Authors: Shijie Wang, Dahun Kim, Ali Taalimi, Chen Sun, Weicheng Kuo
Abstract: Visual grounding tasks aim to localize image regions based on natural language references. In this work, we explore whether generative VLMs predominantly trained on image-text data could be leveraged to scale up the text annotation of visual grounding data. We find that grounding knowledge already exists in generative VLM and can be elicited by proper prompting. We thus prompt a VLM to generate object-level descriptions by feeding it object regions from existing object detection datasets. We further propose attribute modeling to explicitly capture the important object attributes, and spatial relation modeling to capture inter-object relationship, both of which are common linguistic pattern in referring expression. Our constructed dataset (500K images, 1M objects, 16M referring expressions) is one of the largest grounding datasets to date, and the first grounding dataset with purely model-generated queries and human-annotated objects. To verify the quality of this data, we conduct zero-shot transfer experiments to the popular RefCOCO benchmarks for both referring expression comprehension (REC) and segmentation (RES) tasks. On both tasks, our model significantly outperform the state-of-the-art approaches without using human annotated visual grounding data. Our results demonstrate the promise of generative VLM to scale up visual grounding in the real world. Code and models will be released.
Authors: Jialiang Li, Haoyue Wang, Sheng Li, Zhenxing Qian, Xinpeng Zhang, Athanasios V. Vasilakos
Abstract: Recently, a vast number of image generation models have been proposed, which raises concerns regarding the misuse of these artificial intelligence (AI) techniques for generating fake images. To attribute the AI-generated images, existing schemes usually design and train deep neural networks (DNNs) to learn the model fingerprints, which usually requires a large amount of data for effective learning. In this paper, we aim to answer the following two questions for AI-generated image attribution, 1) is it possible to design useful handcrafted filters to facilitate the fingerprint learning? and 2) how we could reduce the amount of training data after we incorporate the handcrafted filters? We first propose a set of Multi-Directional High-Pass Filters (MHFs) which are capable to extract the subtle fingerprints from various directions. Then, we propose a Directional Enhanced Feature Learning network (DEFL) to take both the MHFs and randomly-initialized filters into consideration. The output of the DEFL is fused with the semantic features to produce a compact fingerprint. To make the compact fingerprint discriminative among different models, we propose a Dual-Margin Contrastive (DMC) loss to tune our DEFL. Finally, we propose a reference based fingerprint classification scheme for image attribution. Experimental results demonstrate that it is indeed helpful to use our MHFs for attributing the AI-generated images. The performance of our proposed method is significantly better than the state-of-the-art for both the closed-set and open-set image attribution, where only a small amount of images are required for training.
Authors: Samay Nathani
Abstract: Emotion recognition is a critical aspect of human interaction. This topic garnered significant attention in the field of artificial intelligence. In this study, we investigate the performance of convolutional neural network (CNN) and Modified VGG16 models for emotion recognition tasks across two datasets: FER2013 and AffectNet. Our aim is to measure the effectiveness of these models in identifying emotions and their ability to generalize to different and broader datasets. Our findings reveal that both models achieve reasonable performance on the FER2013 dataset, with the Modified VGG16 model demonstrating slightly increased accuracy. When evaluated on the Affect-Net dataset, performance declines for both models, with the Modified VGG16 model continuing to outperform the CNN. Our study emphasizes the importance of dataset diversity in emotion recognition and discusses open problems and future research directions, including the exploration of multi-modal approaches and the development of more comprehensive datasets.
Authors: Luke Bidulka, Mohsen Gholami, Jiannan Zheng, Martin J. McKeown, Z. Jane Wang
Abstract: Despite recent advances in human pose estimation (HPE), poor generalization to out-of-distribution (OOD) data remains a difficult problem. While previous works have proposed Test-Time Adaptation (TTA) to bridge the train-test domain gap by refining network parameters at inference, the absence of ground-truth annotations makes it highly challenging and existing methods typically increase inference times by one or more orders of magnitude. We observe that 1) not every test time sample is OOD, and 2) HPE errors are significantly larger on distal keypoints (wrist, ankle). To this end, we propose ESCAPE: a lightweight correction and selective adaptation framework which applies a fast, forward-pass correction on most data while reserving costly TTA for OOD data. The free energy function is introduced to separate OOD samples from incoming data and a correction network is trained to estimate the errors of pretrained backbone HPE predictions on the distal keypoints. For OOD samples, we propose a novel self-consistency adaptation loss to update the correction network by leveraging the constraining relationship between distal keypoints and proximal keypoints (shoulders, hips), via a second ``reverse" network. ESCAPE improves the distal MPJPE of five popular HPE models by up to 7% on unseen data, achieves state-of-the-art results on two popular HPE benchmarks, and is significantly faster than existing adaptation methods.
Authors: Hao Xiao
Abstract: Object re-identification is of increasing importance in visual surveillance. Most existing works focus on re-identify individual from multiple cameras while the application of group re-identification (Re-ID) is rarely discussed. We redefine Group Re-identification as a process which includes pedestrian detection, feature extraction, graph model construction, and graph matching. Group re-identification is very challenging since it is not only interfered by view-point and human pose variations in the traditional re-identification tasks, but also suffered from the challenges in group layout change and group member variation. To address the above challenges, this paper introduces a novel approach which leverages the multi-granularity information inside groups to facilitate group re-identification. We first introduce a multi-granularity Re-ID process, which derives features for multi-granularity objects (people/people-subgroups) in a group and iteratively evaluates their importances during group Re-ID, so as to handle group-wise misalignments due to viewpoint change and group dynamics. We further introduce a multi-order matching scheme. It adaptively selects representative people/people-subgroups in each group and integrates the multi-granularity information from these people/people-subgroups to obtain group-wise matching, hence achieving a more reliable matching score between groups. Experimental results on various datasets demonstrate the effectiveness of our approach.
Authors: Srivishnu Vusirikala, Suraj Rajendran
Abstract: We investigate the potential of self-supervision in improving the accuracy of deep learning models trained to classify melanoma patches. Various self-supervision techniques such as rotation prediction, missing patch prediction, and corruption removal were implemented and assessed for their impact on the convolutional neural network's performance. Preliminary results suggest a positive influence of self-supervision methods on the model's accuracy. The study notably demonstrates the efficacy of the corruption removal method in enhancing model performance. Despite observable improvements, we conclude that the self-supervised models have considerable potential for further enhancement, achievable through training over more epochs or expanding the dataset. We suggest exploring other self-supervision methods like Bootstrap Your Own Latent (BYOL) and contrastive learning in future research, emphasizing the cost-benefit trade-off due to their resource-intensive nature. The findings underline the promise of self-supervision in augmenting melanoma detection capabilities of deep learning models.
Authors: Shivansh Sharma, Mathew Huang, Sanat Nair, Alan Wen, Christina Petlowany, Juston Moore, Selma Wanna, Mitch Pryor
Abstract: Industry 4.0 introduced AI as a transformative solution for modernizing manufacturing processes. Its successor, Industry 5.0, envisions humans as collaborators and experts guiding these AI-driven manufacturing solutions. Developing these techniques necessitates algorithms capable of safe, real-time identification of human positions in a scene, particularly their hands, during collaborative assembly. Although substantial efforts have curated datasets for hand segmentation, most focus on residential or commercial domains. Existing datasets targeting industrial settings predominantly rely on synthetic data, which we demonstrate does not effectively transfer to real-world operations. Moreover, these datasets lack uncertainty estimations critical for safe collaboration. Addressing these gaps, we present HAGS: Hand and Glove Segmentation Dataset. This dataset provides 1200 challenging examples to build applications toward hand and glove segmentation in industrial human-robot collaboration scenarios as well as assess out-of-distribution images, constructed via green screen augmentations, to determine ML-classifier robustness. We study state-of-the-art, real-time segmentation models to evaluate existing methods. Our dataset and baselines are publicly available: https://dataverse.tdl.org/dataset.xhtml?persistentId=doi:10.18738/T8/85R7KQ and https://github.com/UTNuclearRoboticsPublic/assembly_glovebox_dataset.
URLs: https://dataverse.tdl.org/dataset.xhtml?persistentId=doi:10.18738/T8/85R7KQ, https://github.com/UTNuclearRoboticsPublic/assembly_glovebox_dataset.
Authors: Soroush Oraki, Harry Zhuang, Jie Liang
Abstract: The complexity of state-of-the-art Transformer-based models for skeleton-based action recognition poses significant challenges in terms of computational efficiency and resource utilization. In this paper, we explore the application of Singular Value Decomposition (SVD) to effectively reduce the model sizes of these pre-trained models, aiming to minimize their resource consumption while preserving accuracy. Our method, LORTSAR (LOw-Rank Transformer for Skeleton-based Action Recognition), also includes a fine-tuning step to compensate for any potential accuracy degradation caused by model compression, and is applied to two leading Transformer-based models, "Hyperformer" and "STEP-CATFormer". Experimental results on the "NTU RGB+D" and "NTU RGB+D 120" datasets show that our method can reduce the number of model parameters substantially with negligible degradation or even performance increase in recognition accuracy. This confirms that SVD combined with post-compression fine-tuning can boost model efficiency, paving the way for more sustainable, lightweight, and high-performance technologies in human action recognition.
Authors: Ayman Beghdadi, Azeddine Beghdadi, Mohib Ullah, Faouzi Alaya Cheikh, Malik Mallem
Abstract: Scene understanding plays an important role in several high-level computer vision applications, such as autonomous vehicles, intelligent video surveillance, or robotics. However, too few solutions have been proposed for indoor/outdoor scene classification to ensure scene context adaptability for computer vision frameworks. We propose the first Lightweight Hybrid Graph Convolutional Neural Network (LH-GCNN)-CNN framework as an add-on to object detection models. The proposed approach uses the output of the CNN object detection model to predict the observed scene type by generating a coherent GCNN representing the semantic and geometric content of the observed scene. This new method, applied to natural scenes, achieves an efficiency of over 90\% for scene classification in a COCO-derived dataset containing a large number of different scenes, while requiring fewer parameters than traditional CNN methods. For the benefit of the scientific community, we will make the source code publicly available: https://github.com/Aymanbegh/Hybrid-GCNN-CNN.
Authors: Zihu Wang, Lingqiao Liu, Scott Ricardo Figueroa Weston, Samuel Tian, Peng Li
Abstract: Self-Supervised Learning (SSL) has become a prominent approach for acquiring visual representations across various tasks, yet its application in fine-grained visual recognition (FGVR) is challenged by the intricate task of distinguishing subtle differences between categories. To overcome this, we introduce an novel strategy that boosts SSL's ability to extract critical discriminative features vital for FGVR. This approach creates synthesized data pairs to guide the model to focus on discriminative features critical for FGVR during SSL. We start by identifying non-discriminative features using two main criteria: features with low variance that fail to effectively separate data and those deemed less important by Grad-CAM induced from the SSL loss. We then introduce perturbations to these non-discriminative features while preserving discriminative ones. A decoder is employed to reconstruct images from both perturbed and original feature vectors to create data pairs. An encoder is trained on such generated data pairs to become invariant to variations in non-discriminative dimensions while focusing on discriminative features, thereby improving the model's performance in FGVR tasks. We demonstrate the promising FGVR performance of the proposed approach through extensive evaluation on a wide variety of datasets.
Authors: Minh-Quan Le, Alexandros Graikos, Srikar Yellapragada, Rajarsi Gupta, Joel Saltz, Dimitris Samaras
Abstract: Synthesizing high-resolution images from intricate, domain-specific information remains a significant challenge in generative modeling, particularly for applications in large-image domains such as digital histopathology and remote sensing. Existing methods face critical limitations: conditional diffusion models in pixel or latent space cannot exceed the resolution on which they were trained without losing fidelity, and computational demands increase significantly for larger image sizes. Patch-based methods offer computational efficiency but fail to capture long-range spatial relationships due to their overreliance on local information. In this paper, we introduce a novel conditional diffusion model in infinite dimensions, $\infty$-Brush for controllable large image synthesis. We propose a cross-attention neural operator to enable conditioning in function space. Our model overcomes the constraints of traditional finite-dimensional diffusion models and patch-based methods, offering scalability and superior capability in preserving global image structures while maintaining fine details. To our best knowledge, $\infty$-Brush is the first conditional diffusion model in function space, that can controllably synthesize images at arbitrary resolutions of up to $4096\times4096$ pixels. The code is available at https://github.com/cvlab-stonybrook/infinity-brush.
Authors: Ryo Fujii, Ryo Hachiuma, Hideo Saito
Abstract: A crowd density forecasting task aims to predict how the crowd density map will change in the future from observed past crowd density maps. However, the past crowd density maps are often incomplete due to the miss-detection of pedestrians, and it is crucial to develop a robust crowd density forecasting model against the miss-detection. This paper presents a MAsked crowd density Completion framework for crowd density forecasting (CrowdMAC), which is simultaneously trained to forecast future crowd density maps from partially masked past crowd density maps (i.e., forecasting maps from past maps with miss-detection) while reconstructing the masked observation maps (i.e., imputing past maps with miss-detection). Additionally, we propose Temporal-Density-aware Masking (TDM), which non-uniformly masks tokens in the observed crowd density map, considering the sparsity of the crowd density maps and the informativeness of the subsequent frames for the forecasting task. Moreover, we introduce multi-task masking to enhance training efficiency. In the experiments, CrowdMAC achieves state-of-the-art performance on seven large-scale datasets, including SDD, ETH-UCY, inD, JRDB, VSCrowd, FDST, and croHD. We also demonstrate the robustness of the proposed method against both synthetic and realistic miss-detections.
Authors: Cuong Pham, Hoang Anh Dung, Cuong C. Nguyen, Trung Le, Dinh Phung, Gustavo Carneiro, Thanh-Toan Do
Abstract: Post-Training Quantization (PTQ) has received significant attention because it requires only a small set of calibration data to quantize a full-precision model, which is more practical in real-world applications in which full access to a large training set is not available. However, it often leads to overfitting on the small calibration dataset. Several methods have been proposed to address this issue, yet they still rely on only the calibration set for the quantization and they do not validate the quantized model due to the lack of a validation set. In this work, we propose a novel meta-learning based approach to enhance the performance of post-training quantization. Specifically, to mitigate the overfitting problem, instead of only training the quantized model using the original calibration set without any validation during the learning process as in previous PTQ works, in our approach, we both train and validate the quantized model using two different sets of images. In particular, we propose a meta-learning based approach to jointly optimize a transformation network and a quantized model through bi-level optimization. The transformation network modifies the original calibration data and the modified data will be used as the training set to learn the quantized model with the objective that the quantized model achieves a good performance on the original calibration data. Extensive experiments on the widely used ImageNet dataset with different neural network architectures demonstrate that our approach outperforms the state-of-the-art PTQ methods.
Authors: Angelly Cabrera, Kleanthis Avramidis, Shrikanth Narayanan
Abstract: Coffee leaf rust, a foliar disease caused by the fungus Hemileia vastatrix, poses a major threat to coffee production, especially in Central America. Climate change further aggravates this issue, as it shortens the latency period between initial infection and the emergence of visible symptoms in diseases like leaf rust. Shortened latency periods can lead to more severe plant epidemics and faster spread of diseases. There is, hence, an urgent need for effective disease management strategies. To address these challenges, we explore the potential of deep learning models for enhancing early disease detection. However, deep learning models require extensive processing power and large amounts of data for model training, resources that are typically scarce. To overcome these barriers, we propose a preprocessing technique that involves convolving training images with a high-pass filter to enhance lesion-leaf contrast, significantly improving model efficacy in resource-limited environments. This method and our model demonstrated a strong performance, achieving over 90% across all evaluation metrics--including precision, recall, F1-score, and the Dice coefficient. Our experiments show that this approach outperforms other methods, including two different image preprocessing techniques and using unaltered, full-color images.
Authors: Yuyang Wanyan, Xiaoshan Yang, Weiming Dong, Changsheng Xu
Abstract: Few-shot action recognition aims to address the high cost and impracticality of manually labeling complex and variable video data in action recognition. It requires accurately classifying human actions in videos using only a few labeled examples per class. Compared to few-shot learning in image scenarios, few-shot action recognition is more challenging due to the intrinsic complexity of video data. Recognizing actions involves modeling intricate temporal sequences and extracting rich semantic information, which goes beyond mere human and object identification in each frame. Furthermore, the issue of intra-class variance becomes particularly pronounced with limited video samples, complicating the learning of representative features for novel action categories. To overcome these challenges, numerous approaches have driven significant advancements in few-shot action recognition, which underscores the need for a comprehensive survey. Unlike early surveys that focus on few-shot image or text classification, we deeply consider the unique challenges of few-shot action recognition. In this survey, we review a wide variety of recent methods and summarize the general framework. Additionally, the survey presents the commonly used benchmarks and discusses relevant advanced topics and promising future directions. We hope this survey can serve as a valuable resource for researchers, offering essential guidance to newcomers and stimulating seasoned researchers with fresh insights.
Authors: Tianwen Zhou, Qihao Duan, Zitong Yu
Abstract: The recovery of high-quality images from images corrupted by lens flare presents a significant challenge in low-level vision. Contemporary deep learning methods frequently entail training a lens flare removing model from scratch. However, these methods, despite their noticeable success, fail to utilize the generative prior learned by pre-trained models, resulting in unsatisfactory performance in lens flare removal. Furthermore, there are only few works considering the physical priors relevant to flare removal. To address these issues, we introduce Difflare, a novel approach designed for lens flare removal. To leverage the generative prior learned by Pre-Trained Diffusion Models (PTDM), we introduce a trainable Structural Guidance Injection Module (SGIM) aimed at guiding the restoration process with PTDM. Towards more efficient training, we employ Difflare in the latent space. To address information loss resulting from latent compression and the stochastic sampling process of PTDM, we introduce an Adaptive Feature Fusion Module (AFFM), which incorporates the Luminance Gradient Prior (LGP) of lens flare to dynamically regulate feature extraction. Extensive experiments demonstrate that our proposed Difflare achieves state-of-the-art performance in real-world lens flare removal, restoring images corrupted by flare with improved fidelity and perceptual quality. The codes will be released soon.
