatible training method that serves as a unified framework for old models of varying qualities. We construct a single loss function applied to mixed old and new features to facilitate backward-compatible training, which adaptively adjusts the constraint domain for new features based on the distribution of old features. We conducted extensive experiments on the large-scale face recognition datasets MS1Mv3 and IJB-C to verify the effectiveness of our method. The experimental results clearly demonstrate its superiority over previous methods. Code is available at https://github.com/yuleung/MixBCT .
Authors: Zhenrong Zhang, Jianan Liu, Yuxuan Xia, Tao Huang, Qing-Long Han, Hongbin Liu
Abstract: Online multi-object tracking (MOT) plays a pivotal role in autonomous systems. The state-of-the-art approaches usually employ a tracking-by-detection method, and data association plays a critical role. This paper proposes a learning and graph-optimized (LEGO) modular tracker to improve data association performance in the existing literature. The proposed LEGO tracker integrates graph optimization and self-attention mechanisms, which efficiently formulate the association score map, facilitating the accurate and efficient matching of objects across time frames. To further enhance the state update process, the Kalman filter is added to ensure consistent tracking by incorporating temporal coherence in the object states. Our proposed method utilizing LiDAR alone has shown exceptional performance compared to other online tracking approaches, including LiDAR-based and LiDAR-camera fusion-based methods. LEGO ranked 1st at the time of submitting results to KITTI object tracking evaluation ranking board and remains 2nd at the time of submitting this paper, among all online trackers in the KITTI MOT benchmark for cars1
Authors: Wenjing Xie, Tao Hu, Neiwen Ling, Guoliang Xing, Chun Jason Xue, Nan Guan
Abstract: Fusing Radar and Lidar sensor data can fully utilize their complementary advantages and provide more accurate reconstruction of the surrounding for autonomous driving systems. Surround Radar/Lidar can provide 360-degree view sampling with the minimal cost, which are promising sensing hardware solutions for autonomous driving systems. However, due to the intrinsic physical constraints, the rotating speed of surround Radar, and thus the frequency to generate Radar data frames, is much lower than surround Lidar. Existing Radar/Lidar fusion methods have to work at the low frequency of surround Radar, which cannot meet the high responsiveness requirement of autonomous driving systems.This paper develops techniques to fuse surround Radar/Lidar with working frequency only limited by the faster surround Lidar instead of the slower surround Radar, based on the state-of-the-art object detection model MVDNet. The basic idea of our approach is simple: we let MVDNet work with temporally unaligned data from Radar/Lidar, so that fusion can take place at any time when a new Lidar data frame arrives, instead of waiting for the slow Radar data frame. However, directly applying MVDNet to temporally unaligned Radar/Lidar data greatly degrades its object detection accuracy. The key information revealed in this paper is that we can achieve high output frequency with little accuracy loss by enhancing the training procedure to explore the temporal redundancy in MVDNet so that it can tolerate the temporal unalignment of input data. We explore several different ways of training enhancement and compare them quantitatively with experiments.
Authors: Xiongfeng Peng, Zhihua Liu, Weiming Li, Ping Tan, SoonYong Cho, Qiang Wang
Abstract: Recent deep learning based visual simultaneous localization and mapping (SLAM) methods have made significant progress. However, how to make full use of visual information as well as better integrate with inertial measurement unit (IMU) in visual SLAM has potential research value. This paper proposes a novel deep SLAM network with dual visual factors. The basic idea is to integrate both photometric factor and re-projection factor into the end-to-end differentiable structure through multi-factor data association module. We show that the proposed network dynamically learns and adjusts the confidence maps of both visual factors and it can be further extended to include the IMU factors as well. Extensive experiments validate that our proposed method significantly outperforms the state-of-the-art methods on several public datasets, including TartanAir, EuRoC and ETH3D-SLAM. Specifically, when dynamically fusing the three factors together, the absolute trajectory error for both monocular and stereo configurations on EuRoC dataset has reduced by 45.3% and 36.2% respectively.
Authors: Qiao Yang, Yu Zhang, Zijing Zhao, Jian Zhang, Shunli Zhang
Abstract: Infrared and visible image fusion (IVIF) is used to generate fusion images with comprehensive features of both images, which is beneficial for downstream vision tasks. However, current methods rarely consider the illumination condition in low-light environments, and the targets in the fused images are often not prominent. To address the above issues, we propose an Illumination-Aware Infrared and Visible Image Fusion Network, named as IAIFNet. In our framework, an illumination enhancement network first estimates the incident illumination maps of input images. Afterwards, with the help of proposed adaptive differential fusion module (ADFM) and salient target aware module (STAM), an image fusion network effectively integrates the salient features of the illumination-enhanced infrared and visible images into a fusion image of high visual quality. Extensive experimental results verify that our method outperforms five state-of-the-art methods of fusing infrared and visible images.
Authors: Chao Huang, Zhao Kang, Hong Wu
Abstract: Image anomaly detection and localization perform not only image-level anomaly classification but also locate pixel-level anomaly regions. Recently, it has received much research attention due to its wide application in various fields. This paper proposes ProtoAD, a prototype-based neural network for image anomaly detection and localization. First, the patch features of normal images are extracted by a deep network pre-trained on nature images. Then, the prototypes of the normal patch features are learned by non-parametric clustering. Finally, we construct an image anomaly localization network (ProtoAD) by appending the feature extraction network with $L2$ feature normalization, a $1\times1$ convolutional layer, a channel max-pooling, and a subtraction operation. We use the prototypes as the kernels of the $1\times1$ convolutional layer; therefore, our neural network does not need a training phase and can conduct anomaly detection and localization in an end-to-end manner. Extensive experiments on two challenging industrial anomaly detection datasets, MVTec AD and BTAD, demonstrate that ProtoAD achieves competitive performance compared to the state-of-the-art methods with a higher inference speed. The source code is available at: https://github.com/98chao/ProtoAD.
Authors: Song Zhang, Qingzhong Wang, Jiang Bian, Haoyi Xiong
Abstract: While models derived from Vision Transformers (ViTs) have been phonemically surging, pre-trained models cannot seamlessly adapt to arbitrary resolution images without altering the architecture and configuration, such as sampling the positional encoding, limiting their flexibility for various vision tasks. For instance, the Segment Anything Model (SAM) based on ViT-Huge requires all input images to be resized to 1024$\times$1024. To overcome this limitation, we propose the Multi-Head Self-Attention Convolution (MSA-Conv) that incorporates Self-Attention within generalized convolutions, including standard, dilated, and depthwise ones. Enabling transformers to handle images of varying sizes without retraining or rescaling, the use of MSA-Conv further reduces computational costs compared to global attention in ViT, which grows costly as image size increases. Later, we present the Vision Transformer in Convolution (TiC) as a proof of concept for image classification with MSA-Conv, where two capacity enhancing strategies, namely Multi-Directional Cyclic Shifted Mechanism and Inter-Pooling Mechanism, have been proposed, through establishing long-distance connections between tokens and enlarging the effective receptive field. Extensive experiments have been carried out to validate the overall effectiveness of TiC. Additionally, ablation studies confirm the performance improvement made by MSA-Conv and the two capacity enhancing strategies separately. Note that our proposal aims at studying an alternative to the global attention used in ViT, while MSA-Conv meets our goal by making TiC comparable to state-of-the-art on ImageNet-1K. Code will be released at https://github.com/zs670980918/MSA-Conv.
Authors: Sanhita Pathak, Vinay Kaushik, Brejesh Lall
Abstract: Image-based virtual try-on aims to fit an in-shop garment onto a clothed person image. Garment warping, which aligns the target garment with the corresponding body parts in the person image, is a crucial step in achieving this goal. Existing methods often use multi-stage frameworks to handle clothes warping, person body synthesis and tryon generation separately or rely on noisy intermediate parser-based labels. We propose a novel single-stage framework that implicitly learns the same without explicit multi-stage learning. Our approach utilizes a novel semantic-contextual fusion attention module for garment-person feature fusion, enabling efficient and realistic cloth warping and body synthesis from target pose keypoints. By introducing a lightweight linear attention framework that attends to garment regions and fuses multiple sampled flow fields, we also address misalignment and artifacts present in previous methods. To achieve simultaneous learning of warped garment and try-on results, we introduce a Warped Cloth Learning Module. Our proposed approach significantly improves the quality and efficiency of virtual try-on methods, providing users with a more reliable and realistic virtual try-on experience.
Authors: Gang Xu, Shuhao Wang, Lingyu Zhao, Xiao Chen, Tongwei Wang, Lang Wang, Zhenwei Luo, Dahan Wang, Zewen Zhang, Aijun Liu, Wei Ba, Zhigang Song, Huaiyin Shi, Dingrong Zhong, Jianpeng Ma
Abstract: Histopathology image analysis plays a crucial role in cancer diagnosis. However, training a clinically applicable segmentation algorithm requires pathologists to engage in labour-intensive labelling. In contrast, weakly supervised learning methods, which only require coarse-grained labels at the image level, can significantly reduce the labeling efforts. Unfortunately, while these methods perform reasonably well in slide-level prediction, their ability to locate cancerous regions, which is essential for many clinical applications, remains unsatisfactory. Previously, we proposed CAMEL, which achieves comparable results to those of fully supervised baselines in pixel-level segmentation. However, CAMEL requires 1,280x1,280 image-level binary annotations for positive WSIs. Here, we present CAMEL2, by introducing a threshold of the cancerous ratio for positive bags, it allows us to better utilize the information, consequently enabling us to scale up the image-level setting from 1,280x1,280 to 5,120x5,120 while maintaining the accuracy. Our results with various datasets, demonstrate that CAMEL2, with the help of 5,120x5,120 image-level binary annotations, which are easy to annotate, achieves comparable performance to that of a fully supervised baseline in both instance- and slide-level classifications.
Authors: Zhenyi Liao, Zhijie Deng
Abstract: Leveraging pre-trained conditional diffusion models for video editing without further tuning has gained increasing attention due to its promise in film production, advertising, etc. Yet, seminal works in this line fall short in generation length, temporal coherence, or fidelity to the source video. This paper aims to bridge the gap, establishing a simple and effective baseline for training-free diffusion model-based long video editing. As suggested by prior arts, we build the pipeline upon ControlNet, which excels at various image editing tasks based on text prompts. To break down the length constraints caused by limited computational memory, we split the long video into consecutive windows and develop a novel cross-window attention mechanism to ensure the consistency of global style and maximize the smoothness among windows. To achieve more accurate control, we extract the information from the source video via DDIM inversion and integrate the outcomes into the latent states of the generations. We also incorporate a video frame interpolation model to mitigate the frame-level flickering issue. Extensive empirical studies verify the superior efficacy of our method over competing baselines across scenarios, including the replacement of the attributes of foreground objects, style transfer, and background replacement. Besides, our method manages to edit videos comprising hundreds of frames according to user requirements. Our project is open-sourced and the project page is at https://github.com/zhijie-group/LOVECon.
Authors: Chen Jin, Ryutaro Tanno, Amrutha Saseendran, Tom Diethe, Philip Teare
Abstract: Textural Inversion, a prompt learning method, learns a singular text embedding for a new "word" to represent image style and appearance, allowing it to be integrated into natural language sentences to generate novel synthesised images. However, identifying multiple unknown object-level concepts within one scene remains a complex challenge. While recent methods have resorted to cropping or masking individual images to learn multiple concepts, these techniques often require prior knowledge of new concepts and are labour-intensive. To address this challenge, we introduce Multi-Concept Prompt Learning (MCPL), where multiple unknown "words" are simultaneously learned from a single sentence-image pair, without any imagery annotations. To enhance the accuracy of word-concept correlation and refine attention mask boundaries, we propose three regularisation techniques: Attention Masking, Prompts Contrastive Loss, and Bind Adjective. Extensive quantitative comparisons with both real-world categories and biomedical images demonstrate that our method can learn new semantically disentangled concepts. Our approach emphasises learning solely from textual embeddings, using less than 10% of the storage space compared to others. The project page, code, and data are available at https://astrazeneca.github.io/mcpl.github.io.
Authors: Yicheng Lin, Shuo Wang, Yunlong Jiang, Bin Han
Abstract: Sparse optical flow is widely used in various computer vision tasks, however assuming brightness consistency limits its performance in High Dynamic Range (HDR) environments. In this work, a lightweight network is used to extract illumination robust convolutional features and corners with strong invariance. Modifying the typical brightness consistency of the optical flow method to the convolutional feature consistency yields the light-robust hybrid optical flow method. The proposed network runs at 190 FPS on a commercial CPU because it uses only four convolutional layers to extract feature maps and score maps simultaneously. Since the shallow network is difficult to train directly, a deep network is designed to compute the reliability map that helps it. An end-to-end unsupervised training mode is used for both networks. To validate the proposed method, we compare corner repeatability and matching performance with origin optical flow under dynamic illumination. In addition, a more accurate visual inertial system is constructed by replacing the optical flow method in VINS-Mono. In a public HDR dataset, it reduces translation errors by 93\%. The code is publicly available at https://github.com/linyicheng1/LET-NET.
Authors: Changdae Oh, Hyesu Lim, Mijoo Kim, Dongyoon Han, Sangdoo Yun, Jaegul Choo, Alexander Hauptmann, Zhi-Qi Cheng, Kyungwoo Song
Abstract: Improving out-of-distribution (OOD) generalization through in-distribution (ID) adaptation is a primary goal of robust fine-tuning methods beyond the naive fine-tuning approach. However, despite decent OOD generalization performance from recent robust fine-tuning methods, OOD confidence calibration for reliable machine learning has not been fully addressed. This work proposes a robust fine-tuning method that improves both OOD accuracy and calibration error in Vision Language Models (VLMs). Firstly, we show that both types of errors have a shared upper bound consisting of two terms of ID data: 1) calibration error and 2) the smallest singular value of the input covariance matrix. Based on this insight, we design a novel framework that conducts fine-tuning with a constrained multimodal contrastive loss enforcing a larger smallest singular value, which is further aided by the self-distillation of a moving averaged model to achieve well-calibrated prediction. Starting from an empirical validation of our theoretical statements, we provide extensive experimental results on ImageNet distribution shift benchmarks that demonstrate the effectiveness of our method.
Authors: Zhe Li, Yipengjing Sun, Zerong Zheng, Lizhen Wang, Shengping Zhang, Yebin Liu
Abstract: Modeling animatable human avatars from RGB videos is a long-standing and challenging problem. Recent works usually adopt MLP-based neural radiance fields (NeRF) to represent 3D humans, but it remains difficult for pure MLPs to regress pose-dependent garment details. To this end, we introduce Animatable Gaussians, a new avatar representation that leverages powerful 2D CNNs and 3D Gaussian splatting to create high-fidelity avatars. To associate 3D Gaussians with the animatable avatar, we learn a parametric template from the input videos, and then parameterize the template on two front & back canonical Gaussian maps where each pixel represents a 3D Gaussian. The learned template is adaptive to the wearing garments for modeling looser clothes like dresses. Such template-guided 2D parameterization enables us to employ a powerful StyleGAN-based CNN to learn the pose-dependent Gaussian maps for modeling detailed dynamic appearances. Furthermore, we introduce a pose projection strategy for better generalization given novel poses. To tackle the realistic relighting of animatable avatars, we introduce physically-based rendering into the avatar representation for decomposing avatar materials and environment illumination. Overall, our method can create lifelike avatars with dynamic, realistic, generalized and relightable appearances. Experiments show that our method outperforms other state-of-the-art approaches.
Authors: Vandan Gorade, Sparsh Mittal, Debesh Jha, Ulas Bagci
Abstract: Knowledge distillation (KD) has demonstrated remarkable success across various domains, but its application to medical imaging tasks, such as kidney and liver tumor segmentation, has encountered challenges. Many existing KD methods are not specifically tailored for these tasks. Moreover, prevalent KD methods often lack a careful consideration of `what' and `from where' to distill knowledge from the teacher to the student. This oversight may lead to issues like the accumulation of training bias within shallower student layers, potentially compromising the effectiveness of KD. To address these challenges, we propose Hierarchical Layer-selective Feedback Distillation (HLFD). HLFD strategically distills knowledge from a combination of middle layers to earlier layers and transfers final layer knowledge to intermediate layers at both the feature and pixel levels. This design allows the model to learn higher-quality representations from earlier layers, resulting in a robust and compact student model. Extensive quantitative evaluations reveal that HLFD outperforms existing methods by a significant margin. For example, in the kidney segmentation task, HLFD surpasses the student model (without KD) by over 10\%, significantly improving its focus on tumor-specific features. From a qualitative standpoint, the student model trained using HLFD excels at suppressing irrelevant information and can focus sharply on tumor-specific details, which opens a new pathway for more efficient and accurate diagnostic tools. Code is available \href{https://github.com/vangorade/RethinkingKD_ISBI24}{here}.
Authors: Sherwin Bahmani, Ivan Skorokhodov, Victor Rong, Gordon Wetzstein, Leonidas Guibas, Peter Wonka, Sergey Tulyakov, Jeong Joon Park, Andrea Tagliasacchi, David B. Lindell
Abstract: Recent breakthroughs in text-to-4D generation rely on pre-trained text-to-image and text-to-video models to generate dynamic 3D scenes. However, current text-to-4D methods face a three-way tradeoff between the quality of scene appearance, 3D structure, and motion. For example, text-to-image models and their 3D-aware variants are trained on internet-scale image datasets and can be used to produce scenes with realistic appearance and 3D structure -- but no motion. Text-to-video models are trained on relatively smaller video datasets and can produce scenes with motion, but poorer appearance and 3D structure. While these models have complementary strengths, they also have opposing weaknesses, making it difficult to combine them in a way that alleviates this three-way tradeoff. Here, we introduce hybrid score distillation sampling, an alternating optimization procedure that blends supervision signals from multiple pre-trained diffusion models and incorporates benefits of each for high-fidelity text-to-4D generation. Using hybrid SDS, we demonstrate synthesis of 4D scenes with compelling appearance, 3D structure, and motion.
Authors: Ziyang Chen, Yongsheng Pan, Yiwen Ye, Mengkang Lu, Yong Xia
Abstract: Distribution shift widely exists in medical images acquired from different medical centres and poses a significant obstacle to deploying the pre-trained semantic segmentation model in real-world applications. Test-time adaptation has proven its effectiveness in tackling the cross-domain distribution shift during inference. However, most existing methods achieve adaptation by updating the pre-trained models, rendering them susceptible to error accumulation and catastrophic forgetting when encountering a series of distribution shifts (i.e., under the continual test-time adaptation setup). To overcome these challenges caused by updating the models, in this paper, we freeze the pre-trained model and propose the Visual Prompt-based Test-Time Adaptation (VPTTA) method to train a specific prompt for each test image to align the statistics in the batch normalization layers. Specifically, we present the low-frequency prompt, which is lightweight with only a few parameters and can be effectively trained in a single iteration. To enhance prompt initialization, we equip VPTTA with a memory bank to benefit the current prompt from previous ones. Additionally, we design a warm-up mechanism, which mixes source and target statistics to construct warm-up statistics, thereby facilitating the training process. Extensive experiments demonstrate the superiority of our VPTTA over other state-of-the-art methods on two medical image segmentation benchmark tasks. The code and weights of pre-trained source models are available at https://github.com/Chen-Ziyang/VPTTA.
Authors: Jiazhong Cen, Jiemin Fang, Chen Yang, Lingxi Xie, Xiaopeng Zhang, Wei Shen, Qi Tian
Abstract: This paper presents SAGA (Segment Any 3D GAussians), a highly efficient 3D promptable segmentation method based on 3D Gaussian Splatting (3D-GS). Given 2D visual prompts as input, SAGA can segment the corresponding 3D target represented by 3D Gaussians within 4 ms. This is achieved by attaching an scale-gated affinity feature to each 3D Gaussian to endow it a new property towards multi-granularity segmentation. Specifically, a scale-aware contrastive training strategy is proposed for the scale-gated affinity feature learning. It 1) distills the segmentation capability of the Segment Anything Model (SAM) from 2D masks into the affinity features and 2) employs a soft scale gate mechanism to deal with multi-granularity ambiguity in 3D segmentation through adjusting the magnitude of each feature channel according to a specified 3D physical scale. Evaluations demonstrate that SAGA achieves real-time multi-granularity segmentation with quality comparable to state-of-the-art methods. As one of the first methods addressing promptable segmentation in 3D-GS, the simplicity and effectiveness of SAGA pave the way for future advancements in this field. Our code will be released.