Authors: Jayanth Mohan, Arrun Sivasubramanian, V Sowmya, Ravi Vinayakumar
Abstract: Skin diseases affect over a third of the global population, yet their impact is often underestimated. Automating skin disease classification to assist doctors with their prognosis might be difficult. Nevertheless, due to efficient feature extraction pipelines, deep learning techniques have shown much promise for various tasks, including dermatological disease identification. This study uses a skin disease dataset with 31 classes and compares it with all versions of Vision Transformers, Swin Transformers and DivoV2. The analysis is also extended to compare with benchmark convolution-based architecture presented in the literature. Transfer learning with ImageNet1k weights on the skin disease dataset contributes to a high test accuracy of 96.48\% and an F1-Score of 0.9727 using DinoV2, which is almost a 10\% improvement over this data's current benchmark results. The performance of DinoV2 was also compared for the HAM10000 and Dermnet datasets to test the model's robustness, and the trained model overcomes the benchmark results by a slight margin in test accuracy and in F1-Score on the 23 and 7 class datasets. The results are substantiated using explainable AI frameworks like GradCAM and SHAP, which provide precise image locations to map the disease, assisting dermatologists in early detection, prompt prognosis, and treatment.
Authors: Xinyu Xu, Shengcheng Luo, Yanchao Yang, Yong-Lu Li, Cewu Lu
Abstract: Building a general-purpose intelligent home-assistant agent skilled in diverse tasks by human commands is a long-term blueprint of embodied AI research, which poses requirements on task planning, environment modeling, and object interaction. In this work, we study primitive mobile manipulations for embodied agents, i.e. how to navigate and interact based on an instructed verb-noun pair. We propose DISCO, which features non-trivial advancements in contextualized scene modeling and efficient controls. In particular, DISCO incorporates differentiable scene representations of rich semantics in object and affordance, which is dynamically learned on the fly and facilitates navigation planning. Besides, we propose dual-level coarse-to-fine action controls leveraging both global and local cues to accomplish mobile manipulation tasks efficiently. DISCO easily integrates into embodied tasks such as embodied instruction following. To validate our approach, we take the ALFRED benchmark of large-scale long-horizon vision-language navigation and interaction tasks as a test bed. In extensive experiments, we make comprehensive evaluations and demonstrate that DISCO outperforms the art by a sizable +8.6% success rate margin in unseen scenes, even without step-by-step instructions. Our code is publicly released at https://github.com/AllenXuuu/DISCO.
Authors: Aryan Singh, Pepijn Van de Ven, Ciar\'an Eising, Patrick Denny
Abstract: In this study, we present the Graph Sub-Graph Network (GSN), a novel hybrid image classification model merging the strengths of Convolutional Neural Networks (CNNs) for feature extraction and Graph Neural Networks (GNNs) for structural modeling. GSN employs k-means clustering to group graph nodes into clusters, facilitating the creation of subgraphs. These subgraphs are then utilized to learn representative `atoms` for dictionary learning, enabling the identification of sparse, class-distinguishable features. This integrated approach is particularly relevant in domains like medical imaging, where discerning subtle feature differences is crucial for accurate classification. To evaluate the performance of our proposed GSN, we conducted experiments on benchmark datasets, including PascalVOC and HAM10000. Our results demonstrate the efficacy of our model in optimizing dictionary configurations across varied classes, which contributes to its effectiveness in medical classification tasks. This performance enhancement is primarily attributed to the integration of CNNs, GNNs, and graph learning techniques, which collectively improve the handling of datasets with limited labeled examples. Specifically, our experiments show that the model achieves a higher accuracy on benchmark datasets such as Pascal VOC and HAM10000 compared to conventional CNN approaches.
Authors: Yuhang Bai, Zichuan Huang, Wenshuo Gao, Shuai Yang, Jiaying Liu
Abstract: Artistic text generation aims to amplify the aesthetic qualities of text while maintaining readability. It can make the text more attractive and better convey its expression, thus enjoying a wide range of application scenarios such as social media display, consumer electronics, fashion, and graphic design. Artistic text generation includes artistic text stylization and semantic typography. Artistic text stylization concentrates on the text effect overlaid upon the text, such as shadows, outlines, colors, glows, and textures. By comparison, semantic typography focuses on the deformation of the characters to strengthen their visual representation by mimicking the semantic understanding within the text. This overview paper provides an introduction to both artistic text stylization and semantic typography, including the taxonomy, the key ideas of representative methods, and the applications in static and dynamic artistic text generation. Furthermore, the dataset and evaluation metrics are introduced, and the future directions of artistic text generation are discussed. A comprehensive list of artistic text generation models studied in this review is available at https://github.com/williamyang1991/Awesome-Artistic-Typography/.
URLs: https://github.com/williamyang1991/Awesome-Artistic-Typography/.
Authors: Ahmed Radwan, Mohamed S. Shehata
Abstract: Federated Domain Generalization (FedDG), aims to tackle the challenge of generalizing to unseen domains at test time while catering to the data privacy constraints that prevent centralized data storage from different domains originating at various clients. Existing approaches can be broadly categorized into four groups: domain alignment, data manipulation, learning strategies, and optimization of model aggregation weights. This paper proposes a novel approach to Federated Domain Generalization that tackles the problem from the perspective of the backbone model architecture. The core principle is that objects, even under substantial domain shifts and appearance variations, maintain a consistent hierarchical structure of parts and wholes. For instance, a photograph and a sketch of a dog share the same hierarchical organization, consisting of a head, body, limbs, and so on. The introduced architecture explicitly incorporates a feature representation for the image parse tree. To the best of our knowledge, this is the first work to tackle Federated Domain Generalization from a model architecture standpoint. Our approach outperforms a convolutional architecture of comparable size by over 12\%, despite utilizing fewer parameters. Additionally, it is inherently interpretable, contrary to the black-box nature of CNNs, which fosters trust in its predictions, a crucial asset in federated learning.
Authors: Junjie Shi, Caozhi Shang, Zhaobin Sun, Li Yu, Xin Yang, Zengqiang Yan
Abstract: Incomplete multi-modal image segmentation is a fundamental task in medical imaging to refine deployment efficiency when only partial modalities are available. However, the common practice that complete-modality data is visible during model training is far from realistic, as modalities can have imbalanced missing rates in clinical scenarios. In this paper, we, for the first time, formulate such a challenging setting and propose Preference-Aware Self-diStillatION (PASSION) for incomplete multi-modal medical image segmentation under imbalanced missing rates. Specifically, we first construct pixel-wise and semantic-wise self-distillation to balance the optimization objective of each modality. Then, we define relative preference to evaluate the dominance of each modality during training, based on which to design task-wise and gradient-wise regularization to balance the convergence rates of different modalities. Experimental results on two publicly available multi-modal datasets demonstrate the superiority of PASSION against existing approaches for modality balancing. More importantly, PASSION is validated to work as a plug-and-play module for consistent performance improvement across different backbones. Code is available at https://github.com/Jun-Jie-Shi/PASSION.
Authors: Bowei Tian, Ruijie Du, Yanning Shen
Abstract: Vision Transformer (ViT) has achieved excellent performance and demonstrated its promising potential in various computer vision tasks. The wide deployment of ViT in real-world tasks requires a thorough understanding of the societal impact of the model. However, most ViT-based works do not take fairness into account and it is unclear whether directly applying CNN-oriented debiased algorithm to ViT is feasible. Moreover, previous works typically sacrifice accuracy for fairness. Therefore, we aim to develop an algorithm that improves accuracy without sacrificing fairness. In this paper, we propose FairViT, a novel accurate and fair ViT framework. To this end, we introduce a novel distance loss and deploy adaptive fairness-aware masks on attention layers updating with model parameters. Experimental results show \sys can achieve accuracy better than other alternatives, even with competitive computational efficiency. Furthermore, \sys achieves appreciable fairness results.
Authors: Di Fu, Thanh Vinh Vo, Haozhe Ma, Tze-Yun Leong
Abstract: Action recognition technology plays a vital role in enhancing security through surveillance systems, enabling better patient monitoring in healthcare, providing in-depth performance analysis in sports, and facilitating seamless human-AI collaboration in domains such as manufacturing and assistive technologies. The dynamic nature of data in these areas underscores the need for models that can continuously adapt to new video data without losing previously acquired knowledge, highlighting the critical role of advanced continual action recognition. To address these challenges, we propose Decoupled Prompt-Adapter Tuning (DPAT), a novel framework that integrates adapters for capturing spatial-temporal information and learnable prompts for mitigating catastrophic forgetting through a decoupled training strategy. DPAT uniquely balances the generalization benefits of prompt tuning with the plasticity provided by adapters in pretrained vision models, effectively addressing the challenge of maintaining model performance amidst continuous data evolution without necessitating extensive finetuning. DPAT consistently achieves state-of-the-art performance across several challenging action recognition benchmarks, thus demonstrating the effectiveness of our model in the domain of continual action recognition.
Authors: Fanxu Min, Shaoxiang Guo, Fan Hao, Junyu Dong
Abstract: Gait recognition is a biometric technology that recognizes the identity of humans through their walking patterns. Existing appearance-based methods utilize CNN or Transformer to extract spatial and temporal features from silhouettes, while model-based methods employ GCN to focus on the special topological structure of skeleton points. However, the quality of silhouettes is limited by complex occlusions, and skeletons lack dense semantic features of the human body. To tackle these problems, we propose a novel gait recognition framework, dubbed Gait Multi-model Aggregation Network (GaitMA), which effectively combines two modalities to obtain a more robust and comprehensive gait representation for recognition. First, skeletons are represented by joint/limb-based heatmaps, and features from silhouettes and skeletons are respectively extracted using two CNN-based feature extractors. Second, a co-attention alignment module is proposed to align the features by element-wise attention. Finally, we propose a mutual learning module, which achieves feature fusion through cross-attention, Wasserstein loss is further introduced to ensure the effective fusion of two modalities. Extensive experimental results demonstrate the superiority of our model on Gait3D, OU-MVLP, and CASIA-B.
Authors: Jiangtao Zhang, Zongsheng Yue, Hui Wang, Qian Zhao, Deyu Meng
Abstract: Blind image deconvolution (BID) is a classic yet challenging problem in the field of image processing. Recent advances in deep image prior (DIP) have motivated a series of DIP-based approaches, demonstrating remarkable success in BID. However, due to the high non-convexity of the inherent optimization process, these methods are notorious for their sensitivity to the initialized kernel. To alleviate this issue and further improve their performance, we propose a new framework for BID that better considers the prior modeling and the initialization for blur kernels, leveraging a deep generative model. The proposed approach pre-trains a generative adversarial network-based kernel generator that aptly characterizes the kernel priors and a kernel initializer that facilitates a well-informed initialization for the blur kernel through latent space encoding. With the pre-trained kernel generator and initializer, one can obtain a high-quality initialization of the blur kernel, and enable optimization within a compact latent kernel manifold. Such a framework results in an evident performance improvement over existing DIP-based BID methods. Extensive experiments on different datasets demonstrate the effectiveness of the proposed method.
Authors: Yukai Shi, Zhipeng Weng, Yupei Lin, Cidan Shi, Xiaojun Yang, Liang Lin
Abstract: In recent years, as computer vision tasks have increasingly relied on high-quality image inputs, the task of image dehazing has received significant attention. Previously, many methods based on priors and deep learning have been proposed to address the task of image dehazing. Ignoring the domain gap between different data, former de-hazing methods usually adopt multiple datasets for explicit training, which often makes the methods themselves be violated. To address this problem, we propose a novel method of internal and external data augmentation to improve the existing dehazing methodology. By using cross-data external augmentor. The dataset inherits samples from different domains that are firmly aligned, making the model learn more robust and generalizable features. By using the internal data augmentation method, the model can fully exploit local information within the images, thereby obtaining more image details. To demonstrate the effectiveness of our proposed method, we conduct training on both the Natural Image Dataset (NID) and the Remote Sensing Image Dataset (RSID). Experimental results show that our method clearly resolves the domain gap in different dehazing datasets and presents a new pipeline for joint training in the dehazing task. Our approach significantly outperforms other advanced methods in dehazing and produces dehazed images that are closest to real haze-free images. The code will be available at: https://github.com/wengzp1/ScaleUpDehazing
Authors: Christos Kyrkou
Abstract: High-efficiency deep learning (DL) models are necessary not only to facilitate their use in devices with limited resources but also to improve resources required for training. Convolutional neural networks (ConvNets) typically exert severe demands on local device resources and this conventionally limits their adoption within mobile and embedded platforms. This brief presents work toward utilizing static convolutional filters generated from the space of local binary patterns (LBPs) and Haar features to design efficient ConvNet architectures. These are referred to as Structured Ternary Patterns (STePs) and can be generated during network initialization in a systematic way instead of having learnable weight parameters thus reducing the total weight updates. The ternary values require significantly less storage and with the appropriate low-level implementation, can also lead to inference improvements. The proposed approach is validated using four image classification datasets, demonstrating that common network backbones can be made more efficient and provide competitive results. It is also demonstrated that it is possible to generate completely custom STeP-based networks that provide good trade-offs for on-device applications such as unmanned aerial vehicle (UAV)-based aerial vehicle detection. The experimental results show that the proposed method maintains high detection accuracy while reducing the trainable parameters by 40-80%. This work motivates further research toward good priors for non-learnable weights that can make DL architectures more efficient without having to alter the network during or after training.
Authors: Harsh Lunia
Abstract: Recent advancements have introduced multiple vision-language models (VLMs) demonstrating impressive commonsense reasoning across various domains. Despite their individual capabilities, the potential of synergizing these complementary VLMs remains underexplored. The Cola Framework addresses this by showcasing how a large language model (LLM) can efficiently coordinate multiple VLMs through natural language communication, leveraging their distinct strengths. We have verified this claim on the challenging A-OKVQA dataset, confirming the effectiveness of such coordination. Building on this, our study investigates whether the same methodology can be applied to surveillance videos for action recognition. Specifically, we explore if leveraging the combined knowledge base of VLMs and LLM can effectively deduce actions from a video when presented with only a few selectively important frames and minimal temporal information. Our experiments demonstrate that LLM, when coordinating different VLMs, can successfully recognize patterns and deduce actions in various scenarios despite the weak temporal signals. However, our findings suggest that to enhance this approach as a viable alternative solution, integrating a stronger temporal signal and exposing the models to slightly more frames would be beneficial.
Authors: Bo Han, Heqing Zou, Haoyang Li, Guangcong Wang, Chng Eng Siong
Abstract: Text-based talking-head video editing aims to efficiently insert, delete, and substitute segments of talking videos through a user-friendly text editing approach. It is challenging because of \textbf{1)} generalizable talking-face representation, \textbf{2)} seamless audio-visual transitions, and \textbf{3)} identity-preserved talking faces. Previous works either require minutes of talking-face video training data and expensive test-time optimization for customized talking video editing or directly generate a video sequence without considering in-context information, leading to a poor generalizable representation, or incoherent transitions, or even inconsistent identity. In this paper, we propose an efficient cascaded conditional diffusion-based framework, which consists of two stages: audio to dense-landmark motion and motion to video. \textit{\textbf{In the first stage}}, we first propose a dynamic weighted in-context diffusion module to synthesize dense-landmark motions given an edited audio. \textit{\textbf{In the second stage}}, we introduce a warping-guided conditional diffusion module. The module first interpolates between the start and end frames of the editing interval to generate smooth intermediate frames. Then, with the help of the audio-to-dense motion images, these intermediate frames are warped to obtain coarse intermediate frames. Conditioned on the warped intermedia frames, a diffusion model is adopted to generate detailed and high-resolution target frames, which guarantees coherent and identity-preserved transitions. The cascaded conditional diffusion model decomposes the complex talking editing task into two flexible generation tasks, which provides a generalizable talking-face representation, seamless audio-visual transitions, and identity-preserved faces on a small dataset. Experiments show the effectiveness and superiority of the proposed method.
Authors: Tianle Zeng, Gerardo Loza Galindo, Junlei Hu, Pietro Valdastri, Dominic Jones
Abstract: Computer vision technologies markedly enhance the automation capabilities of robotic-assisted minimally invasive surgery (RAMIS) through advanced tool tracking, detection, and localization. However, the limited availability of comprehensive surgical datasets for training represents a significant challenge in this field. This research introduces a novel method that employs 3D Gaussian Splatting to generate synthetic surgical datasets. We propose a method for extracting and combining 3D Gaussian representations of surgical instruments and background operating environments, transforming and combining them to generate high-fidelity synthetic surgical scenarios. We developed a data recording system capable of acquiring images alongside tool and camera poses in a surgical scene. Using this pose data, we synthetically replicate the scene, thereby enabling direct comparisons of the synthetic image quality (29.592 PSNR). As a further validation, we compared two YOLOv5 models trained on the synthetic and real data, respectively, and assessed their performance in an unseen real-world test dataset. Comparing the performances, we observe an improvement in neural network performance, with the synthetic-trained model outperforming the real-world trained model by 12%, testing both on real-world data.
Authors: Yueyang Li, Weiming Zeng, Wenhao Dong, Di Han, Lei Chen, Hongyu Chen, Hongjie Yan, Wai Ting Siok, Nizhuan Wang
Abstract: Single-channel electroencephalogram (EEG) is a cost-effective, comfortable, and non-invasive method for monitoring brain activity, widely adopted by researchers, consumers, and clinicians. The increasing number and proportion of articles on single-channel EEG underscore its growing potential. This paper provides a comprehensive review of single-channel EEG, focusing on development trends, devices, datasets, signal processing methods, recent applications, and future directions. Definitions of bipolar and unipolar configurations in single-channel EEG are clarified to guide future advancements. Applications mainly span sleep staging, emotion recognition, educational research, and clinical diagnosis. Ongoing advancements of single-channel EEG in AI-based EEG generation techniques suggest potential parity or superiority over multichannel EEG performance.
Authors: Maximilian E. Tschuchnig, Philipp Steininger, Michael Gadermayr
Abstract: Medical imaging is vital in computer assisted intervention. Particularly cone beam computed tomography (CBCT) with defacto real time and mobility capabilities plays an important role. However, CBCT images often suffer from artifacts, which pose challenges for accurate interpretation, motivating research in advanced algorithms for more effective use in clinical practice. In this work we present CBCTLiTS, a synthetically generated, labelled CBCT dataset for segmentation with paired and aligned, high quality computed tomography data. The CBCT data is provided in 5 different levels of quality, reaching from a large number of projections with high visual quality and mild artifacts to a small number of projections with severe artifacts. This allows thorough investigations with the quality as a degree of freedom. We also provide baselines for several possible research scenarios like uni- and multimodal segmentation, multitask learning and style transfer followed by segmentation of relatively simple, liver to complex liver tumor segmentation. CBCTLiTS is accesssible via https://www.kaggle.com/datasets/maximiliantschuchnig/cbct-liver-and-liver-tumor-segmentation-train-data.
Authors: Francesco Di Luzio, Antonello Rosato, Massimo Panella
Abstract: In the context of artificial intelligence, the inherent human attribute of engaging in logical reasoning to facilitate decision-making is mirrored by the concept of explainability, which pertains to the ability of a model to provide a clear and interpretable account of how it arrived at a particular outcome. This study explores explainability techniques for binary deep neural architectures in the framework of emotion classification through video analysis. We investigate the optimization of input features to binary classifiers for emotion recognition, with face landmarks detection using an improved version of the Integrated Gradients explainability method. The main contribution of this paper consists in the employment of an innovative explainable artificial intelligence algorithm to understand the crucial facial landmarks movements during emotional feeling, using this information also for improving the performances of deep learning-based emotion classifiers. By means of explainability, we can optimize the number and the position of the facial landmarks used as input features for facial emotion recognition, lowering the impact of noisy landmarks and thus increasing the accuracy of the developed models. In order to test the effectiveness of the proposed approach, we considered a set of deep binary models for emotion classification trained initially with a complete set of facial landmarks, which are progressively reduced based on a suitable optimization procedure. The obtained results prove the robustness of the proposed explainable approach in terms of understanding the relevance of the different facial points for the different emotions, also improving the classification accuracy and diminishing the computational cost.
Authors: Yuemei Li, Guojia Hou, Peixian Zhuang, Zhenkuan Pan
Abstract: Underwater images are typically characterized by color cast, haze, blurring, and uneven illumination due to the selective absorption and scattering when light propagates through the water, which limits their practical applications. Underwater image enhancement and restoration (UIER) is one crucial mode to improve the visual quality of underwater images. However, most existing UIER methods concentrate on enhancing contrast and dehazing, rarely pay attention to the local illumination differences within the image caused by illumination variations, thus introducing some undesirable artifacts and unnatural color. To address this issue, an effective variational framework is proposed based on an extended underwater image formation model (UIFM). Technically, dual high-order regularizations are successfully integrated into the variational model to acquire smoothed local ambient illuminance and structure-revealed reflectance in a unified manner. In our proposed framework, the weight factors-based color compensation is combined with the color balance to compensate for the attenuated color channels and remove the color cast. In particular, the local ambient illuminance with strong robustness is acquired by performing the local patch brightest pixel estimation and an improved gamma correction. Additionally, we design an iterative optimization algorithm relying on the alternating direction method of multipliers (ADMM) to accelerate the solution of the proposed variational model. Considerable experiments on three real-world underwater image datasets demonstrate that the proposed method outperforms several state-of-the-art methods with regard to visual quality and quantitative assessments. Moreover, the proposed method can also be extended to outdoor image dehazing, low-light image enhancement, and some high-level vision tasks. The code is available at https://github.com/Hou-Guojia/UDHTV.
Authors: Yanting Yang, Minghao Chen, Qibo Qiu, Jiahao Wu, Wenxiao Wang, Binbin Lin, Ziyu Guan, Xiaofei He
Abstract: For a general-purpose robot to operate in reality, executing a broad range of instructions across various environments is imperative. Central to the reinforcement learning and planning for such robotic agents is a generalizable reward function. Recent advances in vision-language models, such as CLIP, have shown remarkable performance in the domain of deep learning, paving the way for open-domain visual recognition. However, collecting data on robots executing various language instructions across multiple environments remains a challenge. This paper aims to transfer video-language models with robust generalization into a generalizable language-conditioned reward function, only utilizing robot video data from a minimal amount of tasks in a singular environment. Unlike common robotic datasets used for training reward functions, human video-language datasets rarely contain trivial failure videos. To enhance the model's ability to distinguish between successful and failed robot executions, we cluster failure video features to enable the model to identify patterns within. For each cluster, we integrate a newly trained failure prompt into the text encoder to represent the corresponding failure mode. Our language-conditioned reward function shows outstanding generalization to new environments and new instructions for robot planning and reinforcement learning.
Authors: Rui Qin, Ming Sun, Chao Zhou, Bin Wang
Abstract: Recent Blind Image Super-Resolution (BSR) methods have shown proficiency in general images. However, we find that the efficacy of recent methods obviously diminishes when employed on image data with blur, while image data with intentional blur constitute a substantial proportion of general data. To further investigate and address this issue, we developed a new super-resolution dataset specifically tailored for blur images, named the Real-world Blur-kept Super-Resolution (ReBlurSR) dataset, which consists of nearly 3000 defocus and motion blur image samples with diverse blur sizes and varying blur intensities. Furthermore, we propose a new BSR framework for blur images called Perceptual-Blur-adaptive Super-Resolution (PBaSR), which comprises two main modules: the Cross Disentanglement Module (CDM) and the Cross Fusion Module (CFM). The CDM utilizes a dual-branch parallelism to isolate conflicting blur and general data during optimization. The CFM fuses the well-optimized prior from these distinct domains cost-effectively and efficiently based on model interpolation. By integrating these two modules, PBaSR achieves commendable performance on both general and blur data without any additional inference and deployment cost and is generalizable across multiple model architectures. Rich experiments show that PBaSR achieves state-of-the-art performance across various metrics without incurring extra inference costs. Within the widely adopted LPIPS metrics, PBaSR achieves an improvement range of approximately 0.02-0.10 with diverse anchor methods and blur types, across both the ReBlurSR and multiple common general BSR benchmarks. Code here: https://github.com/Imalne/PBaSR.
Authors: Yunlong Lin, Tian Ye, Sixiang Chen, Zhenqi Fu, Yingying Wang, Wenhao Chai, Zhaohu Xing, Lei Zhu, Xinghao Ding
Abstract: Existing low-light image enhancement (LIE) methods have achieved noteworthy success in solving synthetic distortions, yet they often fall short in practical applications. The limitations arise from two inherent challenges in real-world LIE: 1) the collection of distorted/clean image pairs is often impractical and sometimes even unavailable, and 2) accurately modeling complex degradations presents a non-trivial problem. To overcome them, we propose the Attribute Guidance Diffusion framework (AGLLDiff), a training-free method for effective real-world LIE. Instead of specifically defining the degradation process, AGLLDiff shifts the paradigm and models the desired attributes, such as image exposure, structure and color of normal-light images. These attributes are readily available and impose no assumptions about the degradation process, which guides the diffusion sampling process to a reliable high-quality solution space. Extensive experiments demonstrate that our approach outperforms the current leading unsupervised LIE methods across benchmarks in terms of distortion-based and perceptual-based metrics, and it performs well even in sophisticated wild degradation.
Authors: Zhongpai Gao, Abhishek Sharma, Meng Zheng, Benjamin Planche, Terrence Chen, Ziyan Wu
Abstract: Positioning patients for scanning and interventional procedures is a critical task that requires high precision and accuracy. The conventional workflow involves manually adjusting the patient support to align the center of the target body part with the laser projector or other guiding devices. This process is not only time-consuming but also prone to inaccuracies. In this work, we propose an automated patient positioning system that utilizes a camera to detect specific hand gestures from technicians, allowing users to indicate the target patient region to the system and initiate automated positioning. Our approach relies on a novel multi-stage pipeline to recognize and interpret the technicians' gestures, translating them into precise motions of medical devices. We evaluate our proposed pipeline during actual MRI scanning procedures, using RGB-Depth cameras to capture the process. Results show that our system achieves accurate and precise patient positioning with minimal technician intervention. Furthermore, we validate our method on HaGRID, a large-scale hand gesture dataset, demonstrating its effectiveness in hand detection and gesture recognition.
Authors: Rania Saoud, Slimane Larabi
Abstract: This paper presents a visual geo-localization system capable of determining the geographic locations of places (buildings and road intersections) from images without relying on GPS data. Our approach integrates three primary methods: Scale-Invariant Feature Transform (SIFT) for place recognition, traditional image processing for identifying road junction types, and deep learning using the VGG16 model for classifying road junctions. The most effective techniques have been integrated into an offline mobile application, enhancing accessibility for users requiring reliable location information in GPS-denied environments.
Authors: Zhengle Wang, Ruifeng Wang, Minjuan Wang, Tianyun Lai, Man Zhang
Abstract: Pest and disease classification is a challenging issue in agriculture. The performance of deep learning models is intricately linked to training data diversity and quantity, posing issues for plant pest and disease datasets that remain underdeveloped. This study addresses these challenges by constructing a comprehensive dataset and proposing an advanced network architecture that combines Contrastive Learning and Masked Image Modeling (MIM). The dataset comprises diverse plant species and pest categories, making it one of the largest and most varied in the field. The proposed network architecture demonstrates effectiveness in addressing plant pest and disease recognition tasks, achieving notable detection accuracy. This approach offers a viable solution for rapid, efficient, and cost-effective plant pest and disease detection, thereby reducing agricultural production costs. Our code and dataset will be publicly available to advance research in plant pest and disease recognition the GitHub repository at https://github.com/WASSER2545/GPID-22
Authors: Weiqin Jiao, Claudio Persello, George Vosselman
Abstract: Polygonal building outline extraction has been a research focus in recent years. Most existing methods have addressed this challenging task by decomposing it into several subtasks and employing carefully designed architectures. Despite their accuracy, such pipelines often introduce inefficiencies during training and inference. This paper presents an end-to-end framework, denoted as PolyR-CNN, which offers an efficient and fully integrated approach to predict vectorized building polygons and bounding boxes directly from remotely sensed images. Notably, PolyR-CNN leverages solely the features of the Region of Interest (RoI) for the prediction, thereby mitigating the necessity for complex designs. Furthermore, we propose a novel scheme with PolyR-CNN to extract detailed outline information from polygon vertex coordinates, termed vertex proposal feature, to guide the RoI features to predict more regular buildings. PolyR-CNN demonstrates the capacity to deal with buildings with holes through a simple post-processing method on the Inria dataset. Comprehensive experiments conducted on the CrowdAI dataset show that PolyR-CNN achieves competitive accuracy compared to state-of-the-art methods while significantly improving computational efficiency, i.e., achieving 79.2 Average Precision (AP), exhibiting a 15.9 AP gain and operating 2.5 times faster and four times lighter than the well-established end-to-end method PolyWorld. Replacing the backbone with a simple ResNet-50, PolyR-CNN maintains a 71.1 AP while running four times faster than PolyWorld.
Authors: Weiqin Jiao, Hao Cheng, Claudio Persello, George Vosselman
Abstract: Polygonal building outlines are crucial for geographic and cartographic applications. The existing approaches for outline extraction from aerial or satellite imagery are typically decomposed into subtasks, e.g., building masking and vectorization, or treat this task as a sequence-to-sequence prediction of ordered vertices. The former lacks efficiency, and the latter often generates redundant vertices, both resulting in suboptimal performance. To handle these issues, we propose a novel Region-of-Interest (RoI) query-based approach called RoIPoly. Specifically, we formulate each vertex as a query and constrain the query attention on the most relevant regions of a potential building, yielding reduced computational overhead and more efficient vertex level interaction. Moreover, we introduce a novel learnable logit embedding to facilitate vertex classification on the attention map; thus, no post-processing is needed for redundant vertex removal. We evaluated our method on the vectorized building outline extraction dataset CrowdAI and the 2D floorplan reconstruction dataset Structured3D. On the CrowdAI dataset, RoIPoly with a ResNet50 backbone outperforms existing methods with the same or better backbones on most MS-COCO metrics, especially on small buildings, and achieves competitive results in polygon quality and vertex redundancy without any post-processing. On the Structured3D dataset, our method achieves the second-best performance on most metrics among existing methods dedicated to 2D floorplan reconstruction, demonstrating our cross-domain generalization capability. The code will be released upon acceptance of this paper.
Authors: Xiaomeng Chu, Jiajun Deng, Guoliang You, Yifan Duan, Yao Li, Yanyong Zhang
Abstract: The recent advances in query-based multi-camera 3D object detection are featured by initializing object queries in the 3D space, and then sampling features from perspective-view images to perform multi-round query refinement. In such a framework, query points near the same camera ray are likely to sample similar features from very close pixels, resulting in ambiguous query features and degraded detection accuracy. To this end, we introduce RayFormer, a camera-ray-inspired query-based 3D object detector that aligns the initialization and feature extraction of object queries with the optical characteristics of cameras. Specifically, RayFormer transforms perspective-view image features into bird's eye view (BEV) via the lift-splat-shoot method and segments the BEV map to sectors based on the camera rays. Object queries are uniformly and sparsely initialized along each camera ray, facilitating the projection of different queries onto different areas in the image to extract distinct features. Besides, we leverage the instance information of images to supplement the uniformly initialized object queries by further involving additional queries along the ray from 2D object detection boxes. To extract unique object-level features that cater to distinct queries, we design a ray sampling method that suitably organizes the distribution of feature sampling points on both images and bird's eye view. Extensive experiments are conducted on the nuScenes dataset to validate our proposed ray-inspired model design. The proposed RayFormer achieves 55.5% mAP and 63.3% NDS, respectively. Our codes will be made available.
Authors: Georgia Argyrou, Angeliki Dimitriou, Maria Lymperaiou, Giorgos Filandrianos, Giorgos Stamou
Abstract: The advent of artificial intelligence has contributed in a groundbreaking transformation of the fashion industry, redefining creativity and innovation in unprecedented ways. This work investigates methodologies for generating tailored fashion descriptions using two distinct Large Language Models and a Stable Diffusion model for fashion image creation. Emphasizing adaptability in AI-driven fashion creativity, we depart from traditional approaches and focus on prompting techniques, such as zero-shot and few-shot learning, as well as Chain-of-Thought (CoT), which results in a variety of colors and textures, enhancing the diversity of the outputs. Central to our methodology is Retrieval-Augmented Generation (RAG), enriching models with insights from fashion sources to ensure contemporary representations. Evaluation combines quantitative metrics such as CLIPscore with qualitative human judgment, highlighting strengths in creativity, coherence, and aesthetic appeal across diverse styles. Among the participants, RAG and few-shot learning techniques are preferred for their ability to produce more relevant and appealing fashion descriptions. Our code is provided at https://github.com/georgiarg/AutoFashion.
Authors: Sanjeev Muralikrishnan, Niladri Shekhar Dutt, Siddhartha Chaudhuri, Noam Aigerman, Vladimir Kim, Matthew Fisher, Niloy J. Mitra
Abstract: We introduce Temporal Residual Jacobians as a novel representation to enable data-driven motion transfer. Our approach does not assume access to any rigging or intermediate shape keyframes, produces geometrically and temporally consistent motions, and can be used to transfer long motion sequences. Central to our approach are two coupled neural networks that individually predict local geometric and temporal changes that are subsequently integrated, spatially and temporally, to produce the final animated meshes. The two networks are jointly trained, complement each other in producing spatial and temporal signals, and are supervised directly with 3D positional information. During inference, in the absence of keyframes, our method essentially solves a motion extrapolation problem. We test our setup on diverse meshes (synthetic and scanned shapes) to demonstrate its superiority in generating realistic and natural-looking animations on unseen body shapes against SoTA alternatives. Supplemental video and code are available at https://temporaljacobians.github.io/ .
Authors: Md Laraib Salam, Akash S Balsaraf, Gaurav Gupta
Abstract: The use of neural networks and deep learning techniques in image processing has significantly advanced the field, enabling highly accurate recognition results. However, achieving high recognition rates often necessitates complex network models, which can be challenging to train and require substantial computational resources. This research presents a simplified yet effective approach to predicting both the base and exponent from images of mathematical expressions using a multi-output Convolutional Neural Network (CNN). The model is trained on 10,900 synthetically generated images containing exponent expressions, incorporating random noise, font size variations, and blur intensity to simulate real-world conditions. The proposed CNN model demonstrates robust performance with efficient training time. The experimental results indicate that the model achieves high accuracy in predicting the base and exponent values, proving the efficacy of this approach in handling noisy and varied input images.
Authors: Md Zarif Hossain, Ahmed Imteaj
Abstract: Vision-language models (VLMs) have achieved significant strides in recent times specially in multimodal tasks, yet they remain susceptible to adversarial attacks on their vision components. To address this, we propose Sim-CLIP, an unsupervised adversarial fine-tuning method that enhances the robustness of the widely-used CLIP vision encoder against such attacks while maintaining semantic richness and specificity. By employing a Siamese architecture with cosine similarity loss, Sim-CLIP learns semantically meaningful and attack-resilient visual representations without requiring large batch sizes or momentum encoders. Our results demonstrate that VLMs enhanced with Sim-CLIP's fine-tuned CLIP encoder exhibit significantly enhanced robustness against adversarial attacks, while preserving semantic meaning of the perturbed images. Notably, Sim-CLIP does not require additional training or fine-tuning of the VLM itself; replacing the original vision encoder with our fine-tuned Sim-CLIP suffices to provide robustness. This work underscores the significance of reinforcing foundational models like CLIP to safeguard the reliability of downstream VLM applications, paving the way for more secure and effective multimodal systems.