Authors: Anh-Quan Cao, Angela Dai, Raoul de Charette
Abstract: We propose the task of Panoptic Scene Completion (PSC) which extends the recently popular Semantic Scene Completion (SSC) task with instance-level information to produce a richer understanding of the 3D scene. Our PSC proposal utilizes a hybrid mask-based technique on the non-empty voxels from sparse multi-scale completions. Whereas the SSC literature overlooks uncertainty which is critical for robotics applications, we instead propose an efficient ensembling to estimate both voxel-wise and instance-wise uncertainties along PSC. This is achieved by building on a multi-input multi-output (MIMO) strategy, while improving performance and yielding better uncertainty for little additional compute. Additionally, we introduce a technique to aggregate permutation-invariant mask predictions. Our experiments demonstrate that our method surpasses all baselines in both Panoptic Scene Completion and uncertainty estimation on three large-scale autonomous driving datasets. Our code and data are available at https://astra-vision.github.io/PaSCo .
Authors: Xiaochen Zhao, Jingxiang Sun, Lizhen Wang, Jinli Suo, Yebin Liu
Abstract: While high fidelity and efficiency are central to the creation of digital head avatars, recent methods relying on 2D or 3D generative models often experience limitations such as shape distortion, expression inaccuracy, and identity flickering. Additionally, existing one-shot inversion techniques fail to fully leverage multiple input images for detailed feature extraction. We propose a novel framework, \textbf{Incremental 3D GAN Inversion}, that enhances avatar reconstruction performance using an algorithm designed to increase the fidelity from multiple frames, resulting in improved reconstruction quality proportional to frame count. Our method introduces a unique animatable 3D GAN prior with two crucial modifications for enhanced expression controllability alongside an innovative neural texture encoder that categorizes texture feature spaces based on UV parameterization. Differentiating from traditional techniques, our architecture emphasizes pixel-aligned image-to-image translation, mitigating the need to learn correspondences between observation and canonical spaces. Furthermore, we incorporate ConvGRU-based recurrent networks for temporal data aggregation from multiple frames, boosting geometry and texture detail reconstruction. The proposed paradigm demonstrates state-of-the-art performance on one-shot and few-shot avatar animation tasks. Code will be available at https://github.com/XChenZ/invertAvatar.
Authors: Samar Khanna, Patrick Liu, Linqi Zhou, Chenlin Meng, Robin Rombach, Marshall Burke, David Lobell, Stefano Ermon
Abstract: Diffusion models have achieved state-of-the-art results on many modalities including images, speech, and video. However, existing models are not tailored to support remote sensing data, which is widely used in important applications including environmental monitoring and crop-yield prediction. Satellite images are significantly different from natural images -- they can be multi-spectral, irregularly sampled across time -- and existing diffusion models trained on images from the Web do not support them. Furthermore, remote sensing data is inherently spatio-temporal, requiring conditional generation tasks not supported by traditional methods based on captions or images. In this paper, we present DiffusionSat, to date the largest generative foundation model trained on a collection of publicly available large, high-resolution remote sensing datasets. As text-based captions are sparsely available for satellite images, we incorporate the associated metadata such as geolocation as conditioning information. Our method produces realistic samples and can be used to solve multiple generative tasks including temporal generation, superresolution given multi-spectral inputs and in-painting. Our method outperforms previous state-of-the-art methods for satellite image generation and is the first large-scale generative foundation model for satellite imagery. The project website can be found here: https://samar-khanna.github.io/DiffusionSat/
Authors: Xudong Li, Timin Gao, Runze Hu, Yan Zhang, Shengchuan Zhang, Xiawu Zheng, Jingyuan Zheng, Yunhang Shen, Ke Li, Yutao Liu, Pingyang Dai, Rongrong Ji
Abstract: The current state-of-the-art No-Reference Image Quality Assessment (NR-IQA) methods typically rely on feature extraction from upstream semantic backbone networks, assuming that all extracted features are relevant. However, we make a key observation that not all features are beneficial, and some may even be harmful, necessitating careful selection. Empirically, we find that many image pairs with small feature spatial distances can have vastly different quality scores, indicating that the extracted features may contain a significant amount of quality-irrelevant noise. To address this issue, we propose a Quality-Aware Feature Matching IQA Metric (QFM-IQM) that employs an adversarial perspective to remove harmful semantic noise features from the upstream task. Specifically, QFM-IQM enhances the semantic noise distinguish capabilities by matching image pairs with similar quality scores but varying semantic features as adversarial semantic noise and adaptively adjusting the upstream task's features by reducing sensitivity to adversarial noise perturbation. Furthermore, we utilize a distillation framework to expand the dataset and improve the model's generalization ability. Our approach achieves superior performance to the state-of-the-art NR-IQA methods on eight standard IQA datasets.
Authors: H. S. Tan, Kuancheng Wang, Rafe Mcbeth
Abstract: In this work, we study various hybrid models of entropy-based and representativeness sampling techniques in the context of active learning in medical segmentation, in particular examining the role of UMAP (Uniform Manifold Approximation and Projection) as a technique for capturing representativeness. Although UMAP has been shown viable as a general purpose dimension reduction method in diverse areas, its role in deep learning-based medical segmentation has yet been extensively explored. Using the cardiac and prostate datasets in the Medical Segmentation Decathlon for validation, we found that a novel hybrid combination of Entropy-UMAP sampling technique achieved a statistically significant Dice score advantage over the random baseline ($3.2 \%$ for cardiac, $4.5 \%$ for prostate), and attained the highest Dice coefficient among the spectrum of 10 distinct active learning methodologies we examined. This provides preliminary evidence that there is an interesting synergy between entropy-based and UMAP methods when the former precedes the latter in a hybrid model of active learning.
Authors: Amit Rozner, Barak Battash, Ofir Lindenbaum, Lior Wolf
Abstract: We study the problem of performing face verification with an efficient neural model $f$. The efficiency of $f$ stems from simplifying the face verification problem from an embedding nearest neighbor search into a binary problem; each user has its own neural network $f$. To allow information sharing between different individuals in the training set, we do not train $f$ directly but instead generate the model weights using a hypernetwork $h$. This leads to the generation of a compact personalized model for face identification that can be deployed on edge devices. Key to the method's success is a novel way of generating hard negatives and carefully scheduling the training objectives. Our model leads to a substantially small $f$ requiring only 23k parameters and 5M floating point operations (FLOPS). We use six face verification datasets to demonstrate that our method is on par or better than state-of-the-art models, with a significantly reduced number of parameters and computational burden. Furthermore, we perform an extensive ablation study to demonstrate the importance of each element in our method.
Authors: Zhaoyang Wei, Pengfei Chen, Xuehui Yu, Guorong Li, Jianbin Jiao, Zhenjun Han
Abstract: Single-point annotation in visual tasks, with the goal of minimizing labelling costs, is becoming increasingly prominent in research. Recently, visual foundation models, such as Segment Anything (SAM), have gained widespread usage due to their robust zero-shot capabilities and exceptional annotation performance. However, SAM's class-agnostic output and high confidence in local segmentation introduce 'semantic ambiguity', posing a challenge for precise category-specific segmentation. In this paper, we introduce a cost-effective category-specific segmenter using SAM. To tackle this challenge, we have devised a Semantic-Aware Instance Segmentation Network (SAPNet) that integrates Multiple Instance Learning (MIL) with matching capability and SAM with point prompts. SAPNet strategically selects the most representative mask proposals generated by SAM to supervise segmentation, with a specific focus on object category information. Moreover, we introduce the Point Distance Guidance and Box Mining Strategy to mitigate inherent challenges: 'group' and 'local' issues in weakly supervised segmentation. These strategies serve to further enhance the overall segmentation performance. The experimental results on Pascal VOC and COCO demonstrate the promising performance of our proposed SAPNet, emphasizing its semantic matching capabilities and its potential to advance point-prompted instance segmentation. The code will be made publicly available.
Authors: Paul K. Mandal, Cole Leo, Connor Hurley
Abstract: In the modern world, the amount of visual data recorded has been rapidly increasing. In many cases, data is stored in geographically distinct locations and thus requires a large amount of time and space to consolidate. Sometimes, there are also regulations for privacy protection which prevent data consolidation. In this work, we present federated implementations for object detection and recognition using a federated Faster R-CNN (FRCNN) and image segmentation using a federated Fully Convolutional Network (FCN). Our FRCNN was trained on 5000 examples of the COCO2017 dataset while our FCN was trained on the entire train set of the CamVid dataset. The proposed federated models address the challenges posed by the increasing volume and decentralized nature of visual data, offering efficient solutions in compliance with privacy regulations.
Authors: Byeonghyeon Lee, Howoong Lee, Xiangyu Sun, Usman Ali, Eunbyung Park
Abstract: Recent studies in Radiance Fields have paved the robust way for novel view synthesis with their photorealistic rendering quality. Nevertheless, they usually employ neural networks and volumetric rendering, which are costly to train and impede their broad use in various real-time applications due to the lengthy rendering time. Lately 3D Gaussians splatting-based approach has been proposed to model the 3D scene, and it achieves remarkable visual quality while rendering the images in real-time. However, it suffers from severe degradation in the rendering quality if the training images are blurry. Blurriness commonly occurs due to the lens defocusing, object motion, and camera shake, and it inevitably intervenes in clean image acquisition. Several previous studies have attempted to render clean and sharp images from blurry input images using neural fields. The majority of those works, however, are designed only for volumetric rendering-based neural radiance fields and are not straightforwardly applicable to rasterization-based 3D Gaussian splatting methods. Thus, we propose a novel real-time deblurring framework, Deblurring 3D Gaussian Splatting, using a small Multi-Layer Perceptron (MLP) that manipulates the covariance of each 3D Gaussian to model the scene blurriness. While Deblurring 3D Gaussian Splatting can still enjoy real-time rendering, it can reconstruct fine and sharp details from blurry images. A variety of experiments have been conducted on the benchmark, and the results have revealed the effectiveness of our approach for deblurring. Qualitative results are available at https://benhenryl.github.io/Deblurring-3D-Gaussian-Splatting/
URLs: https://benhenryl.github.io/Deblurring-3D-Gaussian-Splatting/
Authors: Eytan Kats, Jochen G. Hirsch, Mattias P. Heinrich
Abstract: This paper demonstrates a self-supervised framework for learning voxel-wise coarse-to-fine representations tailored for dense downstream tasks. Our approach stems from the observation that existing methods for hierarchical representation learning tend to prioritize global features over local features due to inherent architectural bias. To address this challenge, we devise a training strategy that balances the contributions of features from multiple scales, ensuring that the learned representations capture both coarse and fine-grained details. Our strategy incorporates 3-fold improvements: (1) local data augmentations, (2) a hierarchically balanced architecture, and (3) a hybrid contrastive-restorative loss function. We evaluate our method on CT and MRI data and demonstrate that our new approach particularly beneficial for fine-tuning with limited annotated data and consistently outperforms the baseline counterpart in linear evaluation settings.
Authors: Yue Liu, Yunjie Tian, Yuzhong Zhao, Hongtian Yu, Lingxi Xie, Yaowei Wang, Qixiang Ye, Yunfan Liu
Abstract: Designing computationally efficient network architectures persists as an ongoing necessity in computer vision. In this paper, we transplant Mamba, a state-space language model, into VMamba, a vision backbone that works in linear time complexity. At the core of VMamba lies a stack of Visual State-Space (VSS) blocks with the 2D Selective Scan (SS2D) module. By traversing along four scanning routes, SS2D helps bridge the gap between the ordered nature of 1D selective scan and the non-sequential structure of 2D vision data, which facilitates the gathering of contextual information from various sources and perspectives. Based on the VSS blocks, we develop a family of VMamba architectures and accelerate them through a succession of architectural and implementation enhancements. Extensive experiments showcase VMamba's promising performance across diverse visual perception tasks, highlighting its advantages in input scaling efficiency compared to existing benchmark models. Source code is available at https://github.com/MzeroMiko/VMamba.
Authors: Yaokun Li, Chao Gou, Guang Tan
Abstract: Implicit neural representations, represented by Neural Radiance Fields (NeRF), have dominated research in 3D computer vision by virtue of high-quality visual results and data-driven benefits. However, their realistic applications are hindered by the need for dense inputs and per-scene optimization. To solve this problem, previous methods implement generalizable NeRFs by extracting local features from sparse inputs as conditions for the NeRF decoder. However, although this way can allow feed-forward reconstruction, they suffer from the inherent drawback of yielding sub-optimal results caused by erroneous reprojected features. In this paper, we focus on this problem and aim to address it by introducing pre-trained generative priors to enable high-quality generalizable novel view synthesis. Specifically, we propose a novel Indirect Diffusion-guided NeRF framework, termed ID-NeRF, which leverages pre-trained diffusion priors as a guide for the reprojected features created by the previous paradigm. Notably, to enable 3D-consistent predictions, the proposed ID-NeRF discards the way of direct supervision commonly used in prior 3D generative models and instead adopts a novel indirect prior injection strategy. This strategy is implemented by distilling pre-trained knowledge into an imaginative latent space via score-based distillation, and an attention-based refinement module is then proposed to leverage the embedded priors to improve reprojected features extracted from sparse inputs. We conduct extensive experiments on multiple datasets to evaluate our method, and the results demonstrate the effectiveness of our method in synthesizing novel views in a generalizable manner, especially in sparse settings.
Authors: Nantheera Anantrasirichai, Ruirui Lin, Alexandra Malyugina, David Bull
Abstract: Low-light videos often exhibit spatiotemporal incoherent noise, leading to poor visibility and compromised performance across various computer vision applications. One significant challenge in enhancing such content using modern technologies is the scarcity of training data. This paper introduces a novel low-light video dataset, consisting of 40 scenes captured in various motion scenarios under two distinct low-lighting conditions, incorporating genuine noise and temporal artifacts. We provide fully registered ground truth data captured in normal light using a programmable motorized dolly, and subsequently, refine them via image-based post-processing to ensure the pixel-wise alignment of frames in different light levels. This paper also presents an exhaustive analysis of the low-light dataset, and demonstrates the extensive and representative nature of our dataset in the context of supervised learning. Our experimental results demonstrate the significance of fully registered video pairs in the development of low-light video enhancement methods and the need for comprehensive evaluation. Our dataset is available at DOI:10.21227/mzny-8c77.
Authors: Yuzhu Wang, Lechao Cheng, Chaowei Fang, Dingwen Zhang, Manni Duan, Meng Wang
Abstract: Visual prompt tuning (VPT) is a promising solution incorporating learnable prompt tokens to customize pre-trained models for downstream tasks. However, VPT and its variants often encounter challenges like prompt initialization, prompt length, and subpar performance in self-supervised pretraining, hindering successful contextual adaptation. This study commences by exploring the correlation evolvement between prompts and patch tokens during proficient training. Inspired by the observation that the prompt tokens tend to share high mutual information with patch tokens, we propose initializing prompts with downstream token prototypes. The strategic initialization, a stand-in for the previous initialization, substantially improves performance in fine-tuning. To refine further, we optimize token construction with a streamlined pipeline that maintains excellent performance with almost no increase in computational expenses compared to VPT. Exhaustive experiments show our proposed approach outperforms existing methods by a remarkable margin. For instance, it surpasses full fine-tuning in 19 out of 24 tasks, using less than 0.4% of learnable parameters on the FGVC and VTAB-1K benchmarks. Notably, our method significantly advances the adaptation for self-supervised pretraining, achieving impressive task performance gains of at least 10% to 30%. Besides, the experimental results demonstrate the proposed SPT is robust to prompt lengths and scales well with model capacity and training data size. We finally provide an insightful exploration into the amount of target data facilitating the adaptation of pre-trained models to downstream tasks. The code is available at https://github.com/WangYZ1608/Self-Prompt-Tuning.
Authors: Sheng Luo, Wei Chen, Wanxin Tian, Rui Liu, Luanxuan Hou, Xiubao Zhang, Haifeng Shen, Ruiqi Wu, Shuyi Geng, Yi Zhou, Ling Shao, Yi Yang, Bojun Gao, Qun Li, Guobin Wu
Abstract: Foundation models have indeed made a profound impact on various fields, emerging as pivotal components that significantly shape the capabilities of intelligent systems. In the context of intelligent vehicles, leveraging the power of foundation models has proven to be transformative, offering notable advancements in visual understanding. Equipped with multi-modal and multi-task learning capabilities, multi-modal multi-task visual understanding foundation models (MM-VUFMs) effectively process and fuse data from diverse modalities and simultaneously handle various driving-related tasks with powerful adaptability, contributing to a more holistic understanding of the surrounding scene. In this survey, we present a systematic analysis of MM-VUFMs specifically designed for road scenes. Our objective is not only to provide a comprehensive overview of common practices, referring to task-specific models, unified multi-modal models, unified multi-task models, and foundation model prompting techniques, but also to highlight their advanced capabilities in diverse learning paradigms. These paradigms include open-world understanding, efficient transfer for road scenes, continual learning, interactive and generative capability. Moreover, we provide insights into key challenges and future trends, such as closed-loop driving systems, interpretability, embodied driving agents, and world models. To facilitate researchers in staying abreast of the latest developments in MM-VUFMs for road scenes, we have established a continuously updated repository at https://github.com/rolsheng/MM-VUFM4DS
Authors: Hyoungseob Park, Anjali Gupta, Alex Wong
Abstract: It is common to observe performance degradation when transferring models trained on some (source) datasets to target testing data due to a domain gap between them. Existing methods for bridging this gap, such as domain adaptation (DA), may require the source data on which the model was trained (often not available), while others, i.e., source-free DA, require many passes through the testing data. We propose an online test-time adaptation method for depth completion, the task of inferring a dense depth map from a single image and associated sparse depth map, that closes the performance gap in a single pass. We first present a study on how the domain shift in each data modality affects model performance. Based on our observations that the sparse depth modality exhibits a much smaller covariate shift than the image, we design an embedding module trained in the source domain that preserves a mapping from features encoding only sparse depth to those encoding image and sparse depth. During test time, sparse depth features are projected using this map as a proxy for source domain features and are used as guidance to train a set of auxiliary parameters (i.e., adaptation layer) to align image and sparse depth features from the target test domain to that of the source domain. We evaluate our method on indoor and outdoor scenarios and show that it improves over baselines by an average of 21.1%.
Authors: Markus Hiller, Krista A. Ehinger, Tom Drummond
Abstract: We present a novel bi-directional Transformer architecture (BiXT) which scales linearly with input size in terms of computational cost and memory consumption, but does not suffer the drop in performance or limitation to only one input modality seen with other efficient Transformer-based approaches. BiXT is inspired by the Perceiver architectures but replaces iterative attention with an efficient bi-directional cross-attention module in which input tokens and latent variables attend to each other simultaneously, leveraging a naturally emerging attention-symmetry between the two. This approach unlocks a key bottleneck experienced by Perceiver-like architectures and enables the processing and interpretation of both semantics ('what') and location ('where') to develop alongside each other over multiple layers -- allowing its direct application to dense and instance-based tasks alike. By combining efficiency with the generality and performance of a full Transformer architecture, BiXT can process longer sequences like point clouds, text or images at higher feature resolutions and achieves competitive performance across a range of tasks like point cloud part segmentation, semantic image segmentation, image classification, hierarchical sequence modeling and document retrieval. Our experiments demonstrate that BiXT models outperform larger competitors by leveraging longer sequences more efficiently on vision tasks like classification and segmentation, and perform on par with full Transformer variants on sequence modeling and document retrieval -- but require $28\%$ fewer FLOPs and are up to $8.4\times$ faster.