Authors: Mohammad Saeed Ebrahimi Saadabadi, Sahar Rahimi Malakshan, Ali Dabouei, Nasser M. Nasrabadi
Abstract: Aiming to enhance Face Recognition (FR) on Low-Quality (LQ) inputs, recent studies suggest incorporating synthetic LQ samples into training. Although promising, the quality factors that are considered in these works are general rather than FR-specific, \eg, atmospheric turbulence, resolution, \etc. Motivated by the observation of the vulnerability of current FR models to even small Face Alignment Errors (FAE) in LQ images, we present a simple yet effective method that considers FAE as another quality factor that is tailored to FR. We seek to improve LQ FR by enhancing FR models' robustness to FAE. To this aim, we formalize the problem as a combination of differentiable spatial transformations and adversarial data augmentation in FR. We perturb the alignment of the training samples using a controllable spatial transformation and enrich the training with samples expressing FAE. We demonstrate the benefits of the proposed method by conducting evaluations on IJB-B, IJB-C, IJB-S (+4.3\% Rank1), and TinyFace (+2.63\%). \href{https://github.com/msed-Ebrahimi/ARoFace}{https://github.com/msed-Ebrahimi/ARoFace}
URLs: https://github.com/msed-Ebrahimi/ARoFace, https://github.com/msed-Ebrahimi/ARoFace
Authors: Jae Joong Lee, Bedrich Benes
Abstract: We introduce RGB2Point, an unposed single-view RGB image to a 3D point cloud generation based on Transformer. RGB2Point takes an input image of an object and generates a dense 3D point cloud. Contrary to prior works based on CNN layers and diffusion denoising approaches, we use pre-trained Transformer layers that are fast and generate high-quality point clouds with consistent quality over available categories. Our generated point clouds demonstrate high quality on a real-world dataset, as evidenced by improved Chamfer distance (51.15%) and Earth Mover's distance (45.96%) metrics compared to the current state-of-the-art. Additionally, our approach shows a better quality on a synthetic dataset, achieving better Chamfer distance (39.26%), Earth Mover's distance (26.95%), and F-score (47.16%). Moreover, our method produces 63.1% more consistent high-quality results across various object categories compared to prior works. Furthermore, RGB2Point is computationally efficient, requiring only 2.3GB of VRAM to reconstruct a 3D point cloud from a single RGB image, and our implementation generates the results 15,133x faster than a SOTA diffusion-based model.
Authors: Jingzhi Gong, Sisi Li, Giordano d'Aloisio, Zishuo Ding, Yulong Ye, William B. Langdon, Federica Sarro
Abstract: Tuning the parameters and prompts for improving AI-based text-to-image generation has remained a substantial yet unaddressed challenge. Hence we introduce GreenStableYolo, which improves the parameters and prompts for Stable Diffusion to both reduce GPU inference time and increase image generation quality using NSGA-II and Yolo. Our experiments show that despite a relatively slight trade-off (18%) in image quality compared to StableYolo (which only considers image quality), GreenStableYolo achieves a substantial reduction in inference time (266% less) and a 526% higher hypervolume, thereby advancing the state-of-the-art for text-to-image generation.
Authors: Ghazi Gharsallah, Georges Kaddoum
Abstract: As wireless communication technology progresses towards the sixth generation (6G), high-frequency millimeter-wave (mmWave) communication has emerged as a promising candidate for enabling vehicular networks. It offers high data rates and low-latency communication. However, obstacles such as buildings, trees, and other vehicles can cause signal attenuation and blockage, leading to communication failures that can result in fatal accidents or traffic congestion. Predicting blockages is crucial for ensuring reliable and efficient communications. Furthermore, the advent of 6G technology is anticipated to integrate advanced sensing capabilities, utilizing a variety of sensor types. These sensors, ranging from traditional RF sensors to cameras and Lidar sensors, are expected to provide access to rich multimodal data, thereby enriching communication systems with a wealth of additional contextual information. Leveraging this multimodal data becomes essential for making precise network management decisions, including the crucial task of blockage detection. In this paper, we propose a Deep Learning (DL)-based approach that combines Convolutional Neural Networks (CNNs) and customized Vision Transformers (ViTs) to effectively extract essential information from multimodal data and predict blockages in vehicular networks. Our method capitalizes on the synergistic strengths of CNNs and ViTs to extract features from time-series multimodal data, which include images and beam vectors. To capture temporal dependencies between the extracted features and the blockage state at future time steps, we employ a Gated Recurrent Unit (GRU)-based architecture. Our results show that the proposed approach achieves high accuracy and outperforms state-of-the-art solutions, achieving more than $95\%$ accurate predictions.
Authors: Bolivar Solarte, Chin-Hsuan Wu, Jin-Cheng Jhang, Jonathan Lee, Yi-Hsuan Tsai, Min Sun
Abstract: In this paper, we introduce a novel geometry-aware self-training framework for room layout estimation models on unseen scenes with unlabeled data. Our approach utilizes a ray-casting formulation to aggregate multiple estimates from different viewing positions, enabling the computation of reliable pseudo-labels for self-training. In particular, our ray-casting approach enforces multi-view consistency along all ray directions and prioritizes spatial proximity to the camera view for geometry reasoning. As a result, our geometry-aware pseudo-labels effectively handle complex room geometries and occluded walls without relying on assumptions such as Manhattan World or planar room walls. Evaluation on publicly available datasets, including synthetic and real-world scenarios, demonstrates significant improvements in current state-of-the-art layout models without using any human annotation.
Authors: Navyansh Mahla, Annie D'souza, Shubh Gupta, Bhavik Kanekar, Kshitij Sharad Jadhav
Abstract: The application of large-scale models in medical image segmentation demands substantial quantities of meticulously annotated data curated by experts along with high computational resources, both of which are challenges in resource-poor settings. In this study, we present the Medical Segment Anything Model with Galore MedSAGa where we adopt the Segment Anything Model (SAM) to achieve memory-efficient, few-shot medical image segmentation by applying Gradient Low-Rank Projection GaLore to the parameters of the image encoder of SAM. Meanwhile, the weights of the prompt encoder and mask decoder undergo full parameter fine-tuning using standard optimizers. We further assess MedSAGa's few-shot learning capabilities, reporting on its memory efficiency and segmentation performance across multiple standard medical image segmentation datasets. We compare it with several baseline models, including LoRA fine-tuned SAM (SAMed) and DAE-Former. Experiments across multiple datasets and these baseline models with different number of images for fine tuning demonstrated that the GPU memory consumption of MedSAGa is significantly less than that of the baseline models, achieving an average memory efficiency of 66% more than current state-of-the-art (SOTA) models for medical image segmentation. The combination of substantially lower memory requirements and comparable to SOTA results in few-shot learning for medical image segmentation positions MedSAGa as an optimal solution for deployment in resource-constrained settings.
Authors: Shivprasad Sagare, Hemachandran S, Kinshuk Sarabhai, Prashant Ullegaddi, Rajeshkumar SA
Abstract: Recent advances in multimodal LLMs, have led to several video-text models being proposed for critical video-related tasks. However, most of the previous works support visual input only, essentially muting the audio signal in the video. Few models that support both audio and visual input, are not explicitly trained on audio data. Hence, the effect of audio towards video understanding is largely unexplored. To this end, we propose a model architecture that handles audio-visual inputs explicitly. We train our model with both audio and visual data from a video instruction-tuning dataset. Comparison with vision-only baselines, and other audio-visual models showcase that training on audio data indeed leads to improved grounding of responses. For better evaluation of audio-visual models, we also release a human-annotated benchmark dataset, with audio-aware question-answer pairs.
Authors: Jianxin Liang, Xiaojun Meng, Yueqian Wang, Chang Liu, Qun Liu, Dongyan Zhao
Abstract: Video Question Answering (VideoQA) has emerged as a challenging frontier in the field of multimedia processing, requiring intricate interactions between visual and textual modalities. Simply uniformly sampling frames or indiscriminately aggregating frame-level visual features often falls short in capturing the nuanced and relevant contexts of videos to well perform VideoQA. To mitigate these issues, we propose VidF4, a novel VideoQA framework equipped with tailored frame selection strategy for effective and efficient VideoQA. We propose three frame-scoring mechanisms that consider both question relevance and inter-frame similarity to evaluate the importance of each frame for a given question on the video. Furthermore, we design a differentiable adaptive frame sampling mechanism to facilitate end-to-end training for the frame selector and answer generator. The experimental results across three widely adopted benchmarks demonstrate that our model consistently outperforms existing VideoQA methods, establishing a new SOTA across NExT-QA (+0.3%), STAR (+0.9%), and TVQA (+1.0%). Furthermore, through both quantitative and qualitative analyses, we validate the effectiveness of each design choice.
Authors: Yiyang Jiang, Wengyu Zhang, Xulu Zhang, Xiaoyong Wei, Chang Wen Chen, Qing Li
Abstract: In this paper, we investigate the feasibility of leveraging large language models (LLMs) for integrating general knowledge and incorporating pseudo-events as priors for temporal content distribution in video moment retrieval (VMR) models. The motivation behind this study arises from the limitations of using LLMs as decoders for generating discrete textual descriptions, which hinders their direct application to continuous outputs like salience scores and inter-frame embeddings that capture inter-frame relations. To overcome these limitations, we propose utilizing LLM encoders instead of decoders. Through a feasibility study, we demonstrate that LLM encoders effectively refine inter-concept relations in multimodal embeddings, even without being trained on textual embeddings. We also show that the refinement capability of LLM encoders can be transferred to other embeddings, such as BLIP and T5, as long as these embeddings exhibit similar inter-concept similarity patterns to CLIP embeddings. We present a general framework for integrating LLM encoders into existing VMR architectures, specifically within the fusion module. Through experimental validation, we demonstrate the effectiveness of our proposed methods by achieving state-of-the-art performance in VMR. The source code can be accessed at https://github.com/fletcherjiang/LLMEPET.
Authors: Bowen Zhang, Cheng Yang, Xuanhui Liu
Abstract: In recent years, advancements in AIGC (Artificial Intelligence Generated Content) technology have significantly enhanced the capabilities of large text-to-image models. Despite these improvements, controllable image generation remains a challenge. Current methods, such as training, forward guidance, and backward guidance, have notable limitations. The first two approaches either demand substantial computational resources or produce subpar results. The third approach depends on phenomena specific to certain model architectures, complicating its application to large-scale image generation.To address these issues, we propose a novel controllable generation framework that offers a generalized interpretation of backward guidance without relying on specific assumptions. Leveraging this framework, we introduce LSReGen, a large-scale layout-to-image method designed to generate high-quality, layout-compliant images. Experimental results show that LSReGen outperforms existing methods in the large-scale layout-to-image task, underscoring the effectiveness of our proposed framework. Our code and models will be open-sourced.
Authors: Yashashwee Chakrabarty, Smruti Ranjan Sarangi
Abstract: Autonomous mobile robots like self-flying drones and industrial robots heavily depend on depth images to perform tasks such as 3D reconstruction and visual SLAM. However, the presence of inaccuracies in these depth images can greatly hinder the effectiveness of these applications, resulting in sub-optimal results. Depth images produced by commercially available cameras frequently exhibit noise, which manifests as flickering pixels and erroneous patches. ML-based methods to rectify these images are unsuitable for edge devices that have very limited computational resources. Non-ML methods are much faster but have limited accuracy, especially for correcting errors that are a result of occlusion and camera movement. We propose a scheme called VoxDepth that is fast, accurate, and runs very well on edge devices. It relies on a host of novel techniques: 3D point cloud construction and fusion, and using it to create a template that can fix erroneous depth images. VoxDepth shows superior results on both synthetic and real-world datasets. We demonstrate a 31% improvement in quality as compared to state-of-the-art methods on real-world depth datasets, while maintaining a competitive framerate of 27 FPS (frames per second).
Authors: Yuelang Xu, Lizhen Wang, Zerong Zheng, Zhaoqi Su, Yebin Liu
Abstract: Creating high-fidelity 3D human head avatars is crucial for applications in VR/AR, telepresence, digital human interfaces, and film production. Recent advances have leveraged morphable face models to generate animated head avatars from easily accessible data, representing varying identities and expressions within a low-dimensional parametric space. However, existing methods often struggle with modeling complex appearance details, e.g., hairstyles and accessories, and suffer from low rendering quality and efficiency. This paper introduces a novel approach, 3D Gaussian Parametric Head Model, which employs 3D Gaussians to accurately represent the complexities of the human head, allowing precise control over both identity and expression. Additionally, it enables seamless face portrait interpolation and the reconstruction of detailed head avatars from a single image. Unlike previous methods, the Gaussian model can handle intricate details, enabling realistic representations of varying appearances and complex expressions. Furthermore, this paper presents a well-designed training framework to ensure smooth convergence, providing a guarantee for learning the rich content. Our method achieves high-quality, photo-realistic rendering with real-time efficiency, making it a valuable contribution to the field of parametric head models.
Authors: Jiajun Hu, Jian Zhang, Lei Qi, Yinghuan Shi, Yang Gao
Abstract: Domain generalization (DG) aims to avoid the performance degradation of the model when the distribution shift between the limited training data and unseen test data occurs. Recently, foundation models with enormous parameters have been pre-trained with huge datasets, demonstrating strong generalization ability and showing promising direction for solving the DG problem. However, fully Fine-Tuning (FT) the foundation models results in unsatisfactory out-of-distribution accuracy due to the destroyed pre-trained generalized features. Recently, Parameter-Efficient Fine-Tuning (PEFT) alleviates the above problem by fine-tuning a small portion of the model parameters while keeping the rest frozen, which achieves better generalization performance compared to FT. Nevertheless, PEFT still suffers from the issue of overfitting to the training domains. To address the above issue, we propose Parameter-Efficient Group with Orthogonal regularization (PEGO) for vision transformers, which effectively preserves the generalization ability of the pre-trained network and learns more diverse knowledge compared with conventional PEFT. Specifically, we inject a group of trainable Low-Rank Adaptation (LoRA) modules into the pre-trained model and propose an orthogonal regularization loss to enhance the generalization ability of the model. Our framework achieves SOTA performance on five DG benchmarks, while only requiring training a small number of parameters without adding additional testing cost.
Authors: Sheng Fan, Rui Liu, Wenguan Wang, Yi Yang
Abstract: Navigation instruction generation, which requires embodied agents to describe the navigation routes, has been of great interest in robotics and human-computer interaction. Existing studies directly map the sequence of 2D perspective observations to route descriptions. Though straightforward, they overlook the geometric information and object semantics of the 3D environment. To address these challenges, we propose BEVInstructor, which incorporates Bird's Eye View (BEV) features into Multi-Modal Large Language Models (MLLMs) for instruction generation. Specifically, BEVInstructor constructs a PerspectiveBEVVisual Encoder for the comprehension of 3D environments through fusing BEV and perspective features. To leverage the powerful language capabilities of MLLMs, the fused representations are used as visual prompts for MLLMs, and perspective-BEV prompt tuning is proposed for parameter-efficient updating. Based on the perspective-BEV prompts, BEVInstructor further adopts an instance-guided iterative refinement pipeline, which improves the instructions in a progressive manner. BEVInstructor achieves impressive performance across diverse datasets (i.e., R2R, REVERIE, and UrbanWalk).
Authors: Zhaotong Yang, Zicheng Jiang, Xinzhe Li, Huiyu Zhou, Junyu Dong, Huaidong Zhang, Yong Du
Abstract: In this paper, we introduce D$^4$-VTON, an innovative solution for image-based virtual try-on. We address challenges from previous studies, such as semantic inconsistencies before and after garment warping, and reliance on static, annotation-driven clothing parsers. Additionally, we tackle the complexities in diffusion-based VTON models when handling simultaneous tasks like inpainting and denoising. Our approach utilizes two key technologies: Firstly, Dynamic Semantics Disentangling Modules (DSDMs) extract abstract semantic information from garments to create distinct local flows, improving precise garment warping in a self-discovered manner. Secondly, by integrating a Differential Information Tracking Path (DITP), we establish a novel diffusion-based VTON paradigm. This path captures differential information between incomplete try-on inputs and their complete versions, enabling the network to handle multiple degradations independently, thereby minimizing learning ambiguities and achieving realistic results with minimal overhead. Extensive experiments demonstrate that D$^4$-VTON significantly outperforms existing methods in both quantitative metrics and qualitative evaluations, demonstrating its capability in generating realistic images and ensuring semantic consistency.
Authors: Duo Su, Junjie Hou, Weizhi Gao, Yingjie Tian, Bowen Tang
Abstract: Dataset distillation offers a lightweight synthetic dataset for fast network training with promising test accuracy. To imitate the performance of the original dataset, most approaches employ bi-level optimization and the distillation space relies on the matching architecture. Nevertheless, these approaches either suffer significant computational costs on large-scale datasets or experience performance decline on cross-architectures. We advocate for designing an economical dataset distillation framework that is independent of the matching architectures. With empirical observations, we argue that constraining the consistency of the real and synthetic image spaces will enhance the cross-architecture generalization. Motivated by this, we introduce Dataset Distillation via Disentangled Diffusion Model (D$^4$M), an efficient framework for dataset distillation. Compared to architecture-dependent methods, D$^4$M employs latent diffusion model to guarantee consistency and incorporates label information into category prototypes. The distilled datasets are versatile, eliminating the need for repeated generation of distinct datasets for various architectures. Through comprehensive experiments, D$^4$M demonstrates superior performance and robust generalization, surpassing the SOTA methods across most aspects.
Authors: Yechan Kim, JongHyun Park, SooYeon Kim, Moongu Jeon
Abstract: Recently, numerous methods have achieved impressive performance in remote sensing object detection, relying on convolution or transformer architectures. Such detectors typically have a feature backbone to extract useful features from raw input images. For the remote sensing domain, a common practice among current detectors is to initialize the backbone with pre-training on ImageNet consisting of natural scenes. Fine-tuning the backbone is typically required to generate features suitable for remote-sensing images. However, this could hinder the extraction of basic visual features in long-term training, thus restricting performance improvement. To mitigate this issue, we propose a novel method named DBF (Dynamic Backbone Freezing) for feature backbone fine-tuning on remote sensing object detection. Our method aims to handle the dilemma of whether the backbone should extract low-level generic features or possess specific knowledge of the remote sensing domain, by introducing a module called 'Freezing Scheduler' to dynamically manage the update of backbone features during training. Extensive experiments on DOTA and DIOR-R show that our approach enables more accurate model learning while substantially reducing computational costs. Our method can be seamlessly adopted without additional effort due to its straightforward design.
Authors: Idan Kligvasser, Regev Cohen, George Leifman, Ehud Rivlin, Michael Elad
Abstract: Video generation has drawn significant interest recently, pushing the development of large-scale models capable of producing realistic videos with coherent motion. Due to memory constraints, these models typically generate short video segments that are then combined into long videos. The merging process poses a significant challenge, as it requires ensuring smooth transitions and overall consistency. In this paper, we introduce Anchored Diffusion, a novel method for synthesizing relatively long and seamless videos. We extend Diffusion Transformers (DiTs) to incorporate temporal information, creating our sequence-DiT (sDiT) model for generating short video segments. Unlike previous works, we train our model on video sequences with random non-uniform temporal spacing and incorporate temporal information via external guidance, increasing flexibility and allowing it to capture both short and long-term relationships. Furthermore, during inference, we leverage the transformer architecture to modify the diffusion process, generating a batch of non-uniform sequences anchored to a common frame, ensuring consistency regardless of temporal distance. To demonstrate our method, we focus on face reenactment, the task of creating a video from a source image that replicates the facial expressions and movements from a driving video. Through comprehensive experiments, we show our approach outperforms current techniques in producing longer consistent high-quality videos while offering editing capabilities.
Authors: Yunyi Xuan, Weijie Chen, Shicai Yang, Di Xie, Luojun Lin, Yueting Zhuang
Abstract: Data-Free Knowledge Distillation (DFKD) has shown great potential in creating a compact student model while alleviating the dependency on real training data by synthesizing surrogate data. However, prior arts are seldom discussed under distribution shifts, which may be vulnerable in real-world applications. Recent Vision-Language Foundation Models, e.g., CLIP, have demonstrated remarkable performance in zero-shot out-of-distribution generalization, yet consuming heavy computation resources. In this paper, we discuss the extension of DFKD to Vision-Language Foundation Models without access to the billion-level image-text datasets. The objective is to customize a student model for distribution-agnostic downstream tasks with given category concepts, inheriting the out-of-distribution generalization capability from the pre-trained foundation models. In order to avoid generalization degradation, the primary challenge of this task lies in synthesizing diverse surrogate images driven by text prompts. Since not only category concepts but also style information are encoded in text prompts, we propose three novel Prompt Diversification methods to encourage image synthesis with diverse styles, namely Mix-Prompt, Random-Prompt, and Contrastive-Prompt. Experiments on out-of-distribution generalization datasets demonstrate the effectiveness of the proposed methods, with Contrastive-Prompt performing the best.
Authors: Fuying Wang, Shenghui Du, Lequan Yu
Abstract: Radiology reports provide detailed descriptions of medical imaging integrated with patients' medical histories, while report writing is traditionally labor-intensive, increasing radiologists' workload and the risk of diagnostic errors. Recent efforts in automating this process seek to mitigate these issues by enhancing accuracy and clinical efficiency. Emerging research in automating this process promises to alleviate these challenges by reducing errors and streamlining clinical workflows. However, existing automated approaches are based on a single timestamp and often neglect the critical temporal aspect of patients' imaging histories, which is essential for accurate longitudinal analysis. To address this gap, we propose a novel History Enhanced Radiology Report Generation (HERGen) framework that employs a employs a group causal transformer to efficiently integrate longitudinal data across patient visits. Our approach not only allows for comprehensive analysis of varied historical data but also improves the quality of generated reports through an auxiliary contrastive objective that aligns image sequences with their corresponding reports. More importantly, we introduce a curriculum learning-based strategy to adeptly handle the inherent complexity of longitudinal radiology data and thus stabilize the optimization of our framework. The extensive evaluations across three datasets demonstrate that our framework surpasses existing methods in generating accurate radiology reports and effectively predicting disease progression from medical images.
Authors: Junwen Luo, Chengyong Jiang, Qingyuan Chen, Dongqi Han, Yansen Wang, Biao Yan, Dongsheng Li, Jiayi Zhang
Abstract: Effective visual brain-machine interfaces (BMI) is based on reliable and stable EEG biomarkers. However, traditional adaptive filter-based approaches may suffer from individual variations in EEG signals, while deep neural network-based approaches may be hindered by the non-stationarity of EEG signals caused by biomarker attenuation and background oscillations. To address these challenges, we propose the Visual Evoked Potential Booster (VEP Booster), a novel closed-loop AI framework that generates reliable and stable EEG biomarkers under visual stimulation protocols. Our system leverages an image generator to refine stimulus images based on real-time feedback from human EEG signals, generating visual stimuli tailored to the preferences of primary visual cortex (V1) neurons and enabling effective targeting of neurons most responsive to stimuli. We validated our approach by implementing a system and employing steady-state visual evoked potential (SSVEP) visual protocols in five human subjects. Our results show significant enhancements in the reliability and utility of EEG biomarkers for all individuals, with the largest improvement in SSVEP response being 105%, the smallest being 28%, and the average increase being 76.5%. These promising results have implications for both clinical and technological applications
Authors: Zhu Huang, Gang Pan, Chao Kang, YaoZhi Lv
Abstract: In sewer pipe Closed-Circuit Television (CCTV) inspection, accurate temporal defect localization is essential for effective defect classification, detection, segmentation and quantification. Industry standards typically do not require time-interval annotations, even though they are more informative than time-point annotations for defect localization, resulting in additional annotation costs when fully supervised methods are used. Additionally, differences in scene types and camera motion patterns between pipe inspections and Temporal Action Localization (TAL) hinder the effective transfer of point-supervised TAL methods. Therefore, this study introduces a Semi-supervised multi-Prototype-based method incorporating visual Odometry for enhanced attention guidance (PipeSPO). PipeSPO fully leverages unlabeled data through unsupervised pretext tasks and utilizes time-point annotated data with a weakly supervised multi-prototype-based method, relying on visual odometry features to capture camera pose information. Experiments on real-world datasets demonstrate that PipeSPO achieves 41.89% average precision across Intersection over Union (IoU) thresholds of 0.1-0.7, improving by 8.14% over current state-of-the-art methods.
Authors: Jingyi Xu, Hieu Le, Dimitris Samaras
Abstract: Advances in generative models increase the need for sample quality assessment. To do so, previous methods rely on a pre-trained feature extractor to embed the generated samples and real samples into a common space for comparison. However, different feature extractors might lead to inconsistent assessment outcomes. Moreover, these methods are not applicable for domains where a robust, universal feature extractor does not yet exist, such as medical images or 3D assets. In this paper, we propose to directly examine the latent space of the trained generative model to infer generated sample quality. This is feasible because the quality a generated sample directly relates to the amount of training data resembling it, and we can infer this information by examining the density of the latent space. Accordingly, we use a latent density score function to quantify sample quality. We show that the proposed score correlates highly with the sample quality for various generative models including VAEs, GANs and Latent Diffusion Models. Compared with previous quality assessment methods, our method has the following advantages: 1) pre-generation quality estimation with reduced computational cost, 2) generalizability to various domains and modalities, and 3) applicability to latent-based image editing and generation methods. Extensive experiments demonstrate that our proposed methods can benefit downstream tasks such as few-shot image classification and latent face image editing. Code is available at https://github.com/cvlab-stonybrook/LS-sample-quality.
URLs: https://github.com/cvlab-stonybrook/LS-sample-quality.
Authors: Ruoyu Feng, Tao Yu, Xin Jin, Xiaoyuan Yu, Lei Xiao, Zhibo Chen
Abstract: In recent studies on domain adaptation, significant emphasis has been placed on the advancement of learning shared knowledge from a source domain to a target domain. Recently, the large vision-language pre-trained model, i.e., CLIP has shown strong ability on zero-shot recognition, and parameter efficient tuning can further improve its performance on specific tasks. This work demonstrates that a simple domain prior boosts CLIP's zero-shot recognition in a specific domain. Besides, CLIP's adaptation relies less on source domain data due to its diverse pre-training dataset. Furthermore, we create a benchmark for zero-shot adaptation and pseudo-labeling based self-training with CLIP. Last but not least, we propose to improve the task generalization ability of CLIP from multiple unlabeled domains, which is a more practical and unique scenario. We believe our findings motivate a rethinking of domain adaptation benchmarks and the associated role of related algorithms in the era of CLIP.
Authors: Haiyang Zhou, Xinhua Cheng, Wangbo Yu, Yonghong Tian, Li Yuan
Abstract: 3D scene generation is in high demand across various domains, including virtual reality, gaming, and the film industry. Owing to the powerful generative capabilities of text-to-image diffusion models that provide reliable priors, the creation of 3D scenes using only text prompts has become viable, thereby significantly advancing researches in text-driven 3D scene generation. In order to obtain multiple-view supervision from 2D diffusion models, prevailing methods typically employ the diffusion model to generate an initial local image, followed by iteratively outpainting the local image using diffusion models to gradually generate scenes. Nevertheless, these outpainting-based approaches prone to produce global inconsistent scene generation results without high degree of completeness, restricting their broader applications. To tackle these problems, we introduce HoloDreamer, a framework that first generates high-definition panorama as a holistic initialization of the full 3D scene, then leverage 3D Gaussian Splatting (3D-GS) to quickly reconstruct the 3D scene, thereby facilitating the creation of view-consistent and fully enclosed 3D scenes. Specifically, we propose Stylized Equirectangular Panorama Generation, a pipeline that combines multiple diffusion models to enable stylized and detailed equirectangular panorama generation from complex text prompts. Subsequently, Enhanced Two-Stage Panorama Reconstruction is introduced, conducting a two-stage optimization of 3D-GS to inpaint the missing region and enhance the integrity of the scene. Comprehensive experiments demonstrated that our method outperforms prior works in terms of overall visual consistency and harmony as well as reconstruction quality and rendering robustness when generating fully enclosed scenes.
Authors: Jingwei Guo, Meihui Wang, Ilya Ilyankou, Natchapon Jongwiriyanurak, Xiaowei Gao, Nicola Christie, James Haworth
Abstract: Panoramic cycling videos can record 360{\deg} views around the cyclists. Thus, it is essential to conduct automatic road user analysis on them using computer vision models to provide data for studies on cycling safety. However, the features of panoramic data such as severe distortions, large number of small objects and boundary continuity have brought great challenges to the existing CV models, including poor performance and evaluation methods that are no longer applicable. In addition, due to the lack of data with annotations, it is not easy to re-train the models. In response to these problems, the project proposed and implemented a three-step methodology: (1) improve the prediction performance of the pre-trained object detection models on panoramic data by projecting the original image into 4 perspective sub-images; (2) introduce supports for boundary continuity and category information into DeepSORT, a commonly used multiple object tracking model, and set an improved detection model as its detector; (3) using the tracking results, develop an application for detecting the overtaking behaviour of the surrounding vehicles. Evaluated on the panoramic cycling dataset built by the project, the proposed methodology improves the average precision of YOLO v5m6 and Faster RCNN-FPN under any input resolution setting. In addition, it raises MOTA and IDF1 of DeepSORT by 7.6\% and 9.7\% respectively. When detecting the overtakes in the test videos, it achieves the F-score of 0.88. The code is available on GitHub at github.com/cuppp1998/360_object_tracking to ensure the reproducibility and further improvements of results.
Authors: Kaziwa Saleh, S\'andor Sz\'en\'asi, Zolt\'an V\'amossy
Abstract: We present a model to reconstruct partially visible objects. The model takes a mask as an input, which we call weighted mask. The mask is utilized by gated convolutions to assign more weight to the visible pixels of the occluded instance compared to the background, while ignoring the features of the invisible pixels. By drawing more attention from the visible region, our model can predict the invisible patch more effectively than the baseline models, especially in instances with uniform texture. The model is trained on COCOA dataset and two subsets of it in a self-supervised manner. The results demonstrate that our model generates higher quality and more texture-rich outputs compared to baseline models. Code is available at: https://github.com/KaziwaSaleh/mask-guided.
Authors: Yiqun Zhao, Chenming Wu, Binbin Huang, Yihao Zhi, Chen Zhao, Jingdong Wang, Shenghua Gao
Abstract: Efficient and accurate reconstruction of a relightable, dynamic clothed human avatar from a monocular video is crucial for the entertainment industry. This paper introduces the Surfel-based Gaussian Inverse Avatar (SGIA) method, which introduces efficient training and rendering for relightable dynamic human reconstruction. SGIA advances previous Gaussian Avatar methods by comprehensively modeling Physically-Based Rendering (PBR) properties for clothed human avatars, allowing for the manipulation of avatars into novel poses under diverse lighting conditions. Specifically, our approach integrates pre-integration and image-based lighting for fast light calculations that surpass the performance of existing implicit-based techniques. To address challenges related to material lighting disentanglement and accurate geometry reconstruction, we propose an innovative occlusion approximation strategy and a progressive training approach. Extensive experiments demonstrate that SGIA not only achieves highly accurate physical properties but also significantly enhances the realistic relighting of dynamic human avatars, providing a substantial speed advantage. We exhibit more results in our project page: \href{https://GS-IA.github.io}{https://GS-IA.github.io}.
Authors: Yancheng Wang, Yingzhen Yang
Abstract: Self-attention and transformers have been widely used in deep learning. Recent efforts have been devoted to incorporating transformer blocks into different neural architectures, including those with convolutions, leading to various visual transformers for computer vision tasks. In this paper, we propose a novel and compact transformer block, Transformer with Learnable Token Merging (LTM), or LTM-Transformer. LTM-Transformer performs token merging in a learnable scheme. LTM-Transformer is compatible with many popular and compact transformer networks, and it reduces the FLOPs and the inference time of the visual transformers while maintaining or even improving the prediction accuracy. In the experiments, we replace all the transformer blocks in popular visual transformers, including MobileViT, EfficientViT, ViT-S/16, and Swin-T, with LTM-Transformer blocks, leading to LTM-Transformer networks with different backbones. The LTM-Transformer is motivated by reduction of Information Bottleneck, and a novel and separable variational upper bound for the IB loss is derived. The architecture of mask module in our LTM blocks which generate the token merging mask is designed to reduce the derived upper bound for the IB loss. Extensive results on computer vision tasks evidence that LTM-Transformer renders compact and efficient visual transformers with comparable or much better prediction accuracy than the original visual transformers. The code of the LTM-Transformer is available at \url{https://github.com/Statistical-Deep-Learning/LTM}.
Authors: P{\aa}l Anders Floor, Ivar Farup, Marius Pedersen
Abstract: As the number of people affected by diseases in the gastrointestinal system is ever-increasing, a higher demand on preventive screening is inevitable. This will significantly increase the workload on gastroenterologists. To help reduce the workload, tools from computer vision may be helpful. In this paper, we investigate the possibility of constructing 3D models of whole sections of the human colon using image sequences from wireless capsule endoscope video, providing enhanced viewing for gastroenterologists. As capsule endoscope images contain distortion and artifacts non-ideal for many 3D reconstruction algorithms, the problem is challenging. However, recent developments of virtual graphics-based models of the human gastrointestinal system, where distortion and artifacts can be enabled or disabled, makes it possible to ``dissect'' the problem. The graphical model also provides a ground truth, enabling computation of geometric distortion introduced by the 3D reconstruction method. In this paper, most distortions and artifacts are left out to determine if it is feasible to reconstruct whole sections of the human gastrointestinal system by existing methods. We demonstrate that 3D reconstruction is possible using simultaneous localization and mapping. Further, to reconstruct the gastrointestinal wall surface from resulting point clouds, varying greatly in density, Poisson surface reconstruction is a good option. The results are promising, encouraging further research on this problem.
Authors: Yu Li, Yifan Chen, Gongye Liu, Jie Wu, Yujiu Yang
Abstract: Layout generation is the foundation task of intelligent design, which requires the integration of visual aesthetics and harmonious expression of content delivery. However, existing methods still face challenges in generating precise and visually appealing layouts, including blocking, overlap, or spatial misalignment between layouts, which are closely related to the spatial structure of graphic layouts. We find that these methods overly focus on content information and lack constraints on layout spatial structure, resulting in an imbalance of learning content-aware and graphic-aware features. To tackle this issue, we propose Content and Graphic Balance Layout Generation with Transformer-based Diffusion Model (CGB-DM). Specifically, we first design a regulator that balances the predicted content and graphic weight, overcoming the tendency of paying more attention to the content on canvas. Secondly, we introduce a graphic constraint of saliency bounding box to further enhance the alignment of geometric features between layout representations and images. In addition, we adapt a transformer-based diffusion model as the backbone, whose powerful generation capability ensures the quality in layout generation. Extensive experimental results indicate that our method has achieved state-of-the-art performance in both quantitative and qualitative evaluations. Our model framework can also be expanded to other graphic design fields.