Authors: Sankarshanaa Sagaram, Krish Didwania, Laven Srivastava, Aditya Kasliwal, Pallavi Kailas, Ujjwal Verma
Abstract: The increasing adoption of solar energy necessitates advanced methodologies for monitoring and maintenance to ensure optimal performance of solar panel installations. A critical component in this context is the accurate segmentation of solar panels from aerial or satellite imagery, which is essential for identifying operational issues and assessing efficiency. This paper addresses the significant challenges in panel segmentation, particularly the scarcity of annotated data and the labour-intensive nature of manual annotation for supervised learning. We explore and apply Self-Supervised Learning (SSL) to solve these challenges. We demonstrate that SSL significantly enhances model generalization under various conditions and reduces dependency on manually annotated data, paving the way for robust and adaptable solar panel segmentation solutions.
Authors: Jiazhao Zhang, Kunyu Wang, Rongtao Xu, Gengze Zhou, Yicong Hong, Xiaomeng Fang, Qi Wu, Zhizheng Zhang, He Wang
Abstract: Vision-and-language navigation (VLN) stands as a key research problem of Embodied AI, aiming at enabling agents to navigate in unseen environments following linguistic instructions. In this field, generalization is a long-standing challenge, either to out-of-distribution scenes or from Sim to Real. In this paper, we propose NaVid, a video-based large vision language model (VLM), to mitigate such a generalization gap. NaVid makes the first endeavor to showcase the capability of VLMs to achieve state-of-the-art level navigation performance without any maps, odometers, or depth inputs. Following human instruction, NaVid only requires an on-the-fly video stream from a monocular RGB camera equipped on the robot to output the next-step action. Our formulation mimics how humans navigate and naturally gets rid of the problems introduced by odometer noises, and the Sim2Real gaps from map or depth inputs. Moreover, our video-based approach can effectively encode the historical observations of robots as spatio-temporal contexts for decision making and instruction following. We train NaVid with 510k navigation samples collected from continuous environments, including action-planning and instruction-reasoning samples, along with 763k large-scale web data. Extensive experiments show that NaVid achieves state-of-the-art performance in simulation environments and the real world, demonstrating superior cross-dataset and Sim2Real transfer. We thus believe our proposed VLM approach plans the next step for not only the navigation agents but also this research field.
Authors: Tan-Hanh Pham, Kim-Doang Nguyen
Abstract: Precision devices play an important role in enhancing production quality and productivity in agricultural systems. Therefore, the optimization of these devices is essential in precision agriculture. Recently, with the advancements of deep learning, there have been several studies aiming to harness its capabilities for improving spray system performance. However, the effectiveness of these methods heavily depends on the size of the training dataset, which is expensive and time-consuming to collect. To address the challenge of insufficient training samples, we developed an image generator named DropletGAN to generate images of droplets. The DropletGAN model is trained by using a small dataset captured by a high-speed camera and capable of generating images with progressively increasing resolution. The results demonstrate that the model can generate high-quality images with the size of 1024x1024. The generated images from the DropletGAN are evaluated using the Fr\'echet inception distance (FID) with an FID score of 11.29. Furthermore, this research leverages recent advancements in computer vision and deep learning to develop a light droplet detector using the synthetic dataset. As a result, the detection model achieves a 16.06% increase in mean average precision (mAP) when utilizing the synthetic dataset. To the best of our knowledge, this work stands as the first to employ a generative model for augmenting droplet detection. Its significance lies not only in optimizing nozzle design for constructing efficient spray systems but also in addressing the common challenge of insufficient data in various precision agriculture tasks. This work offers a critical contribution to conserving resources while striving for optimal and sustainable agricultural practices.
Authors: Jiangming Shi, Xiangbo Yin, Yaoxing Wang, Xiaofeng Liu, Yuan Xie, Yanyun Qu
Abstract: Unsupervised visible-infrared person re-identification (USVI-ReID) aims to match specified people in infrared images to visible images without annotation, and vice versa. USVI-ReID is a challenging yet under-explored task. Most existing methods address the USVI-ReID problem using cluster-based contrastive learning, which simply employs the cluster center as a representation of a person. However, the cluster center primarily focuses on shared information, overlooking disparity. To address the problem, we propose a Progressive Contrastive Learning with Multi-Prototype (PCLMP) method for USVI-ReID. In brief, we first generate the hard prototype by selecting the sample with the maximum distance from the cluster center. This hard prototype is used in the contrastive loss to emphasize disparity. Additionally, instead of rigidly aligning query images to a specific prototype, we generate the dynamic prototype by randomly picking samples within a cluster. This dynamic prototype is used to retain the natural variety of features while reducing instability in the simultaneous learning of both common and disparate information. Finally, we introduce a progressive learning strategy to gradually shift the model's attention towards hard samples, avoiding cluster deterioration. Extensive experiments conducted on the publicly available SYSU-MM01 and RegDB datasets validate the effectiveness of the proposed method. PCLMP outperforms the existing state-of-the-art method with an average mAP improvement of 3.9%. The source codes will be released.
Authors: ShengYun Peng, Aishwarya Chakravarthy, Seongmin Lee, Xiaojing Wang, Rajarajeswari Balasubramaniyan, Duen Horng Chau
Abstract: Tables convey factual and quantitative data with implicit conventions created by humans that are often challenging for machines to parse. Prior work on table recognition (TR) has mainly centered around complex task-specific combinations of available inputs and tools. We present UniTable, a training framework that unifies both the training paradigm and training objective of TR. Its training paradigm combines the simplicity of purely pixel-level inputs with the effectiveness and scalability empowered by self-supervised pretraining from diverse unannotated tabular images. Our framework unifies the training objectives of all three TR tasks - extracting table structure, cell content, and cell bounding box - into a unified task-agnostic training objective: language modeling. Extensive quantitative and qualitative analyses highlight UniTable's state-of-the-art (SOTA) performance on four of the largest TR datasets. UniTable's table parsing capability has surpassed both existing TR methods and general large vision-language models, e.g., GPT-4o, GPT-4-turbo with vision, and LLaVA. Our code is publicly available at https://github.com/poloclub/unitable, featuring a Jupyter Notebook that includes the complete inference pipeline, fine-tuned across multiple TR datasets, supporting all three TR tasks.
Authors: Qihao Zhao, Yalun Dai, Hao Li, Wei Hu, Fan Zhang, Jun Liu
Abstract: Long-tail recognition is challenging because it requires the model to learn good representations from tail categories and address imbalances across all categories. In this paper, we propose a novel generative and fine-tuning framework, LTGC, to handle long-tail recognition via leveraging generated content. Firstly, inspired by the rich implicit knowledge in large-scale models (e.g., large language models, LLMs), LTGC leverages the power of these models to parse and reason over the original tail data to produce diverse tail-class content. We then propose several novel designs for LTGC to ensure the quality of the generated data and to efficiently fine-tune the model using both the generated and original data. The visualization demonstrates the effectiveness of the generation module in LTGC, which produces accurate and diverse tail data. Additionally, the experimental results demonstrate that our LTGC outperforms existing state-of-the-art methods on popular long-tailed benchmarks.
Authors: Sana Ayromlou, Arash Afkanpour, Vahid Reza Khazaie, Fereshteh Forghani
Abstract: The rapid advancement in self-supervised learning (SSL) has highlighted its potential to leverage unlabeled data for learning rich visual representations. However, the existing SSL techniques, particularly those employing different augmentations of the same image, often rely on a limited set of simple transformations that are not representative of real-world data variations. This constrains the diversity and quality of samples, which leads to sub-optimal representations. In this paper, we introduce a novel framework that enriches the SSL paradigm by utilizing generative models to produce semantically consistent image augmentations. By directly conditioning generative models on a source image representation, our method enables the generation of diverse augmentations while maintaining the semantics of the source image, thus offering a richer set of data for self-supervised learning. Our extensive experimental results on various SSL methods demonstrate that our framework significantly enhances the quality of learned visual representations by up to 10\% Top-1 accuracy in downstream tasks. This research demonstrates that incorporating generative models into the SSL workflow opens new avenues for exploring the potential of synthetic data. This development paves the way for more robust and versatile representation learning techniques.
Authors: Patrick Knab, Sascha Marton, Christian Bartelt
Abstract: Explainable Artificial Intelligence is critical in unraveling decision-making processes in complex machine learning models. LIME (Local Interpretable Model-agnostic Explanations) is a well-known XAI framework for image analysis. It utilizes image segmentation to create features to identify relevant areas for classification. Consequently, poor segmentation can compromise the consistency of the explanation and undermine the importance of the segments, affecting the overall interpretability. Addressing these challenges, we introduce DSEG-LIME (Data-Driven Segmentation LIME), featuring: i) a data-driven segmentation for human-recognized feature generation, and ii) a hierarchical segmentation procedure through composition. We benchmark DSEG-LIME on pre-trained models with images from the ImageNet dataset - scenarios without domain-specific knowledge. The analysis includes a quantitative evaluation using established XAI metrics, complemented by a qualitative assessment through a user study. Our findings demonstrate that DSEG outperforms in most of the XAI metrics and enhances the alignment of explanations with human-recognized concepts, significantly improving interpretability. The code is available under: https://github. com/patrick-knab/DSEG-LIME
URLs: https://github.
Authors: Qing Lin, Jingfeng Zhang, Yew Soon Ong, Mengmi Zhang
Abstract: Despite the rapid progress in image generation, emotional image editing remains under-explored. The semantics, context, and structure of an image can evoke emotional responses, making emotional image editing techniques valuable for various real-world applications, including treatment of psychological disorders, commercialization of products, and artistic design. For the first time, we present a novel challenge of emotion-evoked image generation, aiming to synthesize images that evoke target emotions while retaining the semantics and structures of the original scenes. To address this challenge, we propose a diffusion model capable of effectively understanding and editing source images to convey desired emotions and sentiments. Moreover, due to the lack of emotion editing datasets, we provide a unique dataset consisting of 340,000 pairs of images and their emotion annotations. Furthermore, we conduct human psychophysics experiments and introduce four new evaluation metrics to systematically benchmark all the methods. Experimental results demonstrate that our method surpasses all competitive baselines. Our diffusion model is capable of identifying emotional cues from original images, editing images that elicit desired emotions, and meanwhile, preserving the semantic structure of the original images. All code, model, and dataset will be made public.
Authors: Kang Chen, Shiyan Chen, Jiyuan Zhang, Baoyue Zhang, Yajing Zheng, Tiejun Huang, Zhaofei Yu
Abstract: Reconstructing a sequence of sharp images from the blurry input is crucial for enhancing our insights into the captured scene and poses a significant challenge due to the limited temporal features embedded in the image. Spike cameras, sampling at rates up to 40,000 Hz, have proven effective in capturing motion features and beneficial for solving this ill-posed problem. Nonetheless, existing methods fall into the supervised learning paradigm, which suffers from notable performance degradation when applied to real-world scenarios that diverge from the synthetic training data domain. Moreover, the quality of reconstructed images is capped by the generated images based on motion analysis interpolation, which inherently differs from the actual scene, affecting the generalization ability of these methods in real high-speed scenarios. To address these challenges, we propose the first self-supervised framework for the task of spike-guided motion deblurring. Our approach begins with the formulation of a spike-guided deblurring model that explores the theoretical relationships among spike streams, blurry images, and their corresponding sharp sequences. We subsequently develop a self-supervised cascaded framework to alleviate the issues of spike noise and spatial-resolution mismatching encountered in the deblurring model. With knowledge distillation and re-blurring loss, we further design a lightweight deblur network to generate high-quality sequences with brightness and texture consistency with the original input. Quantitative and qualitative experiments conducted on our real-world and synthetic datasets with spikes validate the superior generalization of the proposed framework. Our code, data and trained models will be available at \url{https://github.com/chenkang455/S-SDM}.
Authors: Hyuck Lee, Heeyoung Kim
Abstract: Pseudo-label-based semi-supervised learning (SSL) algorithms trained on a class-imbalanced set face two cascading challenges: 1) Classifiers tend to be biased towards majority classes, and 2) Biased pseudo-labels are used for training. It is difficult to appropriately re-balance the classifiers in SSL because the class distribution of an unlabeled set is often unknown and could be mismatched with that of a labeled set. We propose a novel class-imbalanced SSL algorithm called class-distribution-mismatch-aware debiasing (CDMAD). For each iteration of training, CDMAD first assesses the classifier's biased degree towards each class by calculating the logits on an image without any patterns (e.g., solid color image), which can be considered irrelevant to the training set. CDMAD then refines biased pseudo-labels of the base SSL algorithm by ensuring the classifier's neutrality. CDMAD uses these refined pseudo-labels during the training of the base SSL algorithm to improve the quality of the representations. In the test phase, CDMAD similarly refines biased class predictions on test samples. CDMAD can be seen as an extension of post-hoc logit adjustment to address a challenge of incorporating the unknown class distribution of the unlabeled set for re-balancing the biased classifier under class distribution mismatch. CDMAD ensures Fisher consistency for the balanced error. Extensive experiments verify the effectiveness of CDMAD.
Authors: Dongmin Park, Zhaofang Qian, Guangxing Han, Ser-Nam Lim
Abstract: Mitigating hallucinations of Large Vision Language Models,(LVLMs) is crucial to enhance their reliability for general-purpose assistants. This paper shows that such hallucinations of LVLMs can be significantly exacerbated by preceding user-system dialogues. To precisely measure this, we first present an evaluation benchmark by extending popular multi-modal benchmark datasets with prepended hallucinatory dialogues powered by our novel Adversarial Question Generator (AQG), which can automatically generate image-related yet adversarial dialogues by adopting adversarial attacks on LVLMs. On our benchmark, the zero-shot performance of state-of-the-art LVLMs drops significantly for both the VQA and Captioning tasks. Next, we further reveal this hallucination is mainly due to the prediction bias toward preceding dialogues rather than visual content. To reduce this bias, we propose Adversarial Instruction Tuning (AIT) that robustly fine-tunes LVLMs against hallucinatory dialogues. Extensive experiments show our proposed approach successfully reduces dialogue hallucination while maintaining performance.
Authors: Yifan Wu, Jiawei Du, Ping Liu, Yuewei Lin, Wenqing Cheng, Wei Xu
Abstract: Dataset distillation is an advanced technique aimed at compressing datasets into significantly smaller counterparts, while preserving formidable training performance. Significant efforts have been devoted to promote evaluation accuracy under limited compression ratio while overlooked the robustness of distilled dataset. In this work, we introduce a comprehensive benchmark that, to the best of our knowledge, is the most extensive to date for evaluating the adversarial robustness of distilled datasets in a unified way. Our benchmark significantly expands upon prior efforts by incorporating a wider range of dataset distillation methods, including the latest advancements such as TESLA and SRe2L, a diverse array of adversarial attack methods, and evaluations across a broader and more extensive collection of datasets such as ImageNet-1K. Moreover, we assessed the robustness of these distilled datasets against representative adversarial attack algorithms like PGD and AutoAttack, while exploring their resilience from a frequency perspective. We also discovered that incorporating distilled data into the training batches of the original dataset can yield to improvement of robustness.
Authors: Lei Jiang, Weixin Yang, Xin Zhang, Hao Ni
Abstract: Skeleton-based action recognition (SAR) in videos is an important but challenging task in computer vision. The recent state-of-the-art (SOTA) models for SAR are primarily based on graph convolutional neural networks (GCNs), which are powerful in extracting the spatial information of skeleton data. However, it is yet clear that such GCN-based models can effectively capture the temporal dynamics of human action sequences. To this end, we propose the G-Dev layer, which exploits the path development -- a principled and parsimonious representation for sequential data by leveraging the Lie group structure. By integrating the G-Dev layer, the hybrid G-DevLSTM module enhances the traditional LSTM to reduce the time dimension while retaining high-frequency information. It can be conveniently applied to any temporal graph data, complementing existing advanced GCN-based models. Our empirical studies on the NTU60, NTU120 and Chalearn2013 datasets demonstrate that our proposed GCN-DevLSTM network consistently improves the strong GCN baseline models and achieves SOTA results with superior robustness in SAR tasks. The code is available at https://github.com/DeepIntoStreams/GCN-DevLSTM.
Authors: Hao Ai, Lin Wang
Abstract: 360 depth estimation has recently received great attention for 3D reconstruction owing to its omnidirectional field of view (FoV). Recent approaches are predominantly focused on cross-projection fusion with geometry-based re-projection: they fuse 360 images with equirectangular projection (ERP) and another projection type, e.g., cubemap projection to estimate depth with the ERP format. However, these methods suffer from 1) limited local receptive fields, making it hardly possible to capture large FoV scenes, and 2) prohibitive computational cost, caused by the complex cross-projection fusion module design. In this paper, we propose Elite360D, a novel framework that inputs the ERP image and icosahedron projection (ICOSAP) point set, which is undistorted and spatially continuous. Elite360D is superior in its capacity in learning a representation from a local-with-global perspective. With a flexible ERP image encoder, it includes an ICOSAP point encoder, and a Bi-projection Bi-attention Fusion (B2F) module (totally ~1M parameters). Specifically, the ERP image encoder can take various perspective image-trained backbones (e.g., ResNet, Transformer) to extract local features. The point encoder extracts the global features from the ICOSAP. Then, the B2F module captures the semantic- and distance-aware dependencies between each pixel of the ERP feature and the entire ICOSAP feature set. Without specific backbone design and obvious computational cost increase, Elite360D outperforms the prior arts on several benchmark datasets.
Authors: Yuqi Yang, Peng-Tao Jiang, Qibin Hou, Hao Zhang, Jinwei Chen, Bo Li
Abstract: Previous multi-task dense prediction methods based on the Mixture of Experts (MoE) have received great performance but they neglect the importance of explicitly modeling the global relations among all tasks. In this paper, we present a novel decoder-focused method for multi-task dense prediction, called Mixture-of-Low-Rank-Experts (MLoRE). To model the global task relationships, MLoRE adds a generic convolution path to the original MoE structure, where each task feature can go through this path for explicit parameter sharing. Furthermore, to control the parameters and computational cost brought by the increase in the number of experts, we take inspiration from LoRA and propose to leverage the low-rank format of a vanilla convolution in the expert network. Since the low-rank experts have fewer parameters and can be dynamically parameterized into the generic convolution, the parameters and computational cost do not change much with the increase of experts. Benefiting from this design, we increase the number of experts and its reception field to enlarge the representation capacity, facilitating multiple dense tasks learning in a unified network. Extensive experiments on the PASCAL-Context and NYUD-v2 benchmarks show that our MLoRE achieves superior performance compared to previous state-of-the-art methods on all metrics. Our code is available at https://github.com/YuqiYang213/MLoRE.
Authors: Qiuhong Shen, Zike Wu, Xuanyu Yi, Pan Zhou, Hanwang Zhang, Shuicheng Yan, Xinchao Wang
Abstract: We tackle the challenge of efficiently reconstructing a 3D asset from a single image at millisecond speed. Existing methods for single-image 3D reconstruction are primarily based on Score Distillation Sampling (SDS) with Neural 3D representations. Despite promising results, these approaches encounter practical limitations due to lengthy optimizations and significant memory consumption. In this work, we introduce Gamba, an end-to-end 3D reconstruction model from a single-view image, emphasizing two main insights: (1) Efficient Backbone Design: introducing a Mamba-based GambaFormer network to model 3D Gaussian Splatting (3DGS) reconstruction as sequential prediction with linear scalability of token length, thereby accommodating a substantial number of Gaussians; (2) Robust Gaussian Constraints: deriving radial mask constraints from multi-view masks to eliminate the need for warmup supervision of 3D point clouds in training. We trained Gamba on Objaverse and assessed it against existing optimization-based and feed-forward 3D reconstruction approaches on the GSO Dataset, among which Gamba is the only end-to-end trained single-view reconstruction model with 3DGS. Experimental results demonstrate its competitive generation capabilities both qualitatively and quantitatively and highlight its remarkable speed: Gamba completes reconstruction within 0.05 seconds on a single NVIDIA A100 GPU, which is about $1,000\times$ faster than optimization-based methods. Please see our project page at https://florinshen.github.io/gamba-project.