Authors: Mariya Hendriksen, Shuo Zhang, Ridho Reinanda, Mohamed Yahya, Edgar Meij, Maarten de Rijke
Abstract: Image-text retrieval (ITR), an important task in information retrieval (IR), is driven by pretrained vision-language models (VLMs) that consistently achieve state-of-the-art performance. However, a significant challenge lies in the brittleness of existing ITR benchmarks. In standard datasets for the task, captions often provide broad summaries of scenes, neglecting detailed information about specific concepts. Additionally, the current evaluation setup assumes simplistic binary matches between images and texts and focuses on intra-modality rather than cross-modal relationships, which can lead to misinterpretations of model performance. Motivated by this gap, in this study, we focus on examining the brittleness of the ITR evaluation pipeline with a focus on concept granularity. We start by analyzing two common benchmarks, MS-COCO and Flickr30k, and compare them with their augmented versions, MS-COCO-FG and Flickr30k-FG, given a specified set of linguistic features capturing concept granularity. We discover that Flickr30k-FG and MS COCO-FG consistently achieve higher scores across all the selected features. To investigate the performance of VLMs on coarse and fine-grained datasets, we introduce a taxonomy of perturbations. We apply these perturbations to the selected datasets. We evaluate four state-of-the-art models - ALIGN, AltCLIP, CLIP, and GroupViT - on the standard and fine-grained datasets under zero-shot conditions, with and without the applied perturbations. The results demonstrate that although perturbations generally degrade model performance, the fine-grained datasets exhibit a smaller performance drop than their standard counterparts. Moreover, the relative performance drop across all setups is consistent across all models and datasets, indicating that the issue lies within the benchmarks. We conclude the paper by providing an agenda for improving ITR evaluation pipelines.
Authors: Hanjun Luo, Haoyu Huang, Ziye Deng, Xuecheng Liu, Ruizhe Chen, Zuozhu Liu
Abstract: Text-to-Image (T2I) generative models are becoming more crucial in terms of their ability to generate complex and high-quality images, which also raises concerns about the social biases in their outputs, especially in human generation. Sociological research has established systematic classifications of bias; however, existing research of T2I models often conflates different types of bias, hindering the progress of these methods. In this paper, we introduce BIGbench, a unified benchmark for Biases of Image Generation with a well-designed dataset. In contrast to existing benchmarks, BIGbench classifies and evaluates complex biases into four dimensions: manifestation of bias, visibility of bias, acquired attributes, and protected attributes. Additionally, BIGbench applies advanced multi-modal large language models (MLLM), achieving fully automated evaluation while maintaining high accuracy. We apply BIGbench to evaluate eight recent general T2I models and three debiased methods. We also conduct human evaluation, whose results demonstrated the effectiveness of BIGbench in aligning images and identifying various biases. Besides, our study also revealed new research directions about biases, including the side-effect of irrelevant protected attributes and distillation. Our dataset and benchmark is openly accessible to the research community to ensure the reproducibility.
Authors: Gelei Xu, Ruiyang Qin, Zhi Zheng, Yiyu Shi
Abstract: Timely stress detection is crucial for protecting vulnerable groups from long-term detrimental effects by enabling early intervention. Wearable devices, by collecting real-time physiological signals, offer a solution for accurate stress detection accommodating individual differences. This position paper introduces an adaptive framework for personalized stress detection using PPG and EDA signals. Unlike traditional methods that rely on a generalized model, which may suffer performance drops when applied to new users due to domain shifts, this framework aims to provide each user with a personalized model for higher stress detection accuracy. The framework involves three stages: developing a generalized model offline with an initial dataset, adapting the model to the user's unlabeled data, and fine-tuning it with a small set of labeled data obtained through user interaction. This approach not only offers a foundation for mobile applications that provide personalized stress detection and intervention but also has the potential to address a wider range of mental health issues beyond stress detection using physiological signals.
Authors: Janmesh Ukey, Tushar Kataria, Shireen Y. Elhabian
Abstract: Statistical Shape Models (SSMs) excel at identifying population level anatomical variations, which is at the core of various clinical and biomedical applications, including morphology-based diagnostics and surgical planning. However, the effectiveness of SSM is often constrained by the necessity for expert-driven manual segmentation, a process that is both time-intensive and expensive, thereby restricting their broader application and utility. Recent deep learning approaches enable the direct estimation of Statistical Shape Models (SSMs) from unsegmented images. While these models can predict SSMs without segmentation during deployment, they do not address the challenge of acquiring the manual annotations needed for training, particularly in resource-limited settings. Semi-supervised and foundation models for anatomy segmentation can mitigate the annotation burden. Yet, despite the abundance of available approaches, there are no established guidelines to inform end-users on their effectiveness for the downstream task of constructing SSMs. In this study, we systematically evaluate the potential of weakly supervised methods as viable alternatives to manual segmentation's for building SSMs. We establish a new performance benchmark by employing various semi-supervised and foundational model methods for anatomy segmentation under low annotation settings, utilizing the predicted segmentation's for the task of SSM. We compare the modes of shape variation and use quantitative metrics to compare against a shape model derived from a manually annotated dataset. Our results indicate that some methods produce noisy segmentation, which is very unfavorable for SSM tasks, while others can capture the correct modes of variations in the population cohort with 60-80\% reduction in required manual annotation.
Authors: Haowei Liu, Xi Zhang, Haiyang Xu, Yaya Shi, Chaoya Jiang, Ming Yan, Ji Zhang, Fei Huang, Chunfeng Yuan, Bing Li, Weiming Hu
Abstract: Built on the power of LLMs, numerous multimodal large language models (MLLMs) have recently achieved remarkable performance on various vision-language tasks across multiple benchmarks. However, most existing MLLMs and benchmarks primarily focus on single-image input scenarios, leaving the performance of MLLMs when handling realistic multiple images remain underexplored. Although a few benchmarks consider multiple images, their evaluation dimensions and samples are very limited. Therefore, in this paper, we propose a new benchmark MIBench, to comprehensively evaluate fine-grained abilities of MLLMs in multi-image scenarios. Specifically, MIBench categorizes the multi-image abilities into three scenarios: multi-image instruction (MII), multimodal knowledge-seeking (MKS) and multimodal in-context learning (MIC), and constructs 13 tasks with a total of 13K annotated samples. During data construction, for MII and MKS, we extract correct options from manual annotations and create challenging distractors to obtain multiple-choice questions. For MIC, to enable an in-depth evaluation, we set four sub-tasks and transform the original datasets into in-context learning formats. We evaluate several open-source MLLMs and close-source MLLMs on the proposed MIBench. The results reveal that although current models excel in single-image tasks, they exhibit significant shortcomings when faced with multi-image inputs, such as confused fine-grained perception, limited multi-image reasoning, and unstable in-context learning. The annotated data in MIBench is available at https://huggingface.co/datasets/StarBottle/MIBench.
Authors: Xiaoyang Wu, Xiang Xu, Lingdong Kong, Liang Pan, Ziwei Liu, Tong He, Wanli Ouyang, Hengshuang Zhao
Abstract: In this technical report, we detail our first-place solution for the 2024 Waymo Open Dataset Challenge's semantic segmentation track. We significantly enhanced the performance of Point Transformer V3 on the Waymo benchmark by implementing cutting-edge, plug-and-play training and inference technologies. Notably, our advanced version, Point Transformer V3 Extreme, leverages multi-frame training and a no-clipping-point policy, achieving substantial gains over the original PTv3 performance. Additionally, employing a straightforward model ensemble strategy further boosted our results. This approach secured us the top position on the Waymo Open Dataset semantic segmentation leaderboard, markedly outperforming other entries.
Authors: Jacob Thrasher, Annahita Amireskandari, Prashnna Gyawali
Abstract: Eye diseases are common in older Americans and can lead to decreased vision and blindness. Recent advancements in imaging technologies allow clinicians to capture high-quality images of the retinal blood vessels via Optical Coherence Tomography Angiography (OCTA), which contain vital information for diagnosing these diseases and expediting preventative measures. OCTA provides detailed vascular imaging as compared to the solely structural information obtained by common OCT imaging. Although there have been considerable studies on OCT imaging, there have been limited to no studies exploring the role of artificial intelligence (AI) and machine learning (ML) approaches for predictive modeling with OCTA images. In this paper, we explore the use of deep learning to identify eye disease in OCTA images. However, due to the lack of labeled data, the straightforward application of deep learning doesn't necessarily yield good generalization. To this end, we utilize active learning to select the most valuable subset of data to train our model. We demonstrate that active learning subset selection greatly outperforms other strategies, such as inverse frequency class weighting, random undersampling, and oversampling, by up to 49% in F1 evaluation.
Authors: Mohammad Reza Taesiri, Cor-Paul Bezemer
Abstract: Large multimodal models (LMMs) hold substantial promise across various domains, from personal assistance in daily tasks to sophisticated applications like medical diagnostics. However, their capabilities have limitations in the video game domain, such as challenges with scene understanding, hallucinations, and inaccurate descriptions of video game content, especially in open-source models. This paper describes the development of VideoGameBunny, a LLaVA-style model based on Bunny, specifically tailored for understanding images from video games. We release intermediate checkpoints, training logs, and an extensive dataset comprising 185,259 video game images from 413 titles, along with 389,565 image-instruction pairs that include image captions, question-answer pairs, and a JSON representation of 16 elements of 136,974 images. Our experiments show that our high quality game-related data has the potential to make a relatively small model outperform the much larger state-of-the-art model LLaVa-1.6-34b (which has more than 4x the number of parameters). Our study paves the way for future research in video game understanding on tasks such as playing, commentary, and debugging. Code and data are available at https://videogamebunny.github.io/
Authors: Kwanyong Park, Kuniaki Saito, Donghyun Kim
Abstract: Vision-language (VL) models often exhibit a limited understanding of complex expressions of visual objects (e.g., attributes, shapes, and their relations), given complex and diverse language queries. Traditional approaches attempt to improve VL models using hard negative synthetic text, but their effectiveness is limited. In this paper, we harness the exceptional compositional understanding capabilities of generative foundational models. We introduce a novel method for structured synthetic data generation aimed at enhancing the compositional understanding of VL models in language-based object detection. Our framework generates densely paired positive and negative triplets (image, text descriptions, and bounding boxes) in both image and text domains. By leveraging these synthetic triplets, we transform 'weaker' VL models into 'stronger' models in terms of compositional understanding, a process we call "Weak-to-Strong Compositional Learning" (WSCL). To achieve this, we propose a new compositional contrastive learning formulation that discovers semantics and structures in complex descriptions from synthetic triplets. As a result, VL models trained with our synthetic data generation exhibit a significant performance boost in the Omnilabel benchmark by up to +5AP and the D3 benchmark by +6.9AP upon existing baselines.
Authors: Weiping Ding, Tianyi Zhou, Jiashuang Huang, Shu Jiang, Tao Hou, Chin-Teng Lin
Abstract: Histopathological image classification constitutes a pivotal task in computer-aided diagnostics. The precise identification and categorization of histopathological images are of paramount significance for early disease detection and treatment. In the diagnostic process of pathologists, a multi-tiered approach is typically employed to assess abnormalities in cell regions at different magnifications. However, feature extraction is often performed at a single granularity, overlooking the multi-granular characteristics of cells. To address this issue, we propose the Fuzzy-guided Multi-granularity Deep Neural Network (FMDNN). Inspired by the multi-granular diagnostic approach of pathologists, we perform feature extraction on cell structures at coarse, medium, and fine granularity, enabling the model to fully harness the information in histopathological images. We incorporate the theory of fuzzy logic to address the challenge of redundant key information arising during multi-granular feature extraction. Cell features are described from different perspectives using multiple fuzzy membership functions, which are fused to create universal fuzzy features. A fuzzy-guided cross-attention module guides universal fuzzy features toward multi-granular features. We propagate these features through an encoder to all patch tokens, aiming to achieve enhanced classification accuracy and robustness. In experiments on multiple public datasets, our model exhibits a significant improvement in accuracy over commonly used classification methods for histopathological image classification and shows commendable interpretability.
Authors: Kaiyu Li, Jiawei Jiang, Andrea Codegoni, Chengxi Han, Yupeng Deng, Keyan Chen, Zhuo Zheng, Hao Chen, Zhengxia Zou, Zhenwei Shi, Sheng Fang, Deyu Meng, Zhi Wang, Xiangyong Cao
Abstract: We present Open-CD, a change detection toolbox that contains a rich set of change detection methods as well as related components and modules. The toolbox started from a series of open source general vision task tools, including OpenMMLab Toolkits, PyTorch Image Models, etc. It gradually evolves into a unified platform that covers many popular change detection methods and contemporary modules. It not only includes training and inference codes, but also provides some useful scripts for data analysis. We believe this toolbox is by far the most complete change detection toolbox. In this report, we introduce the various features, supported methods and applications of Open-CD. In addition, we also conduct a benchmarking study on different methods and components. We wish that the toolbox and benchmark could serve the growing research community by providing a flexible toolkit to reimplement existing methods and develop their own new change detectors. Code and models are available at \url{https://github.com/likyoo/open-cd}. Pioneeringly, this report also includes brief descriptions of the algorithms supported in Open-CD, mainly contributed by their authors. We sincerely encourage researchers in this field to participate in this project and work together to create a more open community. This toolkit and report will be kept updated.
Authors: Xiao Liu, Xiaoliu Guan, Yu Wu, Jiaxu Miao
Abstract: Diffusion models, known for their tremendous ability to generate novel and high-quality samples, have recently raised concerns due to their data memorization behavior, which poses privacy risks. Recent approaches for memory mitigation either only focused on the text modality problem in cross-modal generation tasks or utilized data augmentation strategies. In this paper, we propose a novel training framework for diffusion models from the perspective of visual modality, which is more generic and fundamental for mitigating memorization. To facilitate ``forgetting'' of stored information in diffusion model parameters, we propose an iterative ensemble training strategy by splitting the data into multiple shards for training multiple models and intermittently aggregating these model parameters. Moreover, practical analysis of losses illustrates that the training loss for easily memorable images tends to be obviously lower. Thus, we propose an anti-gradient control method to exclude the sample with a lower loss value from the current mini-batch to avoid memorizing. Extensive experiments and analysis on \crnote{four} datasets are conducted to illustrate the effectiveness of our method, and results show that our method successfully reduces memory capacity while even improving the performance slightly. Moreover, to save the computing cost, we successfully apply our method to fine-tune the well-trained diffusion models by limited epochs, demonstrating the applicability of our method. Code is available in https://github.com/liuxiao-guan/IET_AGC.
Authors: Yiran Yang, Xu Gao, Tong Wang, Xin Hao, Yifeng Shi, Xiao Tan, Xiaoqing Ye, Jingdong Wang
Abstract: Camera and LiDAR serve as informative sensors for accurate and robust autonomous driving systems. However, these sensors often exhibit heterogeneous natures, resulting in distributional modality gaps that present significant challenges for fusion. To address this, a robust fusion technique is crucial, particularly for enhancing 3D object detection. In this paper, we introduce a dynamic adjustment technology aimed at aligning modal distributions and learning effective modality representations to enhance the fusion process. Specifically, we propose a triphase domain aligning module. This module adjusts the feature distributions from both the camera and LiDAR, bringing them closer to the ground truth domain and minimizing differences. Additionally, we explore improved representation acquisition methods for dynamic fusion, which includes modal interaction and specialty enhancement. Finally, an adaptive learning technique that merges the semantics and geometry information for dynamical instance optimization. Extensive experiments in the nuScenes dataset present competitive performance with state-of-the-art approaches. Our code will be released in the future.
Authors: Yvette Y. Lin, Xin-Yi Pan, Sara Fridovich-Keil, Gordon Wetzstein
Abstract: Thermal imaging has a variety of applications, from agricultural monitoring to building inspection to imaging under poor visibility, such as in low light, fog, and rain. However, reconstructing thermal scenes in 3D presents several challenges due to the comparatively lower resolution and limited features present in long-wave infrared (LWIR) images. To overcome these challenges, we propose a unified framework for scene reconstruction from a set of LWIR and RGB images, using a multispectral radiance field to represent a scene viewed by both visible and infrared cameras, thus leveraging information across both spectra. We calibrate the RGB and infrared cameras with respect to each other, as a preprocessing step using a simple calibration target. We demonstrate our method on real-world sets of RGB and LWIR photographs captured from a handheld thermal camera, showing the effectiveness of our method at scene representation across the visible and infrared spectra. We show that our method is capable of thermal super-resolution, as well as visually removing obstacles to reveal objects that are occluded in either the RGB or thermal channels. Please see https://yvette256.github.io/thermalnerf for video results as well as our code and dataset release.
Authors: Wenbin An, Feng Tian, Jiahao Nie, Wenkai Shi, Haonan Lin, Yan Chen, QianYing Wang, Yaqiang Wu, Guang Dai, Ping Chen
Abstract: Knowledge-based Visual Question Answering (KVQA) requires both image and world knowledge to answer questions. Current methods first retrieve knowledge from the image and external knowledge base with the original complex question, then generate answers with Large Language Models (LLMs). However, since the original question contains complex elements that require knowledge from different sources, acquiring different kinds of knowledge in a coupled manner may confuse models and hinder them from retrieving precise knowledge. Furthermore, the ``forward-only'' answering process fails to explicitly capture the knowledge needs of LLMs, which can further hurt answering quality. To cope with the above limitations, we propose DKA: Disentangled Knowledge Acquisition from LLM feedback, a training-free framework that disentangles knowledge acquisition to avoid confusion and uses LLM's feedback to specify the required knowledge. Specifically, DKA requires LLMs to specify what knowledge they need to answer the question and decompose the original complex question into two simple sub-questions: Image-based sub-question and Knowledge-based sub-question. Then we use the two sub-questions to retrieve knowledge from the image and knowledge base, respectively. In this way, two knowledge acquisition models can focus on the content that corresponds to them and avoid disturbance of irrelevant elements in the original complex question, which can help to provide more precise knowledge and better align the knowledge needs of LLMs to yield correct answers. Experiments on benchmark datasets show that DKA significantly outperforms SOTA models. To facilitate future research, our data and code are available at \url{https://github.com/Lackel/DKA}.
Authors: Zhongxing Ma, Shuang Liang, Yongkun Wen, Weixin Lu, Guowei Wan
Abstract: Topology reasoning aims to provide a precise understanding of road scenes, enabling autonomous systems to identify safe and efficient routes. In this paper, we present RoadPainter, an innovative approach for detecting and reasoning the topology of lane centerlines using multi-view images. The core concept behind RoadPainter is to extract a set of points from each centerline mask to improve the accuracy of centerline prediction. We start by implementing a transformer decoder that integrates a hybrid attention mechanism and a real-virtual separation strategy to predict coarse lane centerlines and establish topological associations. Then, we generate centerline instance masks guided by the centerline points from the transformer decoder. Moreover, we derive an additional set of points from each mask and combine them with previously detected centerline points for further refinement. Additionally, we introduce an optional module that incorporates a Standard Definition (SD) map to further optimize centerline detection and enhance topological reasoning performance. Experimental evaluations on the OpenLane-V2 dataset demonstrate the state-of-the-art performance of RoadPainter.
Authors: Quan Kong, Yuki Kawana, Rajat Saini, Ashutosh Kumar, Jingjing Pan, Ta Gu, Yohei Ozao, Balazs Opra, David C. Anastasiu, Yoichi Sato, Norimasa Kobori
Abstract: In this paper, we address the challenge of fine-grained video event understanding in traffic scenarios, vital for autonomous driving and safety. Traditional datasets focus on driver or vehicle behavior, often neglecting pedestrian perspectives. To fill this gap, we introduce the WTS dataset, highlighting detailed behaviors of both vehicles and pedestrians across over 1.2k video events in hundreds of traffic scenarios. WTS integrates diverse perspectives from vehicle ego and fixed overhead cameras in a vehicle-infrastructure cooperative environment, enriched with comprehensive textual descriptions and unique 3D Gaze data for a synchronized 2D/3D view, focusing on pedestrian analysis. We also pro-vide annotations for 5k publicly sourced pedestrian-related traffic videos. Additionally, we introduce LLMScorer, an LLM-based evaluation metric to align inference captions with ground truth. Using WTS, we establish a benchmark for dense video-to-text tasks, exploring state-of-the-art Vision-Language Models with an instance-aware VideoLLM method as a baseline. WTS aims to advance fine-grained video event understanding, enhancing traffic safety and autonomous driving development.
Authors: Zhili Chen, Shuangjie Xu, Maosheng Ye, Zian Qian, Xiaoyi Zou, Dit-Yan Yeung, Qifeng Chen
Abstract: The Bird's-Eye-View (BEV) representation is a critical factor that directly impacts the 3D object detection performance, but the traditional BEV grid representation induces quadratic computational cost as the spatial resolution grows. To address this limitation, we present a new camera-based 3D object detector with high-resolution vector representation: VectorFormer. The presented high-resolution vector representation is combined with the lower-resolution BEV representation to efficiently exploit 3D geometry from multi-camera images at a high resolution through our two novel modules: vector scattering and gathering. To this end, the learned vector representation with richer scene contexts can serve as the decoding query for final predictions. We conduct extensive experiments on the nuScenes dataset and demonstrate state-of-the-art performance in NDS and inference time. Furthermore, we investigate query-BEV-based methods incorporated with our proposed vector representation and observe a consistent performance improvement.
Authors: Shuyi Zhang, Ke Liu, Jingjun Gu, Xiaoxu Cai, Zhihua Wang, Jiajun Bu, Haishuai Wang
Abstract: Implicit Neural Representation (INR) has gained increasing popularity as a data representation method, serving as a prerequisite for innovative generation models. Unlike gradient-based methods, which exhibit lower efficiency in inference, the adoption of hyper-network for generating parameters in Multi-Layer Perceptrons (MLP), responsible for executing INR functions, has surfaced as a promising and efficient alternative. However, as a global continuous function, MLP is challenging in modeling highly discontinuous signals, resulting in slow convergence during the training phase and inaccurate reconstruction performance. Moreover, MLP requires massive representation parameters, which implies inefficiencies in data representation. In this paper, we propose a novel Attention-based Localized INR (ANR) composed of a localized attention layer (LAL) and a global MLP that integrates coordinate features with data features and converts them to meaningful outputs. Subsequently, we design an instance representation framework that delivers a transformer-like hyper-network to represent data instances as a compact representation vector. With instance-specific representation vector and instance-agnostic ANR parameters, the target signals are well reconstructed as a continuous function. We further address aliasing artifacts with variational coordinates when obtaining the super-resolution inference results. Extensive experimentation across four datasets showcases the notable efficacy of our ANR method, e.g. enhancing the PSNR value from 37.95dB to 47.25dB on the CelebA dataset. Code is released at https://github.com/Roninton/ANR.
Authors: Yunpeng Wang, Kang Wang, Yaoyao Zhuo, Weiya Shi, Fei Shan, Lei Liu
Abstract: Rapid and accurate diagnosis of pneumothorax, utilizing chest X-ray and computed tomography (CT), is crucial for assisted diagnosis. Chest X-ray is commonly used for initial localization of pneumothorax, while CT ensures accurate quantification. However, CT scans involve high radiation doses and can be costly. To achieve precise quantitative diagnosis while minimizing radiation exposure, we proposed X-Recon, a CT ultra-sparse reconstruction network based on ortho-lateral chest X-ray images. X-Recon integrates generative adversarial networks (GANs), including a generator with a multi-scale fusion rendering module and a discriminator enhanced by 3D coordinate convolutional layers, designed to facilitate CT reconstruction. To improve precision, a projective spatial transformer is utilized to incorporate multi-angle projection loss. Additionally, we proposed PTX-Seg, a zero-shot pneumothorax segmentation algorithm, combining image processing techniques with deep-learning models for the segmentation of air-accumulated regions and lung structures. Experiments on a large-scale dataset demonstrate its superiority over existing approaches. X-Recon achieved a significantly higher reconstruction resolution with a higher average spatial resolution and a lower average slice thickness. The reconstruction metrics achieved state-of-the-art performance in terms of several metrics including peak signal-to-noise ratio. The zero-shot segmentation algorithm, PTX-Seg, also demonstrated high segmentation precision for the air-accumulated region, the left lung, and the right lung. Moreover, the consistency analysis for the pneumothorax chest occupancy ratio between reconstructed CT and original CT obtained a high correlation coefficient. Code will be available at: https://github.com/wangyunpengbio/X-Recon
Authors: Yingxue Xu, Yihui Wang, Fengtao Zhou, Jiabo Ma, Shu Yang, Huangjing Lin, Xin Wang, Jiguang Wang, Li Liang, Anjia Han, Ronald Cheong Kin Chan, Hao Chen
Abstract: Remarkable strides in computational pathology have been made in the task-agnostic foundation model that advances the performance of a wide array of downstream clinical tasks. Despite the promising performance, there are still several challenges. First, prior works have resorted to either vision-only or vision-captions data, disregarding invaluable pathology reports and gene expression profiles which respectively offer distinct knowledge for versatile clinical applications. Second, the current progress in pathology FMs predominantly concentrates on the patch level, where the restricted context of patch-level pretraining fails to capture whole-slide patterns. Here we curated the largest multimodal dataset consisting of H\&E diagnostic whole slide images and their associated pathology reports and RNA-Seq data, resulting in 26,169 slide-level modality pairs from 10,275 patients across 32 cancer types. To leverage these data for CPath, we propose a novel whole-slide pretraining paradigm which injects multimodal knowledge at the whole-slide context into the pathology FM, called Multimodal Self-TAught PRetraining (mSTAR). The proposed paradigm revolutionizes the workflow of pretraining for CPath, which enables the pathology FM to acquire the whole-slide context. To our knowledge, this is the first attempt to incorporate multimodal knowledge at the slide level for enhancing pathology FMs, expanding the modelling context from unimodal to multimodal knowledge and from patch-level to slide-level. To systematically evaluate the capabilities of mSTAR, extensive experiments including slide-level unimodal and multimodal applications, are conducted across 7 diverse types of tasks on 43 subtasks, resulting in the largest spectrum of downstream tasks. The average performance in various slide-level applications consistently demonstrates significant performance enhancements for mSTAR compared to SOTA FMs.
Authors: Fei Zhou, Maixia Fu, Yulei Qian, Jian Yang, Yimian Dai
Abstract: Infrared small target detection is crucial for the efficacy of infrared search and tracking systems. Current tensor decomposition methods emphasize representing small targets with sparsity but struggle to separate targets from complex backgrounds due to insufficient use of intrinsic directional information and reduced target visibility during decomposition. To address these challenges, this study introduces a Sparse Differential Directionality prior (SDD) framework. SDD leverages the distinct directional characteristics of targets to differentiate them from the background, applying mixed sparse constraints on the differential directional images and continuity difference matrix of the temporal component, both derived from Tucker decomposition. We further enhance target detectability with a saliency coherence strategy that intensifies target contrast against the background during hierarchical decomposition. A Proximal Alternating Minimization-based (PAM) algorithm efficiently solves our proposed model. Experimental results on several real-world datasets validate our method's effectiveness, outperforming ten state-of-the-art methods in target detection and clutter suppression. Our code is available at https://github.com/GrokCV/SDD.
Authors: Junha Song, Tae Soo Kim, Junha Kim, Gunhee Nam, Thijs Kooi, Jaegul Choo
Abstract: This paper aims to adapt the source model to the target environment, leveraging small user feedback (i.e., labeled target data) readily available in real-world applications. We find that existing semi-supervised domain adaptation (SemiSDA) methods often suffer from poorly improved adaptation performance when directly utilizing such feedback data, as shown in Figure 1. We analyze this phenomenon via a novel concept called Negatively Biased Feedback (NBF), which stems from the observation that user feedback is more likely for data points where the model produces incorrect predictions. To leverage this feedback while avoiding the issue, we propose a scalable adapting approach, Retrieval Latent Defending. This approach helps existing SemiSDA methods to adapt the model with a balanced supervised signal by utilizing latent defending samples throughout the adaptation process. We demonstrate the problem caused by NBF and the efficacy of our approach across various benchmarks, including image classification, semantic segmentation, and a real-world medical imaging application. Our extensive experiments reveal that integrating our approach with multiple state-of-the-art SemiSDA methods leads to significant performance improvements.
Authors: Fudong Lin, Jiadong Lou, Xu Yuan, Nian-Feng Tzeng
Abstract: Adversarial training (AT) can help improve the robustness of Vision Transformers (ViT) against adversarial attacks by intentionally injecting adversarial examples into the training data. However, this way of adversarial injection inevitably incurs standard accuracy degradation to some extent, thereby calling for a trade-off between standard accuracy and robustness. Besides, the prominent AT solutions are still vulnerable to adaptive attacks. To tackle such shortcomings, this paper proposes a novel ViT architecture, including a detector and a classifier bridged by our newly developed adaptive ensemble. Specifically, we empirically discover that detecting adversarial examples can benefit from the Guided Backpropagation technique. Driven by this discovery, a novel Multi-head Self-Attention (MSA) mechanism is introduced to enhance our detector to sniff adversarial examples. Then, a classifier with two encoders is employed for extracting visual representations respectively from clean images and adversarial examples, with our adaptive ensemble to adaptively adjust the proportion of visual representations from the two encoders for accurate classification. This design enables our ViT architecture to achieve a better trade-off between standard accuracy and robustness. Besides, our adaptive ensemble technique allows us to mask off a random subset of image patches within input data, boosting our ViT's robustness against adaptive attacks, while maintaining high standard accuracy. Experimental results exhibit that our ViT architecture, on CIFAR-10, achieves the best standard accuracy and adversarial robustness of 90.3% and 49.8%, respectively.
Authors: Jaehyeong Jeon, Kibum Kim, Kanghoon Yoon, Chanyoung Park
Abstract: The scene graph generation (SGG) task involves detecting objects within an image and predicting predicates that represent the relationships between the objects. However, in SGG benchmark datasets, each subject-object pair is annotated with a single predicate even though a single predicate may exhibit diverse semantics (i.e., semantic diversity), existing SGG models are trained to predict the one and only predicate for each pair. This in turn results in the SGG models to overlook the semantic diversity that may exist in a predicate, thus leading to biased predictions. In this paper, we propose a novel model-agnostic Semantic Diversity-aware Prototype-based Learning (DPL) framework that enables unbiased predictions based on the understanding of the semantic diversity of predicates. Specifically, DPL learns the regions in the semantic space covered by each predicate to distinguish among the various different semantics that a single predicate can represent. Extensive experiments demonstrate that our proposed model-agnostic DPL framework brings significant performance improvement on existing SGG models, and also effectively understands the semantic diversity of predicates.
Authors: Kent Fujiwara, Mikihiro Tanaka, Qing Yu
Abstract: With the release of large-scale motion datasets with textual annotations, the task of establishing a robust latent space for language and 3D human motion has recently witnessed a surge of interest. Methods have been proposed to convert human motion and texts into features to achieve accurate correspondence between them. Despite these efforts to align language and motion representations, we claim that the temporal element is often overlooked, especially for compound actions, resulting in chronological inaccuracies. To shed light on the temporal alignment in motion-language latent spaces, we propose Chronologically Accurate Retrieval (CAR) to evaluate the chronological understanding of the models. We decompose textual descriptions into events, and prepare negative text samples by shuffling the order of events in compound action descriptions. We then design a simple task for motion-language models to retrieve the more likely text from the ground truth and its chronologically shuffled version. CAR reveals many cases where current motion-language models fail to distinguish the event chronology of human motion, despite their impressive performance in terms of conventional evaluation metrics. To achieve better temporal alignment between text and motion, we further propose to use these texts with shuffled sequence of events as negative samples during training to reinforce the motion-language models. We conduct experiments on text-motion retrieval and text-to-motion generation using the reinforced motion-language models, which demonstrate improved performance over conventional approaches, indicating the necessity to consider temporal elements in motion-language alignment.
Authors: Seokju Cho, Jiahui Huang, Jisu Nam, Honggyu An, Seungryong Kim, Joon-Young Lee
Abstract: We introduce LocoTrack, a highly accurate and efficient model designed for the task of tracking any point (TAP) across video sequences. Previous approaches in this task often rely on local 2D correlation maps to establish correspondences from a point in the query image to a local region in the target image, which often struggle with homogeneous regions or repetitive features, leading to matching ambiguities. LocoTrack overcomes this challenge with a novel approach that utilizes all-pair correspondences across regions, i.e., local 4D correlation, to establish precise correspondences, with bidirectional correspondence and matching smoothness significantly enhancing robustness against ambiguities. We also incorporate a lightweight correlation encoder to enhance computational efficiency, and a compact Transformer architecture to integrate long-term temporal information. LocoTrack achieves unmatched accuracy on all TAP-Vid benchmarks and operates at a speed almost 6 times faster than the current state-of-the-art.
Authors: Jiahao Lyu, Minghua Zhao, Jing Hu, Runtao Xi, Xuewen Huang, Shuangli Du, Cheng Shi, Tian Ma
Abstract: With the widespread deployment of video surveillance devices and the demand for intelligent system development, video anomaly detection (VAD) has become an important part of constructing intelligent surveillance systems. Expanding the discriminative boundary between normal and abnormal events to enhance performance is the common goal and challenge of VAD. To address this problem, we propose a Bidirectional Skip-frame Prediction (BiSP) network based on a dual-stream autoencoder, from the perspective of learning the intra-domain disparity between different features. The BiSP skips frames in the training phase to achieve the forward and backward frame prediction respectively, and in the testing phase, it utilizes bidirectional consecutive frames to co-predict the same intermediate frames, thus expanding the degree of disparity between normal and abnormal events. The BiSP designs the variance channel attention and context spatial attention from the perspectives of movement patterns and object scales, respectively, thus ensuring the maximization of the disparity between normal and abnormal in the feature extraction and delivery with different dimensions. Extensive experiments from four benchmark datasets demonstrate the effectiveness of the proposed BiSP, which substantially outperforms state-of-the-art competing methods.
Authors: Bowen Liu, Dongjie Chen, Xiao Qi
Abstract: With the rapid growth of the PCB manufacturing industry, there is an increasing demand for computer vision inspection to detect defects during production. Improving the accuracy and generalization of PCB defect detection models remains a significant challenge. This paper proposes a high-precision, robust, and real-time end-to-end method for PCB defect detection based on deep Convolutional Neural Networks (CNN). Traditional methods often suffer from low accuracy and limited applicability. We propose a novel approach combining YOLOv5 and multiscale modules for hierarchical residual-like connections. In PCB defect detection, noise can confuse the background and small targets. The YOLOv5 model provides a strong foundation with its real-time processing and accurate object detection capabilities. The multi-scale module extends traditional approaches by incorporating hierarchical residual-like connections within a single block, enabling multiscale feature extraction. This plug-and-play module significantly enhances performance by extracting features at multiple scales and levels, which are useful for identifying defects of varying sizes and complexities. Our multi-scale architecture integrates feature extraction, defect localization, and classification into a unified network. Experiments on a large-scale PCB dataset demonstrate significant improvements in precision, recall, and F1-score compared to existing methods. This work advances computer vision inspection for PCB defect detection, providing a reliable solution for high-precision, robust, real-time, and domain-adaptive defect detection in the PCB manufacturing industry.
Authors: Bo Yuan, Danpei Zhao, Zhenwei Shi
Abstract: Continual semantic segmentation (CSS) based on incremental learning (IL) is a great endeavour in developing human-like segmentation models. However, current CSS approaches encounter challenges in the trade-off between preserving old knowledge and learning new ones, where they still need large-scale annotated data for incremental training and lack interpretability. In this paper, we present Learning at a Glance (LAG), an efficient, robust, human-like and interpretable approach for CSS. Specifically, LAG is a simple and model-agnostic architecture, yet it achieves competitive CSS efficiency with limited incremental data. Inspired by human-like recognition patterns, we propose a semantic-invariance modelling approach via semantic features decoupling that simultaneously reconciles solid knowledge inheritance and new-term learning. Concretely, the proposed decoupling manner includes two ways, i.e., channel-wise decoupling and spatial-level neuron-relevant semantic consistency. Our approach preserves semantic-invariant knowledge as solid prototypes to alleviate catastrophic forgetting, while also constraining sample-specific contents through an asymmetric contrastive learning method to enhance model robustness during IL steps. Experimental results in multiple datasets validate the effectiveness of the proposed method. Furthermore, we introduce a novel CSS protocol that better reflects realistic data-limited CSS settings, and LAG achieves superior performance under multiple data-limited conditions.