Authors: Ganlin Zhang, Erik Sandstr\"om, Youmin Zhang, Manthan Patel, Luc Van Gool, Martin R. Oswald
Abstract: Recent advancements in RGB-only dense Simultaneous Localization and Mapping (SLAM) have predominantly utilized grid-based neural implicit encodings and/or struggle to efficiently realize global map and pose consistency. To this end, we propose an efficient RGB-only dense SLAM system using a flexible neural point cloud scene representation that adapts to keyframe poses and depth updates, without needing costly backpropagation. Another critical challenge of RGB-only SLAM is the lack of geometric priors. To alleviate this issue, with the aid of a monocular depth estimator, we introduce a novel DSPO layer for bundle adjustment which optimizes the pose and depth of keyframes along with the scale of the monocular depth. Finally, our system benefits from loop closure and online global bundle adjustment and performs either better or competitive to existing dense neural RGB SLAM methods in tracking, mapping and rendering accuracy on the Replica, TUM-RGBD and ScanNet datasets. The source code is available at https://github.com/zhangganlin/GlOIRE-SLAM
Authors: Yassir Bendou, Giulia Lioi, Bastien Pasdeloup, Lukas Mauch, Ghouthi Boukli Hacene, Fabien Cardinaux, Vincent Gripon
Abstract: We consider the problem of zero-shot one-class visual classification, extending traditional one-class classification to scenarios where only the label of the target class is available. This method aims to discriminate between positive and negative query samples without requiring examples from the target class. We propose a two-step solution that first queries large language models for visually confusing objects and then relies on vision-language pre-trained models (e.g., CLIP) to perform classification. By adapting large-scale vision benchmarks, we demonstrate the ability of the proposed method to outperform adapted off-the-shelf alternatives in this setting. Namely, we propose a realistic benchmark where negative query samples are drawn from the same original dataset as positive ones, including a granularity-controlled version of iNaturalist, where negative samples are at a fixed distance in the taxonomy tree from the positive ones. To our knowledge, we are the first to demonstrate the ability to discriminate a single category from other semantically related ones using only its label.
Authors: Zhiyuan Cheng, Zhaoyi Liu, Tengda Guo, Shiwei Feng, Dongfang Liu, Mingjie Tang, Xiangyu Zhang
Abstract: Pixel-wise regression tasks (e.g., monocular depth estimation (MDE) and optical flow estimation (OFE)) have been widely involved in our daily life in applications like autonomous driving, augmented reality and video composition. Although certain applications are security-critical or bear societal significance, the adversarial robustness of such models are not sufficiently studied, especially in the black-box scenario. In this work, we introduce the first unified black-box adversarial patch attack framework against pixel-wise regression tasks, aiming to identify the vulnerabilities of these models under query-based black-box attacks. We propose a novel square-based adversarial patch optimization framework and employ probabilistic square sampling and score-based gradient estimation techniques to generate the patch effectively and efficiently, overcoming the scalability problem of previous black-box patch attacks. Our attack prototype, named BadPart, is evaluated on both MDE and OFE tasks, utilizing a total of 7 models. BadPart surpasses 3 baseline methods in terms of both attack performance and efficiency. We also apply BadPart on the Google online service for portrait depth estimation, causing 43.5% relative distance error with 50K queries. State-of-the-art (SOTA) countermeasures cannot defend our attack effectively.
Authors: Ruowen Zhao, Zhengyi Wang, Yikai Wang, Zihan Zhou, Jun Zhu
Abstract: 3D content generation has wide applications in various fields. One of its dominant paradigms is by sparse-view reconstruction using multi-view images generated by diffusion models. However, since directly reconstructing triangle meshes from multi-view images is challenging, most methodologies opt to an implicit representation (such as NeRF) during the sparse-view reconstruction and acquire the target mesh by a post-processing extraction. However, the implicit representation takes extensive time to train and the post-extraction also leads to undesirable visual artifacts. In this paper, we propose FlexiDreamer, a novel framework that directly reconstructs high-quality meshes from multi-view generated images. We utilize an advanced gradient-based mesh optimization, namely FlexiCubes, for multi-view mesh reconstruction, which enables us to generate 3D meshes in an end-to-end manner. To address the reconstruction artifacts owing to the inconsistencies from generated images, we design a hybrid positional encoding scheme to improve the reconstruction geometry and an orientation-aware texture mapping to mitigate surface ghosting. To further enhance the results, we respectively incorporate eikonal and smooth regularizations to reduce geometric holes and surface noise. Our approach can generate high-fidelity 3D meshes in the single image-to-3D downstream task with approximately 1 minute, significantly outperforming previous methods.
Authors: Maxim Nikolaev, Mikhail Kuznetsov, Dmitry Vetrov, Aibek Alanov
Abstract: Our paper addresses the complex task of transferring a hairstyle from a reference image to an input photo for virtual hair try-on. This task is challenging due to the need to adapt to various photo poses, the sensitivity of hairstyles, and the lack of objective metrics. The current state of the art hairstyle transfer methods use an optimization process for different parts of the approach, making them inexcusably slow. At the same time, faster encoder-based models are of very low quality because they either operate in StyleGAN's W+ space or use other low-dimensional image generators. Additionally, both approaches have a problem with hairstyle transfer when the source pose is very different from the target pose, because they either don't consider the pose at all or deal with it inefficiently. In our paper, we present the HairFast model, which uniquely solves these problems and achieves high resolution, near real-time performance, and superior reconstruction compared to optimization problem-based methods. Our solution includes a new architecture operating in the FS latent space of StyleGAN, an enhanced inpainting approach, and improved encoders for better alignment, color transfer, and a new encoder for post-processing. The effectiveness of our approach is demonstrated on realism metrics after random hairstyle transfer and reconstruction when the original hairstyle is transferred. In the most difficult scenario of transferring both shape and color of a hairstyle from different images, our method performs in less than a second on the Nvidia V100. Our code is available at https://github.com/AIRI-Institute/HairFastGAN.
Authors: Xingyu Zheng, Haotong Qin, Xudong Ma, Mingyuan Zhang, Haojie Hao, Jiakai Wang, Zixiang Zhao, Jinyang Guo, Xianglong Liu
Abstract: With the advancement of diffusion models (DMs) and the substantially increased computational requirements, quantization emerges as a practical solution to obtain compact and efficient low-bit DMs. However, the highly discrete representation leads to severe accuracy degradation, hindering the quantization of diffusion models to ultra-low bit-widths. This paper proposes a novel quantization-aware training approach for DMs, namely BinaryDM. The proposed method pushes DMs' weights toward accurate and efficient binarization, considering the representation and computation properties. From the representation perspective, we present a Learnable Multi-basis Binarizer (LMB) to recover the representations generated by the binarized DM. The LMB enhances detailed information through the flexible combination of dual binary bases while applying to parameter-sparse locations of DM architectures to achieve minor burdens. From the optimization perspective, a Low-rank Representation Mimicking (LRM) is applied to assist the optimization of binarized DMs. The LRM mimics the representations of full-precision DMs in low-rank space, alleviating the direction ambiguity of the optimization process caused by fine-grained alignment. Moreover, a quick progressive warm-up is applied to BinaryDM, avoiding convergence difficulties by layerwisely progressive quantization at the beginning of training. Comprehensive experiments demonstrate that BinaryDM achieves significant accuracy and efficiency gains compared to SOTA quantization methods of DMs under ultra-low bit-widths. With 1.1-bit weight and 4-bit activation (W1.1A4), BinaryDM achieves as low as 7.11 FID and saves the performance from collapse (baseline FID 39.69). As the first binarization method for diffusion models, W1.1A4 BinaryDM achieves impressive 9.3 times OPs and 24.8 times model size savings, showcasing its substantial potential for edge deployment.
Authors: Jiahao Wang, Wenqi Shao, Mengzhao Chen, Chengyue Wu, Yong Liu, Taiqiang Wu, Kaipeng Zhang, Songyang Zhang, Kai Chen, Ping Luo
Abstract: This work examines whether decoder-only Transformers such as LLaMA, which were originally designed for large language models (LLMs), can be adapted to the computer vision field. We first "LLaMAfy" a standard ViT step-by-step to align with LLaMA's architecture, and find that directly applying a causal mask to the self-attention brings an attention collapse issue, resulting in the failure to the network training. We suggest to reposition the class token behind the image tokens with a post-sequence class token technique to overcome this challenge, enabling causal self-attention to efficiently capture the entire image's information. Additionally, we develop a soft mask strategy that gradually introduces a causal mask to the self-attention at the onset of training to facilitate the optimization behavior. The tailored model, dubbed as image LLaMA (iLLaMA), is akin to LLaMA in architecture and enables direct supervised learning. Its causal self-attention boosts computational efficiency and learns complex representation by elevating attention map ranks. iLLaMA rivals the performance with its encoder-only counterparts, achieving 75.1% ImageNet top-1 accuracy with only 5.7M parameters. Scaling the model to $\sim$310M and pre-training on ImageNet-21K further enhances the accuracy to 86.0%. Extensive experiments demonstrate iLLaMA's reliable properties: shape-texture bias, calibration, quantization compatibility, ADE20K segmentation and CIFAR transfer learning. We hope our study can kindle fresh views to visual architectures in the wave of LLMs and inspire the development of unified multimodal models. Pre-trained models and codes are available https://github.com/techmonsterwang/iLLaMA.
Authors: Zhenwei Wang, Qiule Sun, Bingbing Zhang, Pengfei Wang, Jianxin Zhang, Qiang Zhang
Abstract: Few-shot learning has been successfully applied to medical image classification as only very few medical examples are available for training. Due to the challenging problem of limited number of annotated medical images, image representations should not be solely derived from a single image modality which is insufficient for characterizing concept classes. In this paper, we propose a new prompting multi-modal model paradigm on medical image classification based on multi-modal foundation models, called PM2. Besides image modality,PM2 introduces another supplementary text input, known as prompt, to further describe corresponding image or concept classes and facilitate few-shot learning across diverse modalities. To better explore the potential of prompt engineering, we empirically investigate five distinct prompt schemes under the new paradigm. Furthermore, linear probing in multi-modal models acts as a linear classification head taking as input only class token, which ignores completely merits of rich statistics inherent in high-level visual tokens. Thus, we alternatively perform a linear classification on feature distribution of visual tokens and class token simultaneously. To effectively mine such rich statistics, a global covariance pooling with efficient matrix power normalization is used to aggregate visual tokens. Then we study and combine two classification heads. One is shared for class token of image from vision encoder and prompt representation encoded by text encoder. The other is to classification on feature distribution of visual tokens from vision encoder. Extensive experiments on three medical datasets show that our PM2 significantly outperforms counterparts regardless of prompt schemes and achieves state-of-the-art performance.
Authors: Song Xia, Yu Yi, Xudong Jiang, Henghui Ding
Abstract: Randomized Smoothing (RS) has been proven a promising method for endowing an arbitrary image classifier with certified robustness. However, the substantial uncertainty inherent in the high-dimensional isotropic Gaussian noise imposes the curse of dimensionality on RS. Specifically, the upper bound of ${\ell_2}$ certified robustness radius provided by RS exhibits a diminishing trend with the expansion of the input dimension $d$, proportionally decreasing at a rate of $1/\sqrt{d}$. This paper explores the feasibility of providing ${\ell_2}$ certified robustness for high-dimensional input through the utilization of dual smoothing in the lower-dimensional space. The proposed Dual Randomized Smoothing (DRS) down-samples the input image into two sub-images and smooths the two sub-images in lower dimensions. Theoretically, we prove that DRS guarantees a tight ${\ell_2}$ certified robustness radius for the original input and reveal that DRS attains a superior upper bound on the ${\ell_2}$ robustness radius, which decreases proportionally at a rate of $(1/\sqrt m + 1/\sqrt n )$ with $m+n=d$. Extensive experiments demonstrate the generalizability and effectiveness of DRS, which exhibits a notable capability to integrate with established methodologies, yielding substantial improvements in both accuracy and ${\ell_2}$ certified robustness baselines of RS on the CIFAR-10 and ImageNet datasets. Code is available at https://github.com/xiasong0501/DRS.
Authors: Jin Gao, Shubo Lin, Shaoru Wang, Yutong Kou, Zeming Li, Liang Li, Congxuan Zhang, Xiaoqin Zhang, Yizheng Wang, Weiming Hu
Abstract: Masked image modeling (MIM) pre-training for large-scale vision transformers (ViTs) has enabled promising downstream performance on top of the learned self-supervised ViT features. In this paper, we question if the \textit{extremely simple} lightweight ViTs' fine-tuning performance can also benefit from this pre-training paradigm, which is considerably less studied yet in contrast to the well-established lightweight architecture design methodology. We use an observation-analysis-solution flow for our study. We first systematically observe different behaviors among the evaluated pre-training methods with respect to the downstream fine-tuning data scales. Furthermore, we analyze the layer representation similarities and attention maps across the obtained models, which clearly show the inferior learning of MIM pre-training on higher layers, leading to unsatisfactory transfer performance on data-insufficient downstream tasks. This finding is naturally a guide to designing our distillation strategies during pre-training to solve the above deterioration problem. Extensive experiments have demonstrated the effectiveness of our approach. Our pre-training with distillation on pure lightweight ViTs with vanilla/hierarchical design ($5.7M$/$6.5M$) can achieve $79.4\%$/$78.9\%$ top-1 accuracy on ImageNet-1K. It also enables SOTA performance on the ADE20K segmentation task ($42.8\%$ mIoU) and LaSOT tracking task ($66.1\%$ AUC) in the lightweight regime. The latter even surpasses all the current SOTA lightweight CPU-realtime trackers.
Authors: Ziqi Xie, Weidong Zhao, Xianhui Liu, Jian Zhao, Ning Jia
Abstract: Deep learning-based image stitching pipelines are typically divided into three cascading stages: registration, fusion, and rectangling. Each stage requires its own network training and is tightly coupled to the others, leading to error propagation and posing significant challenges to parameter tuning and system stability. This paper proposes the Simple and Robust Stitcher (SRStitcher), which revolutionizes the image stitching pipeline by simplifying the fusion and rectangling stages into a unified inpainting model, requiring no model training or fine-tuning. We reformulate the problem definitions of the fusion and rectangling stages and demonstrate that they can be effectively integrated into an inpainting task. Furthermore, we design the weighted masks to guide the reverse process in a pre-trained largescale diffusion model, implementing this integrated inpainting task in a single inference. Through extensive experimentation, we verify the interpretability and generalization capabilities of this unified model, demonstrating that SRStitcher outperforms state-of-the-art methods in both performance and stability. Code: https://github.com/yayoyo66/SRStitcher
Authors: David Montero, Miguel D. Mahecha, C\'esar Aybar, Clemens Mosig, Sebastian Wieneke
Abstract: The Sentinel-2 (S2) mission from the European Space Agency's Copernicus program provides essential data for Earth surface analysis. Its Level-2A products deliver high-to-medium resolution (10-60 m) surface reflectance (SR) data through the MultiSpectral Instrument (MSI). To enhance the accuracy and comparability of SR data, adjustments simulating a nadir viewing perspective are essential. These corrections address the anisotropic nature of SR and the variability in sun and observation angles, ensuring consistent image comparisons over time and under different conditions. The $c$-factor method, a simple yet effective algorithm, adjusts observed S2 SR by using the MODIS BRDF model to achieve Nadir BRDF Adjusted Reflectance (NBAR). Despite the straightforward application of the $c$-factor to individual images, a cohesive Python framework for its application across multiple S2 images and Earth System Data Cubes (ESDCs) from cloud-stored data has been lacking. Here we introduce sen2nbar, a Python package crafted to convert S2 SR data to NBAR, supporting both individual images and ESDCs derived from cloud-stored data. This package simplifies the conversion of S2 SR data to NBAR via a single function, organized into modules for efficient process management. By facilitating NBAR conversion for both SAFE files and ESDCs from SpatioTemporal Asset Catalogs (STAC), sen2nbar is developed as a flexible tool that can handle diverse data format requirements. We anticipate that sen2nbar will considerably contribute to the standardization and harmonization of S2 data, offering a robust solution for a diverse range of users across various applications. sen2nbar is an open-source tool available at https://github.com/ESDS-Leipzig/sen2nbar.
Authors: Nagabhushan Somraj, Sai Harsha Mupparaju, Adithyan Karanayil, Rajiv Soundararajan
Abstract: Neural Radiance Fields (NeRF) show impressive performance in photo-realistic free-view rendering of scenes. Recent improvements on the NeRF such as TensoRF and ZipNeRF employ explicit models for faster optimization and rendering, as compared to the NeRF that employs an implicit representation. However, both implicit and explicit radiance fields require dense sampling of images in the given scene. Their performance degrades significantly when only a sparse set of views is available. Researchers find that supervising the depth estimated by a radiance field helps train it effectively with fewer views. The depth supervision is obtained either using classical approaches or neural networks pre-trained on a large dataset. While the former may provide only sparse supervision, the latter may suffer from generalization issues. As opposed to the earlier approaches, we seek to learn the depth supervision by designing augmented models and training them along with the main radiance field. Further, we aim to design a framework of regularizations that can work across different implicit and explicit radiance fields. We observe that certain features of these radiance field models overfit to the observed images in the sparse-input scenario. Our key finding is that reducing the capability of the radiance fields with respect to positional encoding, the number of decomposed tensor components or the size of the hash table, constrains the model to learn simpler solutions, which estimate better depth in certain regions. By designing augmented models based on such reduced capabilities, we obtain better depth supervision for the main radiance field. We achieve state-of-the-art view-synthesis performance with sparse input views on popular datasets containing forward-facing and 360$^\circ$ scenes by employing the above regularizations.
Authors: Zhigang Sun, Zixu Wang, Lavdim Halilaj, Juergen Luettin
Abstract: Trajectory prediction in autonomous driving relies on accurate representation of all relevant contexts of the driving scene, including traffic participants, road topology, traffic signs, as well as their semantic relations to each other. Despite increased attention to this issue, most approaches in trajectory prediction do not consider all of these factors sufficiently. We present SemanticFormer, an approach for predicting multimodal trajectories by reasoning over a semantic traffic scene graph using a hybrid approach. It utilizes high-level information in the form of meta-paths, i.e. trajectories on which an agent is allowed to drive from a knowledge graph which is then processed by a novel pipeline based on multiple attention mechanisms to predict accurate trajectories. SemanticFormer comprises a hierarchical heterogeneous graph encoder to capture spatio-temporal and relational information across agents as well as between agents and road elements. Further, it includes a predictor to fuse different encodings and decode trajectories with probabilities. Finally, a refinement module assesses permitted meta-paths of trajectories and speed profiles to obtain final predicted trajectories. Evaluation of the nuScenes benchmark demonstrates improved performance compared to several SOTA methods. In addition, we demonstrate that our knowledge graph can be easily added to two graph-based existing SOTA methods, namely VectorNet and Laformer, replacing their original homogeneous graphs. The evaluation results suggest that by adding our knowledge graph the performance of the original methods is enhanced by 5% and 4%, respectively.
Authors: Yuekun Dai, Dafeng Zhang, Xiaoming Li, Zongsheng Yue, Chongyi Li, Shangchen Zhou, Ruicheng Feng, Peiqing Yang, Zhezhu Jin, Guanqun Liu, Chen Change Loy, Lize Zhang, Shuai Liu, Chaoyu Feng, Luyang Wang, Shuan Chen, Guangqi Shao, Xiaotao Wang, Lei Lei, Qirui Yang, Qihua Cheng, Zhiqiang Xu, Yihao Liu, Huanjing Yue, Jingyu Yang, Florin-Alexandru Vasluianu, Zongwei Wu, George Ciubotariu, Radu Timofte, Zhao Zhang, Suiyi Zhao, Bo Wang, Zhichao Zuo, Yanyan Wei, Kuppa Sai Sri Teja, Jayakar Reddy A, Girish Rongali, Kaushik Mitra, Zhihao Ma, Yongxu Liu, Wanying Zhang, Wei Shang, Yuhong He, Long Peng, Zhongxin Yu, Shaofei Luo, Jian Wang, Yuqi Miao, Baiang Li, Gang Wei, Rakshank Verma, Ritik Maheshwari, Rahul Tekchandani, Praful Hambarde, Satya Narayan Tazi, Santosh Kumar Vipparthi, Subrahmanyam Murala, Haopeng Zhang, Yingli Hou, Mingde Yao, Levin M S, Aniruth Sundararajan, Hari Kumar A
Abstract: The increasing demand for computational photography and imaging on mobile platforms has led to the widespread development and integration of advanced image sensors with novel algorithms in camera systems. However, the scarcity of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). Building on the achievements of the previous MIPI Workshops held at ECCV 2022 and CVPR 2023, we introduce our third MIPI challenge including three tracks focusing on novel image sensors and imaging algorithms. In this paper, we summarize and review the Nighttime Flare Removal track on MIPI 2024. In total, 170 participants were successfully registered, and 14 teams submitted results in the final testing phase. The developed solutions in this challenge achieved state-of-the-art performance on Nighttime Flare Removal. More details of this challenge and the link to the dataset can be found at https://mipi-challenge.org/MIPI2024/.
Authors: Youngdong Jang, Dong In Lee, MinHyuk Jang, Jong Wook Kim, Feng Yang, Sangpil Kim
Abstract: The advances in the Neural Radiance Fields (NeRF) research offer extensive applications in diverse domains, but protecting their copyrights has not yet been researched in depth. Recently, NeRF watermarking has been considered one of the pivotal solutions for safely deploying NeRF-based 3D representations. However, existing methods are designed to apply only to implicit or explicit NeRF representations. In this work, we introduce an innovative watermarking method that can be employed in both representations of NeRF. This is achieved by fine-tuning NeRF to embed binary messages in the rendering process. In detail, we propose utilizing the discrete wavelet transform in the NeRF space for watermarking. Furthermore, we adopt a deferred back-propagation technique and introduce a combination with the patch-wise loss to improve rendering quality and bit accuracy with minimum trade-offs. We evaluate our method in three different aspects: capacity, invisibility, and robustness of the embedded watermarks in the 2D-rendered images. Our method achieves state-of-the-art performance with faster training speed over the compared state-of-the-art methods.
Authors: Mengxian Hu, Minghao Zhu, Xun Zhou, Qingqing Yan, Shu Li, Chengju Liu, Qijun Chen
Abstract: Motion diffusion models excel at text-driven motion generation but struggle with real-time inference since motion sequences are time-axis redundant and solving reverse diffusion trajectory involves tens or hundreds of sequential iterations. In this paper, we propose a Motion Latent Consistency Training (MLCT) framework, which allows for large-scale skip sampling of compact motion latent representation by constraining the consistency of the outputs of adjacent perturbed states on the precomputed trajectory. In particular, we design a flexible motion autoencoder with quantization constraints to guarantee the low-dimensionality, succinctness, and boundednes of the motion embedding space. We further present a conditionally guided consistency training framework based on conditional trajectory simulation without additional pre-training diffusion model, which significantly improves the conditional generation performance with minimal training cost. Experiments on two benchmarks demonstrate our model's state-of-the-art performance with an 80\% inference cost saving and around 14 ms on a single RTX 4090 GPU.
Authors: Hritik Bansal, Yonatan Bitton, Michal Yarom, Idan Szpektor, Aditya Grover, Kai-Wei Chang
Abstract: Recent advances in diffusion-based generative modeling have led to the development of text-to-video (T2V) models that can generate high-quality videos conditioned on a text prompt. Most of these T2V models often produce single-scene video clips that depict an entity performing a particular action (e.g., 'a red panda climbing a tree'). However, it is pertinent to generate multi-scene videos since they are ubiquitous in the real-world (e.g., 'a red panda climbing a tree' followed by 'the red panda sleeps on the top of the tree'). To generate multi-scene videos from a pretrained T2V model, we introduce Time-Aligned Captions (TALC) framework. Specifically, we enhance the text-conditioning mechanism in the T2V architecture to recognize the temporal alignment between the video scenes and scene descriptions. As a result, we show that the pretrained T2V model can generate multi-scene videos that adhere to the multi-scene text descriptions and be visually consistent (e.g., w.r.t entity and background). Our TALC-finetuned model outperforms the baseline methods on multi-scene video-text data by 15.5 points on aggregated score, averaging visual consistency and text adherence using human evaluation. The project website is https://talc-mst2v.github.io/.
Authors: Jiayi Chen, Chunhua Deng
Abstract: With the advancement of video analysis technology, the multi-object tracking (MOT) problem in complex scenes involving pedestrians is gaining increasing importance. This challenge primarily involves two key tasks: pedestrian detection and re-identification. While significant progress has been achieved in pedestrian detection tasks in recent years, enhancing the effectiveness of re-identification tasks remains a persistent challenge. This difficulty arises from the large total number of pedestrian samples in multi-object tracking datasets and the scarcity of individual instance samples. Motivated by recent rapid advancements in meta-learning techniques, we introduce MAML MOT, a meta-learning-based training approach for multi-object tracking. This approach leverages the rapid learning capability of meta-learning to tackle the issue of sample scarcity in pedestrian re-identification tasks, aiming to improve the model's generalization performance and robustness. Experimental results demonstrate that the proposed method achieves high accuracy on mainstream datasets in the MOT Challenge. This offers new perspectives and solutions for research in the field of pedestrian multi-object tracking.
Authors: David Jin, Harry Zhang, Kai Chang
Abstract: We perform detailed theoretical analysis of an expectation-maximization-based algorithm recently proposed in for solving a variation of the 3D registration problem, named multi-model 3D registration. Despite having shown superior empirical results, did not theoretically justify the conditions under which the EM approach converges to the ground truth. In this project, we aim to close this gap by establishing such conditions. In particular, the analysis revolves around the usage of probabilistic tail bounds that are developed and applied in various instances throughout the course. The problem studied in this project stands as another example, different from those seen in the course, in which tail-bounds help advance our algorithmic understanding in a probabilistic way. We provide self-contained background materials on 3D Registration
Authors: Feiran Li, Qianqian Xu, Shilong Bao, Zhiyong Yang, Runmin Cong, Xiaochun Cao, Qingming Huang
Abstract: This paper explores the size-invariance of evaluation metrics in Salient Object Detection (SOD), especially when multiple targets of diverse sizes co-exist in the same image. We observe that current metrics are size-sensitive, where larger objects are focused, and smaller ones tend to be ignored. We argue that the evaluation should be size-invariant because bias based on size is unjustified without additional semantic information. In pursuit of this, we propose a generic approach that evaluates each salient object separately and then combines the results, effectively alleviating the imbalance. We further develop an optimization framework tailored to this goal, achieving considerable improvements in detecting objects of different sizes. Theoretically, we provide evidence supporting the validity of our new metrics and present the generalization analysis of SOD. Extensive experiments demonstrate the effectiveness of our method. The code is available at https://github.com/Ferry-Li/SI-SOD.
Authors: Bike Chen, Chen Gong, Juha R\"oning
Abstract: Point cloud segmentation (PCS) plays an essential role in robot perception and navigation tasks. To efficiently understand large-scale outdoor point clouds, their range image representation is commonly adopted. This image-like representation is compact and structured, making range image-based PCS models practical. However, undesirable missing values in the range images damage the shapes and patterns of objects. This problem creates difficulty for the models in learning coherent and complete geometric information from the objects. Consequently, the PCS models only achieve inferior performance. Delving deeply into this issue, we find that the use of unreasonable projection approaches and deskewing scans mainly leads to unwanted missing values in the range images. Besides, almost all previous works fail to consider filling in the unexpected missing values in the PCS task. To alleviate this problem, we first propose a new projection method, namely scan unfolding++ (SU++), to avoid massive missing values in the generated range images. Then, we introduce a simple yet effective approach, namely range-dependent $K$-nearest neighbor interpolation ($K$NNI), to further fill in missing values. Finally, we introduce the Filling Missing Values Network (FMVNet) and Fast FMVNet. Extensive experimental results on SemanticKITTI, SemanticPOSS, and nuScenes datasets demonstrate that by employing the proposed SU++ and $K$NNI, existing range image-based PCS models consistently achieve better performance than the baseline models. Besides, both FMVNet and Fast FMVNet achieve state-of-the-art performance in terms of the speed-accuracy trade-off. The proposed methods can be applied to other range image-based tasks and practical applications.
Authors: Junzhang Liu, Zhecan Wang, Hammad Ayyubi, Haoxuan You, Chris Thomas, Rui Sun, Shih-Fu Chang, Kai-Wei Chang
Abstract: Despite the widespread adoption of Vision-Language Understanding (VLU) benchmarks such as VQA v2, OKVQA, A-OKVQA, GQA, VCR, SWAG, and VisualCOMET, our analysis reveals a pervasive issue affecting their integrity: these benchmarks contain samples where answers rely on assumptions unsupported by the provided context. Training models on such data foster biased learning and hallucinations as models tend to make similar unwarranted assumptions. To address this issue, we collect contextual data for each sample whenever available and train a context selection module to facilitate evidence-based model predictions. Strong improvements across multiple benchmarks demonstrate the effectiveness of our approach. Further, we develop a general-purpose Context-AwaRe Abstention (CARA) detector to identify samples lacking sufficient context and enhance model accuracy by abstaining from responding if the required context is absent. CARA exhibits generalization to new benchmarks it wasn't trained on, underscoring its utility for future VLU benchmarks in detecting or cleaning samples with inadequate context. Finally, we curate a Context Ambiguity and Sufficiency Evaluation (CASE) set to benchmark the performance of insufficient context detectors. Overall, our work represents a significant advancement in ensuring that vision-language models generate trustworthy and evidence-based outputs in complex real-world scenarios.
Authors: Mengxi Zhang, Kang Rong
Abstract: Current multimodal Large Language Models (MLLMs) suffer from "hallucination", occasionally generating responses that are not grounded in the input images. To tackle this challenge, one promising path is to utilize reinforcement learning from human feedback (RLHF), which steers MLLMs towards learning superior responses while avoiding inferior ones. We rethink the common practice of using binary preferences (i.e., superior, inferior), and find that adopting multi-level preferences (e.g., superior, medium, inferior) is better for two benefits: 1) It narrows the gap between adjacent levels, thereby encouraging MLLMs to discern subtle differences. 2) It further integrates cross-level comparisons (beyond adjacent-level comparisons), thus providing a broader range of comparisons with hallucination examples. To verify our viewpoint, we present the Automated Multi-level Preference (\textbf{AMP}) framework for MLLMs. To facilitate this framework, we first develop an automated dataset generation pipeline that provides high-quality multi-level preference datasets without any human annotators. Furthermore, we design the Multi-level Direct Preference Optimization (MDPO) algorithm to robustly conduct complex multi-level preference learning. Additionally, we propose a new hallucination benchmark, MRHal-Bench. Extensive experiments across public hallucination and general benchmarks, as well as our MRHal-Bench, demonstrate the effectiveness of our proposed method.
Authors: Duo Peng, Qiuhong Ke, Jun Liu
Abstract: Text-to-Image (T2I) models have raised security concerns due to their potential to generate inappropriate or harmful images. In this paper, we propose UPAM, a novel framework that investigates the robustness of T2I models from the attack perspective. Unlike most existing attack methods that focus on deceiving textual defenses, UPAM aims to deceive both textual and visual defenses in T2I models. UPAM enables gradient-based optimization, offering greater effectiveness and efficiency than previous methods. Given that T2I models might not return results due to defense mechanisms, we introduce a Sphere-Probing Learning (SPL) scheme to support gradient optimization even when no results are returned. Additionally, we devise a Semantic-Enhancing Learning (SEL) scheme to finetune UPAM for generating target-aligned images. Our framework also ensures attack stealthiness. Extensive experiments demonstrate UPAM's effectiveness and efficiency.
Authors: Sebastian Schmidt, Leonard Schenk, Leo Schwinn, Stephan G\"unnemann
Abstract: When applying deep learning models in open-world scenarios, active learning (AL) strategies are crucial for identifying label candidates from a nearly infinite amount of unlabeled data. In this context, robust out-of-distribution (OOD) detection mechanisms are essential for handling data outside the target distribution of the application. However, current works investigate both problems separately. In this work, we introduce SISOM as the first unified solution for both AL and OOD detection. By leveraging feature space distance metrics SISOM combines the strengths of the currently independent tasks to solve both effectively. We conduct extensive experiments showing the problems arising when migrating between both tasks. In these evaluations SISOM underlined its effectiveness by achieving first place in two of the widely used OpenOOD benchmarks and second place in the remaining one. In AL, SISOM outperforms others and delivers top-1 performance in three benchmarks
Authors: Yinghao Huang, Leo Ho, Dafei Qin, Mingyi Shi, Taku Komura
Abstract: We address the problem of accurate capture and expressive modelling of interactive behaviors happening between two persons in daily scenarios. Different from previous works which either only consider one person or focus on conversational gestures, we propose to simultaneously model the activities of two persons, and target objective-driven, dynamic, and coherent interactions which often span long duration. To this end, we capture a new dataset dubbed InterAct, which is composed of 241 motion sequences where two persons perform a realistic scenario over the whole sequence. The audios, body motions, and facial expressions of both persons are all captured in our dataset. We also demonstrate the first diffusion model based approach that directly estimates the interactive motions between two persons from their audios alone. All the data and code will be available at: https://hku-cg.github.io/interact.
Authors: Mengqi Lei, Xin Wang
Abstract: Accurate segmentation of polyps in colonoscopy images is essential for early-stage diagnosis and management of colorectal cancer. Despite advancements in deep learning for polyp segmentation, enduring limitations persist. The edges of polyps are typically ambiguous, making them difficult to discern from the background, and the model performance is often compromised by the influence of irrelevant or unimportant features. To alleviate these challenges, we propose a novel model named Edge-Prioritized Polyp Segmentation (EPPS). Specifically, we incorporate an Edge Mapping Engine (EME) aimed at accurately extracting the edges of polyps. Subsequently, an Edge Information Injector (EII) is devised to augment the mask prediction by injecting the captured edge information into Decoder blocks. Furthermore, we introduce a component called Selective Feature Decoupler (SFD) to suppress the influence of noise and extraneous features on the model. Extensive experiments on 3 widely used polyp segmentation benchmarks demonstrate the superior performance of our method compared with other state-of-the-art approaches.
Authors: Weilian Zhou (Cynthia), Sei-Ichiro Kamata (Cynthia), Haipeng Wang (Cynthia), Man-Sing Wong (Cynthia), Huiying (Cynthia), Hou
Abstract: Hyperspectral image (HSI) classification is pivotal in the remote sensing (RS) field, particularly with the advancement of deep learning techniques. Sequential models, adapted from the natural language processing (NLP) field such as Recurrent Neural Networks (RNNs) and Transformers, have been tailored to this task, offering a unique viewpoint. However, several challenges persist 1) RNNs struggle with centric feature aggregation and are sensitive to interfering pixels, 2) Transformers require significant computational resources and often underperform with limited HSI training samples, and 3) Current scanning methods for converting images into sequence-data are simplistic and inefficient. In response, this study introduces the innovative Mamba-in-Mamba (MiM) architecture for HSI classification, the first attempt of deploying State Space Model (SSM) in this task. The MiM model includes 1) A novel centralized Mamba-Cross-Scan (MCS) mechanism for transforming images into sequence-data, 2) A Tokenized Mamba (T-Mamba) encoder that incorporates a Gaussian Decay Mask (GDM), a Semantic Token Learner (STL), and a Semantic Token Fuser (STF) for enhanced feature generation and concentration, and 3) A Weighted MCS Fusion (WMF) module coupled with a Multi-Scale Loss Design to improve decoding efficiency. Experimental results from three public HSI datasets with fixed and disjoint training-testing samples demonstrate that our method outperforms existing baselines and state-of-the-art approaches, highlighting its efficacy and potential in HSI applications.
Authors: Safouane El Ghazouali, Arnaud Gucciardi, Umberto Michelucci
Abstract: Self-rewarding have emerged recently as a powerful tool in the field of Natural Language Processing (NLP), allowing language models to generate high-quality relevant responses by providing their own rewards during training. This innovative technique addresses the limitations of other methods that rely on human preferences. In this paper, we build upon the concept of self-rewarding models and introduce its vision equivalent for Text-to-Image generative AI models. This approach works by fine-tuning diffusion model on a self-generated self-judged dataset, making the fine-tuning more automated and with better data quality. The proposed mechanism makes use of other pre-trained models such as vocabulary based-object detection, image captioning and is conditioned by the a set of object for which the user might need to improve generated data quality. The approach has been implemented, fine-tuned and evaluated on stable diffusion and has led to a performance that has been evaluated to be at least 60\% better than existing commercial and research Text-to-image models. Additionally, the built self-rewarding mechanism allowed a fully automated generation of images, while increasing the visual quality of the generated images and also more efficient following of prompt instructions. The code used in this work is freely available on https://github.com/safouaneelg/SRT2I.
Authors: Yimeng Shan, Malu Zhang, Rui-jie Zhu, Xuerui Qiu, Jason K. Eshraghian, Haicheng Qu
Abstract: Recent advancements in neuroscience research have propelled the development of Spiking Neural Networks (SNNs), which not only have the potential to further advance neuroscience research but also serve as an energy-efficient alternative to Artificial Neural Networks (ANNs) due to their spike-driven characteristics. However, previous studies often neglected the multiscale information and its spatiotemporal correlation between event data, leading SNN models to approximate each frame of input events as static images. We hypothesize that this oversimplification significantly contributes to the performance gap between SNNs and traditional ANNs. To address this issue, we have designed a Spiking Multiscale Attention (SMA) module that captures multiscale spatiotemporal interaction information. Furthermore, we developed a regularization method named Attention ZoneOut (AZO), which utilizes spatiotemporal attention weights to reduce the model's generalization error through pseudo-ensemble training. Our approach has achieved state-of-the-art results on mainstream neural morphology datasets. Additionally, we have reached a performance of 77.1% on the Imagenet-1K dataset using a 104-layer ResNet architecture enhanced with SMA and AZO. This achievement confirms the state-of-the-art performance of SNNs with non-transformer architectures and underscores the effectiveness of our method in bridging the performance gap between SNN models and traditional ANN models.
Authors: Fangqiang Ding, Xiangyu Wen, Yunzhou Zhu, Yiming Li, Chris Xiaoxuan Lu
Abstract: 3D occupancy-based perception pipeline has significantly advanced autonomous driving by capturing detailed scene descriptions and demonstrating strong generalizability across various object categories and shapes. Current methods predominantly rely on LiDAR or camera inputs for 3D occupancy prediction. These methods are susceptible to adverse weather conditions, limiting the all-weather deployment of self-driving cars. To improve perception robustness, we leverage the recent advances in automotive radars and introduce a novel approach that utilizes 4D imaging radar sensors for 3D occupancy prediction. Our method, RadarOcc, circumvents the limitations of sparse radar point clouds by directly processing the 4D radar tensor, thus preserving essential scene details. RadarOcc innovatively addresses the challenges associated with the voluminous and noisy 4D radar data by employing Doppler bins descriptors, sidelobe-aware spatial sparsification, and range-wise self-attention mechanisms. To minimize the interpolation errors associated with direct coordinate transformations, we also devise a spherical-based feature encoding followed by spherical-to-Cartesian feature aggregation. We benchmark various baseline methods based on distinct modalities on the public K-Radar dataset. The results demonstrate RadarOcc's state-of-the-art performance in radar-based 3D occupancy prediction and promising results even when compared with LiDAR- or camera-based methods. Additionally, we present qualitative evidence of the superior performance of 4D radar in adverse weather conditions and explore the impact of key pipeline components through ablation studies.
Authors: Alan Q. Wang, Rachit Saluja, Heejong Kim, Xinzi He, Adrian Dalca, Mert R. Sabuncu
Abstract: We present a keypoint-based foundation model for general purpose brain MRI registration, based on the recently-proposed KeyMorph framework. Our model, called BrainMorph, serves as a tool that supports multi-modal, pairwise, and scalable groupwise registration. BrainMorph is trained on a massive dataset of over 100,000 3D volumes, skull-stripped and non-skull-stripped, from nearly 16,000 unique healthy and diseased subjects. BrainMorph is robust to large misalignments, interpretable via interrogating automatically-extracted keypoints, and enables rapid and controllable generation of many plausible transformations with different alignment types and different degrees of nonlinearity at test-time. We demonstrate the superiority of BrainMorph in solving 3D rigid, affine, and nonlinear registration on a variety of multi-modal brain MRI scans of healthy and diseased subjects, in both the pairwise and groupwise setting. In particular, we show registration accuracy and speeds that surpass current state-of-the-art methods, especially in the context of large initial misalignments and large group settings. All code and models are available at https://github.com/alanqrwang/brainmorph.
Authors: Youcan Xu, Zhen Wang, Jun Xiao, Wei Liu, Long Chen
Abstract: With the advance of diffusion models, various personalized image generation methods have been proposed. However, almost all existing work only focuses on either subject-driven or style-driven personalization. Meanwhile, state-of-the-art methods face several challenges in realizing compositional personalization, i.e., composing different subject and style concepts, such as concept disentanglement, unified reconstruction paradigm, and insufficient training data. To address these issues, we introduce FreeTuner, a flexible and training-free method for compositional personalization that can generate any user-provided subject in any user-provided style (see Figure 1). Our approach employs a disentanglement strategy that separates the generation process into two stages to effectively mitigate concept entanglement. FreeTuner leverages the intermediate features within the diffusion model for subject concept representation and introduces style guidance to align the synthesized images with the style concept, ensuring the preservation of both the subject's structure and the style's aesthetic features. Extensive experiments have demonstrated the generation ability of FreeTuner across various personalization settings.
Authors: Jiannan Wang, Jiarui Fang, Aoyu Li, PengCheng Yang
Abstract: This paper introduces PipeFusion, a novel approach that harnesses multi-GPU parallelism to address the high computational and latency challenges of generating high-resolution images with diffusion transformers (DiT) models. PipeFusion splits images into patches and distributes the network layers across multiple devices. It employs a pipeline parallel manner to orchestrate communication and computations. By leveraging the high similarity between the input from adjacent diffusion steps, PipeFusion eliminates the waiting time in the pipeline by reusing the one-step stale feature maps to provide context for the current step. Our experiments demonstrate that it can generate higher image resolution where existing DiT parallel approaches meet OOM. PipeFusion significantly reduces the required communication bandwidth, enabling DiT inference to be hosted on GPUs connected via PCIe rather than the more costly NVLink infrastructure, which substantially lowers the overall operational expenses for serving DiT models. Our code is publicly available at https://github.com/PipeFusion/PipeFusion.
Authors: Lv Tang, HaoKe Xiao, Peng-Tao Jiang, Hao Zhang, Jinwei Chen, Bo Li
Abstract: Foundational models have significantly advanced in natural language processing (NLP) and computer vision (CV), with the Transformer architecture becoming a standard backbone. However, the Transformer's quadratic complexity poses challenges for handling longer sequences and higher resolution images. To address this challenge, State Space Models (SSMs) like Mamba have emerged as efficient alternatives, initially matching Transformer performance in NLP tasks and later surpassing Vision Transformers (ViTs) in various CV tasks. To improve the performance of SSMs, one crucial aspect is effective serialization of image patches. Existing methods, relying on linear scanning curves, often fail to capture complex spatial relationships and produce repetitive patterns, leading to biases. To address these limitations, we propose using fractal scanning curves for patch serialization. Fractal curves maintain high spatial proximity and adapt to different image resolutions, avoiding redundancy and enhancing SSMs' ability to model complex patterns accurately. We validate our method in image classification, detection, and segmentation tasks, and the superior performance validates its effectiveness.
Authors: Mohamed Debbagh, Yixue Liu, Zhouzhou Zheng, Xintong Jiang, Shangpeng Sun, Mark Lefsrud
Abstract: A plant growth simulation can be characterized as a reconstructed visual representation of a plant or plant system. The phenotypic characteristics and plant structures are controlled by the scene environment and other contextual attributes. Considering the temporal dependencies and compounding effects of various factors on growth trajectories, we formulate a probabilistic approach to the simulation task by solving a frame synthesis and pattern recognition problem. We introduce a sequence-informed plant growth simulation framework (SI-PGS) that employs a conditional generative model to implicitly learn a distribution of possible plant representations within a dynamic scene from a fusion of low dimensional temporal sensor and context data. Methods such as controlled latent sampling and recurrent output connections are used to improve coherence in the plant structures between frames of predictions. In this work, we demonstrate that SI-PGS is able to capture temporal dependencies and continuously generate realistic frames of plant growth.
Authors: Xu Pan, Aaron Philip, Ziqian Xie, Odelia Schwartz
Abstract: Self-attention in vision transformers is often thought to perform perceptual grouping where tokens attend to other tokens with similar embeddings, which could correspond to semantically similar features of an object. However, attending to dissimilar tokens can be beneficial by providing contextual information. We propose to use the Singular Value Decomposition to dissect the query-key interaction (i.e. ${\textbf{W}_q}^\top\textbf{W}_k$). We find that early layers attend more to similar tokens, while late layers show increased attention to dissimilar tokens, providing evidence corresponding to perceptual grouping and contextualization, respectively. Many of these interactions between features represented by singular vectors are interpretable and semantic, such as attention between relevant objects, between parts of an object, or between the foreground and background. This offers a novel perspective on interpreting the attention mechanism, which contributes to understanding how transformer models utilize context and salient features when processing images.
Authors: Yuanhao Cai, Zihao Xiao, Yixun Liang, Minghan Qin, Yulun Zhang, Xiaokang Yang, Yaoyao Liu, Alan Yuille
Abstract: High dynamic range (HDR) novel view synthesis (NVS) aims to create photorealistic images from novel viewpoints using HDR imaging techniques. The rendered HDR images capture a wider range of brightness levels containing more details of the scene than normal low dynamic range (LDR) images. Existing HDR NVS methods are mainly based on NeRF. They suffer from long training time and slow inference speed. In this paper, we propose a new framework, High Dynamic Range Gaussian Splatting (HDR-GS), which can efficiently render novel HDR views and reconstruct LDR images with a user input exposure time. Specifically, we design a Dual Dynamic Range (DDR) Gaussian point cloud model that uses spherical harmonics to fit HDR color and employs an MLP-based tone-mapper to render LDR color. The HDR and LDR colors are then fed into two Parallel Differentiable Rasterization (PDR) processes to reconstruct HDR and LDR views. To establish the data foundation for the research of 3D Gaussian splatting-based methods in HDR NVS, we recalibrate the camera parameters and compute the initial positions for Gaussian point clouds. Experiments demonstrate that our HDR-GS surpasses the state-of-the-art NeRF-based method by 3.84 and 1.91 dB on LDR and HDR NVS while enjoying 1000x inference speed and only requiring 6.3% training time. Code, models, and recalibrated data will be publicly available at https://github.com/caiyuanhao1998/HDR-GS
Authors: Hyungtae Lee, Yan Zhang, Yi-Ting Shen, Heesung Kwon, Shuvra S. Bhattacharyya
Abstract: Aerial-view human detection has a large demand for large-scale data to capture more diverse human appearances compared to ground-view human detection. Therefore, synthetic data can be a good resource to expand data, but the domain gap with real-world data is the biggest obstacle to its use in training. As a common solution to deal with the domain gap, the sim2real transformation is used, and its quality is affected by three factors: i) the real data serving as a reference when calculating the domain gap, ii) the synthetic data chosen to avoid the transformation quality degradation, and iii) the synthetic data pool from which the synthetic data is selected. In this paper, we investigate the impact of these factors on maximizing the effectiveness of synthetic data in training in terms of improving learning performance and acquiring domain generalization ability--two main benefits expected of using synthetic data. As an evaluation metric for the second benefit, we introduce a method for measuring the distribution gap between two datasets, which is derived as the normalized sum of the Mahalanobis distances of all test data. As a result, we have discovered several important findings that have never been investigated or have been used previously without accurate understanding. We expect that these findings can break the current trend of either naively using or being hesitant to use synthetic data in machine learning due to the lack of understanding, leading to more appropriate use in future research.
Authors: Yanwei Zheng, Changrui Li, Chuanlin Lan, Yaling Li, Xiao Zhang, Yifei Zou, Dongxiao Yu, Zhipeng Cai
Abstract: Zero-shot object navigation (ZSON) addresses situation where an agent navigates to an unseen object that does not present in the training set. Previous works mainly train agent using seen objects with known labels, and ignore the seen objects without labels. In this paper, we introduce seen objects without labels, herein termed as ``unknown objects'', into training procedure to enrich the agent's knowledge base with distinguishable but previously overlooked information. Furthermore, we propose the label-wise meta-correlation module (LWMCM) to harness relationships among objects with and without labels, and obtain enhanced objects information. Specially, we propose target feature generator (TFG) to generate the features representation of the unlabeled target objects. Subsequently, the unlabeled object identifier (UOI) module assesses whether the unlabeled target object appears in the current observation frame captured by the camera and produces an adapted target features representation specific to the observed context. In meta contrastive feature modifier (MCFM), the target features is modified via approaching the features of objects within the observation frame while distancing itself from features of unobserved objects. Finally, the meta object-graph learner (MOGL) module is utilized to calculate the relationships among objects based on the features. Experiments conducted on AI2THOR and RoboTHOR platforms demonstrate the effectiveness of our proposed method.
Authors: Khanh-Binh Nguyen, Chae Jung Park
Abstract: Self-supervised learning (SSL) is gaining attention for its ability to learn effective representations with large amounts of unlabeled data. Lightweight models can be distilled from larger self-supervised pre-trained models using contrastive and consistency constraints. Still, the different sizes of the projection heads make it challenging for students to mimic the teacher's embedding accurately. We propose \textsc{Retro}, which reuses the teacher's projection head for students, and our experimental results demonstrate significant improvements over the state-of-the-art on all lightweight models. For instance, when training EfficientNet-B0 using ResNet-50/101/152 as teachers, our approach improves the linear result on ImageNet to $66.9\%$, $69.3\%$, and $69.8\%$, respectively, with significantly fewer parameters.
Authors: Byung-Kwan Lee, Chae Won Kim, Beomchan Park, Yong Man Ro
Abstract: The rapid development of large language and vision models (LLVMs) has been driven by advances in visual instruction tuning. Recently, open-source LLVMs have curated high-quality visual instruction tuning datasets and utilized additional vision encoders or multiple computer vision models in order to narrow the performance gap with powerful closed-source LLVMs. These advancements are attributed to multifaceted information required for diverse capabilities, including fundamental image understanding, real-world knowledge about common-sense and non-object concepts (e.g., charts, diagrams, symbols, signs, and math problems), and step-by-step procedures for solving complex questions. Drawing from the multifaceted information, we present a new efficient LLVM, Mamba-based traversal of rationales (Meteor), which leverages multifaceted rationale to enhance understanding and answering capabilities. To embed lengthy rationales containing abundant information, we employ the Mamba architecture, capable of processing sequential data with linear time complexity. We introduce a new concept of traversal of rationale that facilitates efficient embedding of rationale. Subsequently, the backbone multimodal language model (MLM) is trained to generate answers with the aid of rationale. Through these steps, Meteor achieves significant improvements in vision language performances across multiple evaluation benchmarks requiring diverse capabilities, without scaling up the model size or employing additional vision encoders and computer vision models.
Authors: Nikita Durasov, Nik Dorndorf, Hieu Le, Pascal Fua
Abstract: Whereas the ability of deep networks to produce useful predictions has been amply demonstrated, estimating the reliability of these predictions remains challenging. Sampling approaches such as MC-Dropout and Deep Ensembles have emerged as the most popular ones for this purpose. Unfortunately, they require many forward passes at inference time, which slows them down. Sampling-free approaches can be faster but suffer from other drawbacks, such as lower reliability of uncertainty estimates, difficulty of use, and limited applicability to different types of tasks and data. In this work, we introduce a sampling-free approach that is generic and easy to deploy, while producing reliable uncertainty estimates on par with state-of-the-art methods at a significantly lower computational cost. It is predicated on training the network to produce the same output with and without additional information about it. At inference time, when no prior information is given, we use the network's own prediction as the additional information. We then take the distance between the predictions with and without prior information as our uncertainty measure. We demonstrate our approach on several classification and regression tasks. We show that it delivers results on par with those of Ensembles but at a much lower computational cost.
Authors: Zhongwen Zhang, Yuri Boykov
Abstract: Maximization of mutual information between the model's input and output is formally related to "decisiveness" and "fairness" of the softmax predictions, motivating these unsupervised entropy-based criteria for clustering. First, in the context of linear softmax models, we discuss some general properties of entropy-based clustering. Disproving some earlier claims, we point out fundamental differences with K-means. On the other hand, we prove the margin maximizing property for decisiveness establishing a relation to SVM-based clustering. Second, we propose a new self-labeling formulation of entropy clustering for general softmax models. The pseudo-labels are introduced as auxiliary variables "splitting" the fairness and decisiveness. The derived self-labeling loss includes the reverse cross-entropy robust to pseudo-label errors and allows an efficient EM solver for pseudo-labels. Our algorithm improves the state of the art on several standard benchmarks for deep clustering.
Authors: Nan Gao, Zeyu Zhao, Zhi Zeng, Shuwu Zhang, Dongdong Weng, Yihua Bao
Abstract: Gesture synthesis has gained significant attention as a critical research field, aiming to produce contextually appropriate and natural gestures corresponding to speech or textual input. Although deep learning-based approaches have achieved remarkable progress, they often overlook the rich semantic information present in the text, leading to less expressive and meaningful gestures. In this letter, we propose GesGPT, a novel approach to gesture generation that leverages the semantic analysis capabilities of large language models , such as ChatGPT. By capitalizing on the strengths of LLMs for text analysis, we adopt a controlled approach to generate and integrate professional gestures and base gestures through a text parsing script, resulting in diverse and meaningful gestures. Firstly, our approach involves the development of prompt principles that transform gesture generation into an intention classification problem using ChatGPT. We also conduct further analysis on emphasis words and semantic words to aid in gesture generation. Subsequently, we construct a specialized gesture lexicon with multiple semantic annotations, decoupling the synthesis of gestures into professional gestures and base gestures. Finally, we merge the professional gestures with base gestures. Experimental results demonstrate that GesGPT effectively generates contextually appropriate and expressive gestures.
Authors: Leo Feng, Frederick Tung, Hossein Hajimirsadeghi, Yoshua Bengio, Mohamed Osama Ahmed
Abstract: Neural Processes (NPs) are popular meta-learning methods for efficiently modelling predictive uncertainty. Recent state-of-the-art methods, however, leverage expensive attention mechanisms, limiting their applications, particularly in low-resource settings. In this work, we propose Constant Memory Attentive Neural Processes (CMANPs), an NP variant that only requires constant memory. To do so, we first propose an efficient update operation for Cross Attention. Leveraging the update operation, we propose Constant Memory Attention Block (CMAB), a novel attention block that (i) is permutation invariant, (ii) computes its output in constant memory, and (iii) performs constant computation updates. Finally, building on CMAB, we detail Constant Memory Attentive Neural Processes. Empirically, we show CMANPs achieve state-of-the-art results on popular NP benchmarks while being significantly more memory efficient than prior methods.
Authors: Ivan R. Nabi, Ben Cardoen, Ismail M. Khater, Guang Gao, Timothy H. Wong, Ghassan Hamarneh
Abstract: Super-resolution microscopy, or nanoscopy, enables the use of fluorescent-based molecular localization tools to study molecular structure at the nanoscale level in the intact cell, bridging the mesoscale gap to classical structural biology methodologies. Analysis of super-resolution data by artificial intelligence (AI), such as machine learning, offers tremendous potential for discovery of new biology, that, by definition, is not known and lacks ground truth. Herein, we describe the application of weakly supervised paradigms to super-resolution microscopy and its potential to enable the accelerated exploration of the nanoscale architecture of subcellular macromolecules and organelles.
Authors: Joan Puigcerver, Carlos Riquelme, Basil Mustafa, Neil Houlsby
Abstract: Sparse mixture of expert architectures (MoEs) scale model capacity without significant increases in training or inference costs. Despite their success, MoEs suffer from a number of issues: training instability, token dropping, inability to scale the number of experts, or ineffective finetuning. In this work, we propose Soft MoE, a fully-differentiable sparse Transformer that addresses these challenges, while maintaining the benefits of MoEs. Soft MoE performs an implicit soft assignment by passing different weighted combinations of all input tokens to each expert. As in other MoEs, experts in Soft MoE only process a subset of the (combined) tokens, enabling larger model capacity (and performance) at lower inference cost. In the context of visual recognition, Soft MoE greatly outperforms dense Transformers (ViTs) and popular MoEs (Tokens Choice and Experts Choice). Furthermore, Soft MoE scales well: Soft MoE Huge/14 with 128 experts in 16 MoE layers has over 40x more parameters than ViT Huge/14, with only 2% increased inference time, and substantially better quality.
Authors: Zelin Zang, Hao Luo, Kai Wang, Panpan Zhang, Fan Wang, Stan. Z Li, Yang You
Abstract: Unsupervised Contrastive learning has gained prominence in fields such as vision, and biology, leveraging predefined positive/negative samples for representation learning. Data augmentation, categorized into hand-designed and model-based methods, has been identified as a crucial component for enhancing contrastive learning. However, hand-designed methods require human expertise in domain-specific data while sometimes distorting the meaning of the data. In contrast, generative model-based approaches usually require supervised or large-scale external data, which has become a bottleneck constraining model training in many domains. To address the problems presented above, this paper proposes DiffAug, a novel unsupervised contrastive learning technique with diffusion mode-based positive data generation. DiffAug consists of a semantic encoder and a conditional diffusion model; the conditional diffusion model generates new positive samples conditioned on the semantic encoding to serve the training of unsupervised contrast learning. With the help of iterative training of the semantic encoder and diffusion model, DiffAug improves the representation ability in an uninterrupted and unsupervised manner. Experimental evaluations show that DiffAug outperforms hand-designed and SOTA model-based augmentation methods on DNA sequence, visual, and bio-feature datasets. The code for review is released at \url{https://github.com/zangzelin/code_diffaug}.
Authors: Yixuan Huang, Jialin Yuan, Chanho Kim, Pupul Pradhan, Bryan Chen, Li Fuxin, Tucker Hermans
Abstract: Robots need to have a memory of previously observed, but currently occluded objects to work reliably in realistic environments. We investigate the problem of encoding object-oriented memory into a multi-object manipulation reasoning and planning framework. We propose DOOM and LOOM, which leverage transformer relational dynamics to encode the history of trajectories given partial-view point clouds and an object discovery and tracking engine. Our approaches can perform multiple challenging tasks including reasoning with occluded objects, novel objects appearance, and object reappearance. Throughout our extensive simulation and real-world experiments, we find that our approaches perform well in terms of different numbers of objects and different numbers of distractor actions. Furthermore, we show our approaches outperform an implicit memory baseline.
Authors: Ye Zhu, Yu Wu, Duo Xu, Zhiwei Deng, Yan Yan, Olga Russakovsky
Abstract: In this work, we study the generalization properties of diffusion models in a few-shot setup, introduce a novel tuning-free paradigm to synthesize the target out-of-domain (OOD) data, and demonstrate its advantages compared to existing methods in data-sparse scenarios with large domain gaps. Specifically, given a pre-trained model and a small set of images that are OOD relative to the model's training distribution, we explore whether the frozen model is able to generalize to this new domain. We begin by revealing that Denoising Diffusion Probabilistic Models (DDPMs) trained on single-domain images are already equipped with sufficient representation abilities to reconstruct arbitrary images from the inverted latent encoding following bi-directional deterministic diffusion and denoising trajectories. We then demonstrate through both theoretical and empirical perspectives that the OOD images establish Gaussian priors in latent spaces of the given model, and the inverted latent modes are separable from their initial training domain. We then introduce our novel tuning-free paradigm to synthesize new images of the target unseen domain by discovering qualified OOD latent encodings in the inverted noisy spaces. This is fundamentally different from the current paradigm that seeks to modify the denoising trajectory to achieve the same goal by tuning the model parameters. Extensive cross-model and domain experiments show that our proposed method can expand the latent space and generate unseen images via frozen DDPMs without impairing the quality of generation of their original domain. We also showcase a practical application of our proposed heuristic approach in dramatically different domains using astrophysical data, revealing the great potential of such a generalization paradigm in data spare fields such as scientific explorations.
Authors: Xiang Chen, Duanzheng Song, Honghao Gui, Chenxi Wang, Ningyu Zhang, Yong Jiang, Fei Huang, Chengfei Lv, Dan Zhang, Huajun Chen
Abstract: Despite their impressive generative capabilities, LLMs are hindered by fact-conflicting hallucinations in real-world applications. The accurate identification of hallucinations in texts generated by LLMs, especially in complex inferential scenarios, is a relatively unexplored area. To address this gap, we present FactCHD, a dedicated benchmark designed for the detection of fact-conflicting hallucinations from LLMs. FactCHD features a diverse dataset that spans various factuality patterns, including vanilla, multi-hop, comparison, and set operation. A distinctive element of FactCHD is its integration of fact-based evidence chains, significantly enhancing the depth of evaluating the detectors' explanations. Experiments on different LLMs expose the shortcomings of current approaches in detecting factual errors accurately. Furthermore, we introduce Truth-Triangulator that synthesizes reflective considerations by tool-enhanced ChatGPT and LoRA-tuning based on Llama2, aiming to yield more credible detection through the amalgamation of predictive results and evidence. The benchmark dataset is available at https://github.com/zjunlp/FactCHD.
Authors: Shicheng Xu, Danyang Hou, Liang Pang, Jingcheng Deng, Jun Xu, Huawei Shen, Xueqi Cheng
Abstract: With the advancement of generation models, AI-generated content (AIGC) is becoming more realistic, flooding the Internet. A recent study suggests that this phenomenon causes source bias in text retrieval for web search. Specifically, neural retrieval models tend to rank generated texts higher than human-written texts. In this paper, we extend the study of this bias to cross-modal retrieval. Firstly, we successfully construct a suitable benchmark to explore the existence of the bias. Subsequent extensive experiments on this benchmark reveal that AI-generated images introduce an invisible relevance bias to text-image retrieval models. Specifically, our experiments show that text-image retrieval models tend to rank the AI-generated images higher than the real images, even though the AI-generated images do not exhibit more visually relevant features to the query than real images. This invisible relevance bias is prevalent across retrieval models with varying training data and architectures. Furthermore, our subsequent exploration reveals that the inclusion of AI-generated images in the training data of the retrieval models exacerbates the invisible relevance bias. The above phenomenon triggers a vicious cycle, which makes the invisible relevance bias become more and more serious. To elucidate the potential causes of invisible relevance and address the aforementioned issues, we introduce an effective training method aimed at alleviating the invisible relevance bias. Subsequently, we apply our proposed debiasing method to retroactively identify the causes of invisible relevance, revealing that the AI-generated images induce the image encoder to embed additional information into their representation. This information exhibits a certain consistency across generated images with different semantics and can make the retriever estimate a higher relevance score.
Authors: Anthony Chen, Huanrui Yang, Yulu Gan, Denis A Gudovskiy, Zhen Dong, Haofan Wang, Tomoyuki Okuno, Yohei Nakata, Kurt Keutzer, Shanghang Zhang
Abstract: Uncertainty estimation is crucial for machine learning models to detect out-of-distribution (OOD) inputs. However, the conventional discriminative deep learning classifiers produce uncalibrated closed-set predictions for OOD data. A more robust classifiers with the uncertainty estimation typically require a potentially unavailable OOD dataset for outlier exposure training, or a considerable amount of additional memory and compute to build ensemble models. In this work, we improve on uncertainty estimation without extra OOD data or additional inference costs using an alternative Split-Ensemble method. Specifically, we propose a novel subtask-splitting ensemble training objective, where a common multiclass classification task is split into several complementary subtasks. Then, each subtask's training data can be considered as OOD to the other subtasks. Diverse submodels can therefore be trained on each subtask with OOD-aware objectives. The subtask-splitting objective enables us to share low-level features across submodels to avoid parameter and computational overheads. In particular, we build a tree-like Split-Ensemble architecture by performing iterative splitting and pruning from a shared backbone model, where each branch serves as a submodel corresponding to a subtask. This leads to improved accuracy and uncertainty estimation across submodels under a fixed ensemble computation budget. Empirical study with ResNet-18 backbone shows Split-Ensemble, without additional computation cost, improves accuracy over a single model by 0.8%, 1.8%, and 25.5% on CIFAR-10, CIFAR-100, and Tiny-ImageNet, respectively. OOD detection for the same backbone and in-distribution datasets surpasses a single model baseline by, correspondingly, 2.2%, 8.1%, and 29.6% mean AUROC.
Authors: Viet Dung Nguyen, Michael T. LaCour, Richard D. Komistek
Abstract: Image segmentation in total knee arthroplasty is crucial for precise preoperative planning and accurate implant positioning, leading to improved surgical outcomes and patient satisfaction. The biggest challenges of image segmentation in total knee arthroplasty include accurately delineating complex anatomical structures, dealing with image artifacts and noise, and developing robust algorithms that can handle anatomical variations and pathologies commonly encountered in patients. The potential of using machine learning for image segmentation in total knee arthroplasty lies in its ability to improve segmentation accuracy, automate the process, and provide real-time assistance to surgeons, leading to enhanced surgical planning, implant placement, and patient outcomes. This paper proposes a methodology to use deep learning for robust and real-time total knee arthroplasty image segmentation. The deep learning model, trained on a large dataset, demonstrates outstanding performance in accurately segmenting both the implanted femur and tibia, achieving an impressive mean-Average-Precision (mAP) of 88.83 when compared to the ground truth while also achieving a real-time segmented speed of 20 frames per second (fps). We have introduced a novel methodology for segmenting implanted knee fluoroscopic or x-ray images that showcases remarkable levels of accuracy and speed, paving the way for various potential extended applications.
Authors: Xi Zhang, Xiaolin Wu
Abstract: Recently, DNN models for lossless image coding have surpassed their traditional counterparts in compression performance, reducing the bit rate by about ten percent for natural color images. But even with these advances, mathematically lossless image compression (MLLIC) ratios for natural images still fall short of the bandwidth and cost-effectiveness requirements of most practical imaging and vision systems at present and beyond. To break the bottleneck of MLLIC in compression performance, we question the necessity of MLLIC, as almost all digital sensors inherently introduce acquisition noises, making mathematically lossless compression counterproductive. Therefore, in contrast to MLLIC, we propose a new paradigm of joint denoising and compression called functionally lossless image compression (FLLIC), which performs lossless compression of optimally denoised images (the optimality may be task-specific). Although not literally lossless with respect to the noisy input, FLLIC aims to achieve the best possible reconstruction of the latent noise-free original image. Extensive experiments show that FLLIC achieves state-of-the-art performance in joint denoising and compression of noisy images and does so at a lower computational cost.
Authors: Muthu Chidambaram, Rong Ge
Abstract: Data augmentation has been pivotal in successfully training deep learning models on classification tasks over the past decade. An important subclass of data augmentation techniques - which includes both label smoothing and Mixup - involves modifying not only the input data but also the input label during model training. In this work, we analyze the role played by the label augmentation aspect of such methods. We first prove that linear models on binary classification data trained with label augmentation learn only the minimum variance features in the data, while standard training (which includes weight decay) can learn higher variance features. We then use our techniques to show that even for nonlinear models and general data distributions, the label smoothing and Mixup losses are lower bounded by a function of the model output variance. An important consequence of our results is negative: label smoothing and Mixup can be less robust to spurious correlations in the data. We verify that our theory reflects practice via experiments on image classification benchmarks modified to have spurious correlations.
Authors: Jin-Hwa Kim
Abstract: Recent advancements in visualizing deep neural networks provide insights into their structures and mesh extraction from Continuous Piecewise Affine (CPWA) functions. Meanwhile, developments in neural surface representation learning incorporate non-linear positional encoding, addressing issues like spectral bias; however, this poses challenges in applying mesh extraction techniques based on CPWA functions. Focusing on trilinear interpolating methods as positional encoding, we present theoretical insights and an analytical mesh extraction, showing the transformation of hypersurfaces to flat planes within the trilinear region under the eikonal constraint. Moreover, we introduce a method for approximating intersecting points among three hypersurfaces contributing to broader applications. We empirically validate correctness and parsimony through chamfer distance and efficiency, and angular distance, while examining the correlation between the eikonal loss and the planarity of the hypersurfaces.
Authors: Arturs Berzins, Andreas Radler, Sebastian Sanokowski, Sepp Hochreiter, Johannes Brandstetter
Abstract: Geometry is a ubiquitous language of computer graphics, design, and engineering. However, the lack of large shape datasets limits the application of state-of-the-art supervised learning methods and motivates the exploration of alternative learning strategies. To this end, we introduce geometry-informed neural networks (GINNs) to train shape generative models \emph{without any data}. GINNs combine (i) learning under constraints, (ii) neural fields as a suitable representation, and (iii) generating diverse solutions to under-determined problems. We apply GINNs to several two and three-dimensional problems of increasing levels of complexity. Our results demonstrate the feasibility of training shape generative models in a data-free setting. This new paradigm opens several exciting research directions, expanding the application of generative models into domains where data is sparse.
Authors: Sisipho Hamlomo, Marcellin Atemkeng, Yusuf Brima, Chuneeta Nunhokee, Jeremy Baxter
Abstract: The large volume and complexity of medical imaging datasets are bottlenecks for storage, transmission, and processing. To tackle these challenges, the application of low-rank matrix approximation (LRMA) and its derivative, local LRMA (LLRMA) has demonstrated potential. A detailed analysis of the literature identifies LRMA and LLRMA methods applied to various imaging modalities, and the challenges and limitations associated with existing LRMA and LLRMA methods are addressed. We note a significant shift towards a preference for LLRMA in the medical imaging field since 2015, demonstrating its potential and effectiveness in capturing complex structures in medical data compared to LRMA. Acknowledging the limitations of shallow similarity methods used with LLRMA, we suggest advanced semantic image segmentation for similarity measure, explaining in detail how it can be used to measure similar patches and its feasibility. We note that LRMA and LLRMA are mainly applied to unstructured medical data, and we propose extending their application to different medical data types, including structured and semi-structured. This paper also discusses how LRMA and LLRMA can be applied to regular data with missing entries and the impact of inaccuracies in predicting missing values and their effects. We discuss the impact of patch size and propose the use of random search (RS) to determine the optimal patch size. To enhance feasibility, a hybrid approach using Bayesian optimization and RS is proposed, which could improve the application of LRMA and LLRMA in medical imaging.
Authors: Sam Dauncey, Chris Holmes, Christopher Williams, Fabian Falck
Abstract: Likelihood-based deep generative models such as score-based diffusion models and variational autoencoders are state-of-the-art machine learning models approximating high-dimensional distributions of data such as images, text, or audio. One of many downstream tasks they can be naturally applied to is out-of-distribution (OOD) detection. However, seminal work by Nalisnick et al. which we reproduce showed that deep generative models consistently infer higher log-likelihoods for OOD data than data they were trained on, marking an open problem. In this work, we analyse using the gradient of a data point with respect to the parameters of the deep generative model for OOD detection, based on the simple intuition that OOD data should have larger gradient norms than training data. We formalise measuring the size of the gradient as approximating the Fisher information metric. We show that the Fisher information matrix (FIM) has large absolute diagonal values, motivating the use of chi-square distributed, layer-wise gradient norms as features. We combine these features to make a simple, model-agnostic and hyperparameter-free method for OOD detection which estimates the joint density of the layer-wise gradient norms for a given data point. We find that these layer-wise gradient norms are weakly correlated, rendering their combined usage informative, and prove that the layer-wise gradient norms satisfy the principle of (data representation) invariance. Our empirical results indicate that this method outperforms the Typicality test for most deep generative models and image dataset pairings.
Authors: Marvin Li, Sitan Chen
Abstract: We develop theory to understand an intriguing property of diffusion models for image generation that we term critical windows. Empirically, it has been observed that there are narrow time intervals in sampling during which particular features of the final image emerge, e.g. the image class or background color (Ho et al., 2020b; Meng et al., 2022; Choi et al., 2022; Raya & Ambrogioni, 2023; Georgiev et al., 2023; Sclocchi et al., 2024; Biroli et al., 2024). While this is advantageous for interpretability as it implies one can localize properties of the generation to a small segment of the trajectory, it seems at odds with the continuous nature of the diffusion. We propose a formal framework for studying these windows and show that for data coming from a mixture of strongly log-concave densities, these windows can be provably bounded in terms of certain measures of inter- and intra-group separation. We also instantiate these bounds for concrete examples like well-conditioned Gaussian mixtures. Finally, we use our bounds to give a rigorous interpretation of diffusion models as hierarchical samplers that progressively "decide" output features over a discrete sequence of times. We validate our bounds with synthetic experiments. Additionally, preliminary experiments on Stable Diffusion suggest critical windows may serve as a useful tool for diagnosing fairness and privacy violations in real-world diffusion models.
Authors: Arkajit Datta, Tushar Verma, Rajat Chawla, Mukunda N. S, Ishaan Bhola
Abstract: In recent advancements within the domain of Large Language Models (LLMs), there has been a notable emergence of agents capable of addressing Robotic Process Automation (RPA) challenges through enhanced cognitive capabilities and sophisticated reasoning. This development heralds a new era of scalability and human-like adaptability in goal attainment. In this context, we introduce AUTONODE (Autonomous User-interface Transformation through Online Neuro-graphic Operations and Deep Exploration). AUTONODE employs advanced neuro-graphical techniques to facilitate autonomous navigation and task execution on web interfaces, thereby obviating the necessity for predefined scripts or manual intervention. Our engine empowers agents to comprehend and implement complex workflows, adapting to dynamic web environments with unparalleled efficiency. Our methodology synergizes cognitive functionalities with robotic automation, endowing AUTONODE with the ability to learn from experience. We have integrated an exploratory module, DoRA (Discovery and mapping Operation for graph Retrieval Agent), which is instrumental in constructing a knowledge graph that the engine utilizes to optimize its actions and achieve objectives with minimal supervision. The versatility and efficacy of AUTONODE are demonstrated through a series of experiments, highlighting its proficiency in managing a diverse array of web-based tasks, ranging from data extraction to transaction processing.
Authors: Chun-Tse Chien, Rui-Yang Ju, Kuang-Yi Chou, Jen-Shiun Chiang
Abstract: The introduction of YOLOv9, the latest version of the You Only Look Once (YOLO) series, has led to its widespread adoption across various scenarios. This paper is the first to apply the YOLOv9 algorithm model to the fracture detection task as computer-assisted diagnosis (CAD) to help radiologists and surgeons to interpret X-ray images. Specifically, this paper trained the model on the GRAZPEDWRI-DX dataset and extended the training set using data augmentation techniques to improve the model performance. Experimental results demonstrate that compared to the mAP 50-95 of the current state-of-the-art (SOTA) model, the YOLOv9 model increased the value from 42.16% to 43.73%, with an improvement of 3.7%. The implementation code is publicly available at https://github.com/RuiyangJu/YOLOv9-Fracture-Detection.
URLs: https://github.com/RuiyangJu/YOLOv9-Fracture-Detection.
Authors: Zhicong Tang, Tiankai Hang, Shuyang Gu, Dong Chen, Baining Guo
Abstract: This paper introduces a novel theoretical simplification of the Diffusion Schr\"odinger Bridge (DSB) that facilitates its unification with Score-based Generative Models (SGMs), addressing the limitations of DSB in complex data generation and enabling faster convergence and enhanced performance. By employing SGMs as an initial solution for DSB, our approach capitalizes on the strengths of both frameworks, ensuring a more efficient training process and improving the performance of SGM. We also propose a reparameterization technique that, despite theoretical approximations, practically improves the network's fitting capabilities. Our extensive experimental evaluations confirm the effectiveness of the simplified DSB, demonstrating its significant improvements. We believe the contributions of this work pave the way for advanced generative modeling. The code is available at https://github.com/checkcrab/SDSB.
Authors: Allen Z. Ren, Jaden Clark, Anushri Dixit, Masha Itkina, Anirudha Majumdar, Dorsa Sadigh
Abstract: We consider the problem of Embodied Question Answering (EQA), which refers to settings where an embodied agent such as a robot needs to actively explore an environment to gather information until it is confident about the answer to a question. In this work, we leverage the strong semantic reasoning capabilities of large vision-language models (VLMs) to efficiently explore and answer such questions. However, there are two main challenges when using VLMs in EQA: they do not have an internal memory for mapping the scene to be able to plan how to explore over time, and their confidence can be miscalibrated and can cause the robot to prematurely stop exploration or over-explore. We propose a method that first builds a semantic map of the scene based on depth information and via visual prompting of a VLM - leveraging its vast knowledge of relevant regions of the scene for exploration. Next, we use conformal prediction to calibrate the VLM's question answering confidence, allowing the robot to know when to stop exploration - leading to a more calibrated and efficient exploration strategy. To test our framework in simulation, we also contribute a new EQA dataset with diverse, realistic human-robot scenarios and scenes built upon the Habitat-Matterport 3D Research Dataset (HM3D). Both simulated and real robot experiments show our proposed approach improves the performance and efficiency over baselines that do no leverage VLM for exploration or do not calibrate its confidence. Webpage with experiment videos and code: https://explore-eqa.github.io/
Authors: Zhenhailong Wang, Joy Hsu, Xingyao Wang, Kuan-Hao Huang, Manling Li, Jiajun Wu, Heng Ji
Abstract: While large multimodal models excel in broad vision-language benchmarks, they often struggle with tasks requiring precise perception of low-level visual details, such as comparing line lengths or solving simple mazes. In particular, this failure mode persists in question-answering tasks about vector graphics -- images composed purely of 2D objects and shapes. To address this challenge, we propose the Visually Descriptive Language Model (VDLM), which performs text-based reasoning about vector graphics. VDLM leverages Scalable Vector Graphics (SVG) for a more precise visual description and first uses an off-the-shelf raster-to-SVG algorithm for encoding. Since existing language models cannot understand raw SVGs in a zero-shot setting, VDLM then bridges SVG with pretrained language models through a newly introduced intermediate symbolic representation, Primal Visual Description (PVD), comprising primitive attributes (e.g., shape, position, measurement) with their corresponding predicted values. PVD is task-agnostic and represents visual primitives that are universal across all vector graphics. It can be learned with procedurally generated (SVG, PVD) pairs and also enables the direct use of LLMs for generalization to complex reasoning tasks. By casting an image to a text-based representation, we can leverage the power of language models to learn alignment from SVG to visual primitives and generalize to unseen question-answering tasks. Empirical results show that VDLM achieves stronger zero-shot performance compared to state-of-the-art LMMs, such as GPT-4V, in various low-level multimodal perception and reasoning tasks on vector graphics. We additionally present extensive analyses on VDLM's performance, demonstrating that our framework offers better interpretability due to its disentangled perception and reasoning processes. Project page: https://mikewangwzhl.github.io/VDLM/
Authors: Chi Tran, Huong Le Thanh
Abstract: Large Language Models (LLMs) and Multimodal Large language models (MLLMs) have taken the world by storm with impressive abilities in complex reasoning and linguistic comprehension. Meanwhile there are plethora of works related to Vietnamese Large Language Models, the lack of high-quality resources in multimodality limits the progress of Vietnamese MLLMs. In this paper, we pioneer in address this by introducing LaVy, a state-of-the-art Vietnamese MLLM, and we also introduce LaVy-Bench benchmark designated for evaluating MLLMs's understanding on Vietnamese visual language tasks. Our project is public at https://github.com/baochi0212/LaVy
Authors: Kyle Hsu, Jubayer Ibn Hamid, Kaylee Burns, Chelsea Finn, Jiajun Wu
Abstract: Inductive biases are crucial in disentangled representation learning for narrowing down an underspecified solution set. In this work, we consider endowing a neural network autoencoder with three select inductive biases from the literature: data compression into a grid-like latent space via quantization, collective independence amongst latents, and minimal functional influence of any latent on how other latents determine data generation. In principle, these inductive biases are deeply complementary: they most directly specify properties of the latent space, encoder, and decoder, respectively. In practice, however, naively combining existing techniques instantiating these inductive biases fails to yield significant benefits. To address this, we propose adaptations to the three techniques that simplify the learning problem, equip key regularization terms with stabilizing invariances, and quash degenerate incentives. The resulting model, Tripod, achieves state-of-the-art results on a suite of four image disentanglement benchmarks. We also verify that Tripod significantly improves upon its naive incarnation and that all three of its "legs" are necessary for best performance.
Authors: Xiaofei Wang, Xingxu Huang, Stephen J. Price, Chao Li
Abstract: The recent advancement of spatial transcriptomics (ST) allows to characterize spatial gene expression within tissue for discovery research. However, current ST platforms suffer from low resolution, hindering in-depth understanding of spatial gene expression. Super-resolution approaches promise to enhance ST maps by integrating histology images with gene expressions of profiled tissue spots. However, current super-resolution methods are limited by restoration uncertainty and mode collapse. Although diffusion models have shown promise in capturing complex interactions between multi-modal conditions, it remains a challenge to integrate histology images and gene expression for super-resolved ST maps. This paper proposes a cross-modal conditional diffusion model for super-resolving ST maps with the guidance of histology images. Specifically, we design a multi-modal disentangling network with cross-modal adaptive modulation to utilize complementary information from histology images and spatial gene expression. Moreover, we propose a dynamic cross-attention modelling strategy to extract hierarchical cell-to-tissue information from histology images. Lastly, we propose a co-expression-based gene-correlation graph network to model the co-expression relationship of multiple genes. Experiments show that our method outperforms other state-of-the-art methods in ST super-resolution on three public datasets.
Authors: Ali Nasiri-Sarvi, Vincent Quoc-Huy Trinh, Hassan Rivaz, Mahdi S. Hosseini
Abstract: Representation learning from Gigapixel Whole Slide Images (WSI) poses a significant challenge in computational pathology due to the complicated nature of tissue structures and the scarcity of labeled data. Multi-instance learning methods have addressed this challenge, leveraging image patches to classify slides utilizing pretrained models using Self-Supervised Learning (SSL) approaches. The performance of both SSL and MIL methods relies on the architecture of the feature encoder. This paper proposes leveraging the Vision Mamba (Vim) architecture, inspired by state space models, within the DINO framework for representation learning in computational pathology. We evaluate the performance of Vim against Vision Transformers (ViT) on the Camelyon16 dataset for both patch-level and slide-level classification. Our findings highlight Vim's enhanced performance compared to ViT, particularly at smaller scales, where Vim achieves an 8.21 increase in ROC AUC for models of similar size. An explainability analysis further highlights Vim's capabilities, which reveals that Vim uniquely emulates the pathologist workflow-unlike ViT. This alignment with human expert analysis highlights Vim's potential in practical diagnostic settings and contributes significantly to developing effective representation-learning algorithms in computational pathology. We release the codes and pretrained weights at \url{https://github.com/AtlasAnalyticsLab/Vim4Path}.
Authors: Muhammad Abdullah, Anne Querfurth, Deepak Bhatia, Mahdi Mantash
Abstract: This paper investigates the use of deep learning approaches to estimate the femur caput-collum-diaphyseal (CCD) angle from X-ray images. The CCD angle is an important measurement in the diagnosis of hip problems, and correct prediction can help in the planning of surgical procedures. Manual measurement of this angle, on the other hand, can be time-intensive and vulnerable to inter-observer variability. In this paper, we present a deep-learning algorithm that can reliably estimate the femur CCD angle from X-ray images. To train and test the performance of our model, we employed an X-ray image dataset with associated femur CCD angle measurements. Furthermore, we built a prototype to display the resulting predictions and to allow the user to interact with the predictions. As this is happening in a sterile setting during surgery, we expanded our interface to the possibility of being used only by voice commands. Our results show that our deep learning model predicts the femur CCD angle on X-ray images with great accuracy, with a mean absolute error of 4.3 degrees on the left femur and 4.9 degrees on the right femur on the test dataset. Our results suggest that deep learning has the potential to give a more efficient and accurate technique for predicting the femur CCD angle, which might have substantial therapeutic implications for the diagnosis and management of hip problems.
Authors: Johann Schmidt, Sebastian Stober
Abstract: Deep neural networks are applied in more and more areas of everyday life. However, they still lack essential abilities, such as robustly dealing with spatially transformed input signals. Approaches to mitigate this severe robustness issue are limited to two pathways: Either models are implicitly regularised by increased sample variability (data augmentation) or explicitly constrained by hard-coded inductive biases. The limiting factor of the former is the size of the data space, which renders sufficient sample coverage intractable. The latter is limited by the engineering effort required to develop such inductive biases for every possible scenario. Instead, we take inspiration from human behaviour, where percepts are modified by mental or physical actions during inference. We propose a novel technique to emulate such an inference process for neural nets. This is achieved by traversing a sparsified inverse transformation tree during inference using parallel energy-based evaluations. Our proposed inference algorithm, called Inverse Transformation Search (ITS), is model-agnostic and equips the model with zero-shot pseudo-invariance to spatially transformed inputs. We evaluated our method on several benchmark datasets, including a synthesised ImageNet test set. ITS outperforms the utilised baselines on all zero-shot test scenarios.
Authors: Shijie Liu, Kang Yan, Feiwei Qin, Changmiao Wang, Ruiquan Ge, Kai Zhang, Jie Huang, Yong Peng, Jin Cao
Abstract: Single image super-resolution (SR) is an established pixel-level vision task aimed at reconstructing a high-resolution image from its degraded low-resolution counterpart. Despite the notable advancements achieved by leveraging deep neural networks for SR, most existing deep learning architectures feature an extensive number of layers, leading to high computational complexity and substantial memory demands. These issues become particularly pronounced in the context of infrared image SR, where infrared devices often have stringent storage and computational constraints. To mitigate these challenges, we introduce a novel, efficient, and precise single infrared image SR model, termed the Lightweight Information Split Network (LISN). The LISN comprises four main components: shallow feature extraction, deep feature extraction, dense feature fusion, and high-resolution infrared image reconstruction. A key innovation within this model is the introduction of the Lightweight Information Split Block (LISB) for deep feature extraction. The LISB employs a sequential process to extract hierarchical features, which are then aggregated based on the relevance of the features under consideration. By integrating channel splitting and shift operations, the LISB successfully strikes an optimal balance between enhanced SR performance and a lightweight framework. Comprehensive experimental evaluations reveal that the proposed LISN achieves superior performance over contemporary state-of-the-art methods in terms of both SR quality and model complexity, affirming its efficacy for practical deployment in resource-constrained infrared imaging applications.
Authors: Shao-Yuan Lo, Vishal M. Patel
Abstract: Deep networks are vulnerable to adversarial examples. Adversarial Training (AT) has been a standard foundation of modern adversarial defense approaches due to its remarkable effectiveness. However, AT is extremely time-consuming, refraining it from wide deployment in practical applications. In this paper, we aim at a non-AT defense: How to design a defense method that gets rid of AT but is still robust against strong adversarial attacks? To answer this question, we resort to adaptive Batch Normalization (BN), inspired by the recent advances in test-time domain adaptation. We propose a novel defense accordingly, referred to as the Adaptive Batch Normalization Network (ABNN). ABNN employs a pre-trained substitute model to generate clean BN statistics and sends them to the target model. The target model is exclusively trained on clean data and learns to align the substitute model's BN statistics. Experimental results show that ABNN consistently improves adversarial robustness against both digital and physically realizable attacks on both image and video datasets. Furthermore, ABNN can achieve higher clean data performance and significantly lower training time complexity compared to AT-based approaches.
Authors: Zheyuan Zhang, Elif Keles, Gorkem Durak, Yavuz Taktak, Onkar Susladkar, Vandan Gorade, Debesh Jha, Asli C. Ormeci, Alpay Medetalibeyoglu, Lanhong Yao, Bin Wang, Ilkin Sevgi Isler, Linkai Peng, Hongyi Pan, Camila Lopes Vendrami, Amir Bourhani, Yury Velichko, Boqing Gong, Concetto Spampinato, Ayis Pyrros, Pallavi Tiwari, Derk C. F. Klatte, Megan Engels, Sanne Hoogenboom, Candice W. Bolan, Emil Agarunov, Nassier Harfouch, Chenchan Huang, Marco J. Bruno, Ivo Schoots, Rajesh N. Keswani, Frank H. Miller, Tamas Gonda, Cemal Yazici, Temel Tirkes, Baris Turkbey, Michael B. Wallace, Ulas Bagci
Abstract: Automated volumetric segmentation of the pancreas on cross-sectional imaging is needed for diagnosis and follow-up of pancreatic diseases. While CT-based pancreatic segmentation is more established, MRI-based segmentation methods are understudied, largely due to a lack of publicly available datasets, benchmarking research efforts, and domain-specific deep learning methods. In this retrospective study, we collected a large dataset (767 scans from 499 participants) of T1-weighted (T1W) and T2-weighted (T2W) abdominal MRI series from five centers between March 2004 and November 2022. We also collected CT scans of 1,350 patients from publicly available sources for benchmarking purposes. We developed a new pancreas segmentation method, called PanSegNet, combining the strengths of nnUNet and a Transformer network with a new linear attention module enabling volumetric computation. We tested PanSegNet's accuracy in cross-modality (a total of 2,117 scans) and cross-center settings with Dice and Hausdorff distance (HD95) evaluation metrics. We used Cohen's kappa statistics for intra and inter-rater agreement evaluation and paired t-tests for volume and Dice comparisons, respectively. For segmentation accuracy, we achieved Dice coefficients of 88.3% (std: 7.2%, at case level) with CT, 85.0% (std: 7.9%) with T1W MRI, and 86.3% (std: 6.4%) with T2W MRI. There was a high correlation for pancreas volume prediction with R^2 of 0.91, 0.84, and 0.85 for CT, T1W, and T2W, respectively. We found moderate inter-observer (0.624 and 0.638 for T1W and T2W MRI, respectively) and high intra-observer agreement scores. All MRI data is made available at https://osf.io/kysnj/. Our source code is available at https://github.com/NUBagciLab/PaNSegNet.
URLs: https://osf.io/kysnj/., https://github.com/NUBagciLab/PaNSegNet.
Authors: Saddam Hussain Khan (Artificial Intelligence Lab, Department of Computer Systems Engineering, University of Engineering,Applied Sciences)
Abstract: Alzheimer diseases (ADs) involves cognitive decline and abnormal brain protein accumulation, necessitating timely diagnosis for effective treatment. Therefore, CAD systems leveraging deep learning advancements have demonstrated success in AD detection but pose computational intricacies and the dataset minor contrast, structural, and texture variations. In this regard, a novel hybrid FME-Residual-HSCMT technique is introduced, comprised of residual CNN and Transformer concepts to capture global and local fine-grained AD analysis in MRI. This approach integrates three distinct elements: a novel CNN Meet Transformer (HSCMT), customized residual learning CNN, and a new Feature Map Enhancement (FME) strategy to learn diverse morphological, contrast, and texture variations of ADs. The proposed HSCMT at the initial stage utilizes stem convolution blocks that are integrated with CMT blocks followed by systematic homogenous and structural (HS) operations. The customized CMT block encapsulates each element with global contextual interactions through multi-head attention and facilitates computational efficiency through lightweight. Moreover, inverse residual and stem CNN in customized CMT enables effective extraction of local texture information and handling vanishing gradients. Furthermore, in the FME strategy, residual CNN blocks utilize TL-based generated auxiliary and are combined with the proposed HSCMT channels at the target level to achieve diverse enriched feature space. Finally, diverse enhanced channels are fed into a novel spatial attention mechanism for optimal pixel selection to reduce redundancy and discriminate minor contrast and texture inter-class variation. The proposed achieves an F1-score (98.55%), an accuracy of 98.42% and a sensitivity of 98.50%, a precision of 98.60% on the standard Kaggle dataset, and demonstrates outperformance existing ViTs and CNNs methods.
Authors: Yao Rong, David Scheerer, Enkelejda Kasneci
Abstract: In recent years, model explanation methods have been designed to interpret model decisions faithfully and intuitively so that users can easily understand them. In this paper, we propose a framework, Faithful Attention Explainer (FAE), capable of generating faithful textual explanations regarding the attended-to features. Towards this goal, we deploy an attention module that takes the visual feature maps from the classifier for sentence generation. Furthermore, our method successfully learns the association between features and words, which allows a novel attention enforcement module for attention explanation. Our model achieves promising performance in caption quality metrics and a faithful decision-relevance metric on two datasets (CUB and ACT-X). In addition, we show that FAE can interpret gaze-based human attention, as human gaze indicates the discriminative features that humans use for decision-making, demonstrating the potential of deploying human gaze for advanced human-AI interaction.
Authors: Rub\'en Ballester, Pablo Hern\'andez-Garc\'ia, Mathilde Papillon, Claudio Battiloro, Nina Miolane, Tolga Birdal, Carles Casacuberta, Sergio Escalera, Mustafa Hajij
Abstract: Topological Deep Learning seeks to enhance the predictive performance of neural network models by harnessing topological structures in input data. Topological neural networks operate on spaces such as cell complexes and hypergraphs, that can be seen as generalizations of graphs. In this work, we introduce the Cellular Transformer (CT), a novel architecture that generalizes graph-based transformers to cell complexes. First, we propose a new formulation of the usual self- and cross-attention mechanisms, tailored to leverage incidence relations in cell complexes, e.g., edge-face and node-edge relations. Additionally, we propose a set of topological positional encodings specifically designed for cell complexes. By transforming three graph datasets into cell complex datasets, our experiments reveal that CT not only achieves state-of-the-art performance, but it does so without the need for more complex enhancements such as virtual nodes, in-domain structural encodings, or graph rewiring.
Authors: Yiyang Zhou, Zhiyuan Fan, Dongjie Cheng, Sihan Yang, Zhaorun Chen, Chenhang Cui, Xiyao Wang, Yun Li, Linjun Zhang, Huaxiu Yao
Abstract: Large Vision-Language Models (LVLMs) have made substantial progress by integrating pre-trained large language models (LLMs) and vision models through instruction tuning. Despite these advancements, LVLMs often exhibit the hallucination phenomenon, where generated text responses appear linguistically plausible but contradict the input image, indicating a misalignment between image and text pairs. This misalignment arises because the model tends to prioritize textual information over visual input, even when both the language model and visual representations are of high quality. Existing methods leverage additional models or human annotations to curate preference data and enhance modality alignment through preference optimization. These approaches may not effectively reflect the target LVLM's preferences, making the curated preferences easily distinguishable. Our work addresses these challenges by proposing the Calibrated Self-Rewarding (CSR) approach, which enables the model to self-improve by iteratively generating candidate responses, evaluating the reward for each response, and curating preference data for fine-tuning. In the reward modeling, we employ a step-wise strategy and incorporate visual constraints into the self-rewarding process to place greater emphasis on visual input. Empirical results demonstrate that CSR enhances performance and reduces hallucinations across ten benchmarks and tasks, achieving substantial improvements over existing methods by 7.62%. Our empirical results are further supported by rigorous theoretical analysis, under mild assumptions, verifying the effectiveness of introducing visual constraints into the self-rewarding paradigm. Additionally, CSR shows compatibility with different vision-language models and the ability to incrementally improve performance through iterative fine-tuning. Our data and code are available at https://github.com/YiyangZhou/CSR.
Authors: Royson Lee, Javier Fernandez-Marques, Shell Xu Hu, Da Li, Stefanos Laskaridis, {\L}ukasz Dudziak, Timothy Hospedales, Ferenc Husz\'ar, Nicholas D. Lane
Abstract: Federated learning (FL) has enabled distributed learning of a model across multiple clients in a privacy-preserving manner. One of the main challenges of FL is to accommodate clients with varying hardware capacities; clients have differing compute and memory requirements. To tackle this challenge, recent state-of-the-art approaches leverage the use of early exits. Nonetheless, these approaches fall short of mitigating the challenges of joint learning multiple exit classifiers, often relying on hand-picked heuristic solutions for knowledge distillation among classifiers and/or utilizing additional layers for weaker classifiers. In this work, instead of utilizing multiple classifiers, we propose a recurrent early exit approach named ReeFL that fuses features from different sub-models into a single shared classifier. Specifically, we use a transformer-based early-exit module shared among sub-models to i) better exploit multi-layer feature representations for task-specific prediction and ii) modulate the feature representation of the backbone model for subsequent predictions. We additionally present a per-client self-distillation approach where the best sub-model is automatically selected as the teacher of the other sub-models at each client. Our experiments on standard image and speech classification benchmarks across various emerging federated fine-tuning baselines demonstrate ReeFL's effectiveness over previous works.
Authors: Shiyu Qin, Jinpeng Wang, Yimin Zhou, Bin Chen, Tianci Luo, Baoyi An, Tao Dai, Shutao Xia, Yaowei Wang
Abstract: Learned visual compression is an important and active task in multimedia. Existing approaches have explored various CNN- and Transformer-based designs to model content distribution and eliminate redundancy, where balancing efficacy (i.e., rate-distortion trade-off) and efficiency remains a challenge. Recently, state-space models (SSMs) have shown promise due to their long-range modeling capacity and efficiency. Inspired by this, we take the first step to explore SSMs for visual compression. We introduce MambaVC, a simple, strong and efficient compression network based on SSM. MambaVC develops a visual state space (VSS) block with a 2D selective scanning (2DSS) module as the nonlinear activation function after each downsampling, which helps to capture informative global contexts and enhances compression. On compression benchmark datasets, MambaVC achieves superior rate-distortion performance with lower computational and memory overheads. Specifically, it outperforms CNN and Transformer variants by 9.3% and 15.6% on Kodak, respectively, while reducing computation by 42% and 24%, and saving 12% and 71% of memory. MambaVC shows even greater improvements with high-resolution images, highlighting its potential and scalability in real-world applications. We also provide a comprehensive comparison of different network designs, underscoring MambaVC's advantages.
Authors: Pranab Sahoo, Ayush Kumar Singh, Sriparna Saha, Aman Chadha, Samrat Mondal
Abstract: The mining of adverse drug events (ADEs) is pivotal in pharmacovigilance, enhancing patient safety by identifying potential risks associated with medications, facilitating early detection of adverse events, and guiding regulatory decision-making. Traditional ADE detection methods are reliable but slow, not easily adaptable to large-scale operations, and offer limited information. With the exponential increase in data sources like social media content, biomedical literature, and Electronic Medical Records (EMR), extracting relevant ADE-related information from these unstructured texts is imperative. Previous ADE mining studies have focused on text-based methodologies, overlooking visual cues, limiting contextual comprehension, and hindering accurate interpretation. To address this gap, we present a MultiModal Adverse Drug Event (MMADE) detection dataset, merging ADE-related textual information with visual aids. Additionally, we introduce a framework that leverages the capabilities of LLMs and VLMs for ADE detection by generating detailed descriptions of medical images depicting ADEs, aiding healthcare professionals in visually identifying adverse events. Using our MMADE dataset, we showcase the significance of integrating visual cues from images to enhance overall performance. This approach holds promise for patient safety, ADE awareness, and healthcare accessibility, paving the way for further exploration in personalized healthcare.