Authors: Ruizhe Wang, Chunliang Hua, Tomakayev Shingys, Mengyuan Niu, Qingxin Yang, Lizhong Gao, Yi Zheng, Junyan Yang, Qiao Wang
Abstract: The photorealistic reconstruction and rendering of architectural scenes have extensive applications in industries such as film, games, and transportation. It also plays an important role in urban planning, architectural design, and the city's promotion, especially in protecting historical and cultural relics. The 3D Gaussian Splatting, due to better performance over NeRF, has become a mainstream technology in 3D reconstruction. Its only input is a set of images but it relies heavily on geometric parameters computed by the SfM process. At the same time, there is an existing abundance of raw 3D models, that could inform the structural perception of certain buildings but cannot be applied. In this paper, we propose a straightforward method to harness these raw 3D models to guide 3D Gaussians in capturing the basic shape of the building and improve the visual quality of textures and details when photos are captured non-systematically. This exploration opens up new possibilities for improving the effectiveness of 3D reconstruction techniques in the field of architectural design.
Authors: Rishubh Parihar, Harsh Gupta, Sachidanand VS, R. Venkatesh Babu
Abstract: For a given scene, humans can easily reason for the locations and pose to place objects. Designing a computational model to reason about these affordances poses a significant challenge, mirroring the intuitive reasoning abilities of humans. This work tackles the problem of realistic human insertion in a given background scene termed as \textbf{Semantic Human Placement}. This task is extremely challenging given the diverse backgrounds, scale, and pose of the generated person and, finally, the identity preservation of the person. We divide the problem into the following two stages \textbf{i)} learning \textit{semantic masks} using text guidance for localizing regions in the image to place humans and \textbf{ii)} subject-conditioned inpainting to place a given subject adhering to the scene affordance within the \textit{semantic masks}. For learning semantic masks, we leverage rich object-scene priors learned from the text-to-image generative models and optimize a novel parameterization of the semantic mask, eliminating the need for large-scale training. To the best of our knowledge, we are the first ones to provide an effective solution for realistic human placements in diverse real-world scenes. The proposed method can generate highly realistic scene compositions while preserving the background and subject identity. Further, we present results for several downstream tasks - scene hallucination from a single or multiple generated persons and text-based attribute editing. With extensive comparisons against strong baselines, we show the superiority of our method in realistic human placement.
Authors: Mohammadreza Salehi, Michael Dorkenwald, Fida Mohammad Thoker, Efstratios Gavves, Cees G. M. Snoek, Yuki M. Asano
Abstract: Video-based pretraining offers immense potential for learning strong visual representations on an unprecedented scale. Recently, masked video modeling methods have shown promising scalability, yet fall short in capturing higher-level semantics due to reconstructing predefined low-level targets such as pixels. To tackle this, we present Sinkhorn-guided Masked Video Modelling (SIGMA), a novel video pretraining method that jointly learns the video model in addition to a target feature space using a projection network. However, this simple modification means that the regular L2 reconstruction loss will lead to trivial solutions as both networks are jointly optimized. As a solution, we distribute features of space-time tubes evenly across a limited number of learnable clusters. By posing this as an optimal transport problem, we enforce high entropy in the generated features across the batch, infusing semantic and temporal meaning into the feature space. The resulting cluster assignments are used as targets for a symmetric prediction task where the video model predicts cluster assignment of the projection network and vice versa. Experimental results on ten datasets across three benchmarks validate the effectiveness of SIGMA in learning more performant, temporally-aware, and robust video representations improving upon state-of-the-art methods. Our project website with code is available at: https://quva-lab.github.io/SIGMA.
Authors: Yihao Ai, Yifei Qi, Bo Wang, Yu Cheng, Xinchao Wang, Robby T. Tan
Abstract: Existing 2D human pose estimation research predominantly concentrates on well-lit scenarios, with limited exploration of poor lighting conditions, which are a prevalent aspect of daily life. Recent studies on low-light pose estimation require the use of paired well-lit and low-light images with ground truths for training, which are impractical due to the inherent challenges associated with annotation on low-light images. To this end, we introduce a novel approach that eliminates the need for low-light ground truths. Our primary novelty lies in leveraging two complementary-teacher networks to generate more reliable pseudo labels, enabling our model achieves competitive performance on extremely low-light images without the need for training with low-light ground truths. Our framework consists of two stages. In the first stage, our model is trained on well-lit data with low-light augmentations. In the second stage, we propose a dual-teacher framework to utilize the unlabeled low-light data, where a center-based main teacher produces the pseudo labels for relatively visible cases, while a keypoints-based complementary teacher focuses on producing the pseudo labels for the missed persons of the main teacher. With the pseudo labels from both teachers, we propose a person-specific low-light augmentation to challenge a student model in training to outperform the teachers. Experimental results on real low-light dataset (ExLPose-OCN) show, our method achieves 6.8% (2.4 AP) improvement over the state-of-the-art (SOTA) method, despite no low-light ground-truth data is used in our approach, in contrast to the SOTA method. Our code will be available at:https://github.com/ayh015-dev/DA-LLPose.
Authors: Anis Amziane
Abstract: Multispectral (MS) snapshot cameras equipped with a MS filter array (MSFA), capture multiple spectral bands in a single shot, resulting in a raw mosaic image where each pixel holds only one channel value. The fully-defined MS image is estimated from the raw one through $\textit{demosaicing}$, which inevitably introduces spatio-spectral artifacts. Moreover, training on fully-defined MS images can be computationally intensive, particularly with deep neural networks (DNNs), and may result in features lacking discrimination power due to suboptimal learning of spatio-spectral interactions. Furthermore, outdoor MS image acquisition occurs under varying lighting conditions, leading to illumination-dependent features. This paper presents an original approach to learn discriminant and illumination-robust features directly from raw images. It involves: $\textit{raw spectral constancy}$ to mitigate the impact of illumination, $\textit{MSFA-preserving}$ transformations suited for raw image augmentation to train DNNs on diverse raw textures, and $\textit{raw-mixing}$ to capture discriminant spatio-spectral interactions in raw images. Experiments on MS image classification show that our approach outperforms both handcrafted and recent deep learning-based methods, while also requiring significantly less computational effort.~The source code will be available.
Authors: \.Ismail \"Oz\c{c}\.il, A. Bu\u{g}ra Koku
Abstract: The advancement in computing power has significantly reduced the training times for deep learning, fostering the rapid development of networks designed for object recognition. However, the exploration of object utility, which is the affordance of the object, as opposed to object recognition, has received comparatively less attention. This work focuses on the problem of exploration of object affordances using existing networks trained on the object classification dataset. While pre-trained networks have proven to be instrumental in transfer learning for classification tasks, this work diverges from conventional object classification methods. Instead, it employs pre-trained networks to discern affordance labels without the need for specialized layers, abstaining from modifying the final layers through the addition of classification layers. To facilitate the determination of affordance labels without such modifications, two approaches, i.e. subspace clustering and manifold curvature methods are tested. These methods offer a distinct perspective on affordance label recognition. Especially, manifold curvature method has been successfully tested with nine distinct pre-trained networks, each achieving an accuracy exceeding 95%. Moreover, it is observed that manifold curvature and subspace clustering methods explore affordance labels that are not marked in the ground truth, but object affords in various cases.
Authors: Xinhao Tao, Tianyuan Qiu, Junyan Cao, Li Niu
Abstract: Image harmonization aims to adjust the foreground illumination in a composite image to make it harmonious. The existing harmonization methods can only produce one deterministic result for a composite image, ignoring that a composite image could have multiple plausible harmonization results due to multiple plausible reflectances. In this work, we first propose a reflectance-guided harmonization network, which can achieve better performance with the guidance of ground-truth foreground reflectance. Then, we also design a diverse reflectance generation network to predict multiple plausible foreground reflectances, leading to multiple plausible harmonization results. The extensive experiments on the benchmark datasets demonstrate the effectiveness of our method.
Authors: Matteo Bortolon, Theodore Tsesmelis, Stuart James, Fabio Poiesi, Alessio Del Bue
Abstract: We propose 6DGS to estimate the camera pose of a target RGB image given a 3D Gaussian Splatting (3DGS) model representing the scene. 6DGS avoids the iterative process typical of analysis-by-synthesis methods (e.g. iNeRF) that also require an initialization of the camera pose in order to converge. Instead, our method estimates a 6DoF pose by inverting the 3DGS rendering process. Starting from the object surface, we define a radiant Ellicell that uniformly generates rays departing from each ellipsoid that parameterize the 3DGS model. Each Ellicell ray is associated with the rendering parameters of each ellipsoid, which in turn is used to obtain the best bindings between the target image pixels and the cast rays. These pixel-ray bindings are then ranked to select the best scoring bundle of rays, which their intersection provides the camera center and, in turn, the camera rotation. The proposed solution obviates the necessity of an "a priori" pose for initialization, and it solves 6DoF pose estimation in closed form, without the need for iterations. Moreover, compared to the existing Novel View Synthesis (NVS) baselines for pose estimation, 6DGS can improve the overall average rotational accuracy by 12% and translation accuracy by 22% on real scenes, despite not requiring any initialization pose. At the same time, our method operates near real-time, reaching 15fps on consumer hardware.
Authors: Matteo Nulli, Anesa Ibrahimi, Avik Pal, Hoshe Lee, Ivona Najdenkoska
Abstract: Vision-Language Models (VLMs) have shown remarkable capabilities in a large number of downstream tasks. Nonetheless, compositional image understanding remains a rather difficult task due to the object bias present in training data. In this work, we investigate the reasons for such a lack of capability by performing an extensive bench-marking of compositional understanding in VLMs. We compare contrastive models with generative ones and analyze their differences in architecture, pre-training data, and training tasks and losses. Furthermore, we leverage In-Context Learning (ICL) as a way to improve the ability of VLMs to perform more complex reasoning and understanding given an image. Our extensive experiments demonstrate that our proposed approach outperforms baseline models across multiple compositional understanding datasets.
Authors: Zeyu Wang, Jingyu Lin, Yifei Qian, Yi Huang, Shicen Tian, Bosong Chai, Juncan Deng, Lan Du, Cunjian Chen, Yufei Guo, Kejie Huang
Abstract: Diffusion models have made significant strides in text-driven and layout-driven image generation. However, most diffusion models are limited to visible RGB image generation. In fact, human perception of the world is enriched by diverse viewpoints, including chromatic contrast, thermal illumination, and depth information. In this paper, we introduce a novel diffusion model for general layout-guided cross-modal "RGB+X" generation, called DiffX. We firstly construct the cross-modal image datasets with text descriptions using the LLaVA model for image captioning, supplemented by manual corrections. Notably, DiffX presents a simple yet effective cross-modal generative modeling pipeline, which conducts diffusion and denoising processes in the modality-shared latent space, facilitated by our Dual-Path Variational AutoEncoder (DP-VAE). Furthermore, we incorporate the gated cross-attention mechanism to connect the layout and text conditions, leveraging Long-CLIP for embedding long captions to enhance user guidance. Through extensive experiments, DiffX demonstrates robustness and flexibility in cross-modal generation across three RGB+X datasets: FLIR, MFNet, and COME15K, guided by various layout types. It also shows the potential for adaptive generation of "RGB+X+Y" or more diverse modalities. Our code and processed image captions are available at https://github.com/zeyuwang-zju/DiffX.
Authors: Despina Konstantinidou, Christos Koutlis, Symeon Papadopoulos
Abstract: Generative AI technologies produce hyper-realistic imagery that can be used for nefarious purposes such as producing misleading or harmful content, among others. This makes Synthetic Image Detection (SID) an essential tool for defending against AI-generated harmful content. Current SID methods typically resize input images to a fixed resolution or perform center-cropping due to computational concerns, leading to challenges in effectively detecting artifacts in high-resolution images. To this end, we propose TextureCrop, a novel image pre-processing technique. By focusing on high-frequency image parts where generation artifacts are prevalent, TextureCrop effectively enhances SID accuracy while maintaining manageable memory requirements. Experimental results demonstrate a consistent improvement in AUC across various detectors by 5.7% compared to center cropping and by 14% compared to resizing, across high-resolution images from the Forensynths and Synthbuster datasets.
Authors: Zirui Shao, Feiyu Gao, Hangdi Xing, Zepeng Zhu, Zhi Yu, Jiajun Bu, Qi Zheng, Cong Yao
Abstract: In the era of content creation revolution propelled by advancements in generative models, the field of web design remains unexplored despite its critical role in modern digital communication. The web design process is complex and often time-consuming, especially for those with limited expertise. In this paper, we introduce Web Rendering Parameters Generation (WebRPG), a new task that aims at automating the generation for visual presentation of web pages based on their HTML code. WebRPG would contribute to a faster web development workflow. Since there is no existing benchmark available, we develop a new dataset for WebRPG through an automated pipeline. Moreover, we present baseline models, utilizing VAE to manage numerous elements and rendering parameters, along with custom HTML embedding for capturing essential semantic and hierarchical information from HTML. Extensive experiments, including customized quantitative evaluations for this specific task, are conducted to evaluate the quality of the generated results.
Authors: Stanislav Frolov, Brian B. Moser, Andreas Dengel
Abstract: Generating high-resolution images with generative models has recently been made widely accessible by leveraging diffusion models pre-trained on large-scale datasets. Various techniques, such as MultiDiffusion and SyncDiffusion, have further pushed image generation beyond training resolutions, i.e., from square images to panorama, by merging multiple overlapping diffusion paths or employing gradient descent to maintain perceptual coherence. However, these methods suffer from significant computational inefficiencies due to generating and averaging numerous predictions, which is required in practice to produce high-quality and seamless images. This work addresses this limitation and presents a novel approach that eliminates the need to generate and average numerous overlapping denoising predictions. Our method shifts non-overlapping denoising windows over time, ensuring that seams in one timestep are corrected in the next. This results in coherent, high-resolution images with fewer overall steps. We demonstrate the effectiveness of our approach through qualitative and quantitative evaluations, comparing it with MultiDiffusion, SyncDiffusion, and StitchDiffusion. Our method offers several key benefits, including improved computational efficiency and faster inference times while producing comparable or better image quality.
Authors: Eugenio Lomurno, Matteo Matteucci
Abstract: Generative artificial intelligence has transformed the generation of synthetic data, providing innovative solutions to challenges like data scarcity and privacy, which are particularly critical in fields such as medicine. However, the effective use of this synthetic data to train high-performance models remains a significant challenge. This paper addresses this issue by introducing Knowledge Recycling (KR), a pipeline designed to optimise the generation and use of synthetic data for training downstream classifiers. At the heart of this pipeline is Generative Knowledge Distillation (GKD), the proposed technique that significantly improves the quality and usefulness of the information provided to classifiers through a synthetic dataset regeneration and soft labelling mechanism. The KR pipeline has been tested on a variety of datasets, with a focus on six highly heterogeneous medical image datasets, ranging from retinal images to organ scans. The results show a significant reduction in the performance gap between models trained on real and synthetic data, with models based on synthetic data outperforming those trained on real data in some cases. Furthermore, the resulting models show almost complete immunity to Membership Inference Attacks, manifesting privacy properties missing in models trained with conventional techniques.
Authors: Abdelrahman Seleem (Instituto Superior T\'ecnico - Universidade de Lisboa, Lisbon, Portugal), Andr\'e F. R. Guarda (Instituto de Telecomunica\c{c}\~oes, Portugal), Nuno M. M. Rodrigues (Instituto de Telecomunica\c{c}\~oes, Portugal), Fernando Pereira (Instituto Superior T\'ecnico - Universidade de Lisboa, Lisbon, Portugal)
Abstract: Event cameras have the ability to capture asynchronous per-pixel brightness changes, called "events", offering advantages over traditional frame-based cameras for computer vision applications. Efficiently coding event data is critical for transmission and storage, given the significant volume of events. This paper proposes a novel double deep learning-based architecture for both event data coding and classification, using a point cloud-based representation for events. In this context, the conversions from events to point clouds and back to events are key steps in the proposed solution, and therefore its impact is evaluated in terms of compression and classification performance. Experimental results show that it is possible to achieve a classification performance of compressed events which is similar to one of the original events, even after applying a lossy point cloud codec, notably the recent learning-based JPEG Pleno Point Cloud Coding standard, with a clear rate reduction. Experimental results also demonstrate that events coded using JPEG PCC achieve better classification performance than those coded using the conventional lossy MPEG Geometry-based Point Cloud Coding standard. Furthermore, the adoption of learning-based coding offers high potential for performing computer vision tasks in the compressed domain, which allows skipping the decoding stage while mitigating the impact of coding artifacts.
Authors: Zakaria Laskar, Iaroslav Melekhov, Assia Benbihi, Shuzhe Wang, Juho Kannala
Abstract: Camera relocalization relies on 3D models of the scene with a large memory footprint that is incompatible with the memory budget of several applications. One solution to reduce the scene memory size is map compression by removing certain 3D points and descriptor quantization. This achieves high compression but leads to performance drop due to information loss. To address the memory performance trade-off, we train a light-weight scene-specific auto-encoder network that performs descriptor quantization-dequantization in an end-to-end differentiable manner updating both product quantization centroids and network parameters through back-propagation. In addition to optimizing the network for descriptor reconstruction, we encourage it to preserve the descriptor-matching performance with margin-based metric loss functions. Results show that for a local descriptor memory of only 1MB, the synergistic combination of the proposed network and map compression achieves the best performance on the Aachen Day-Night compared to existing compression methods.
Authors: Minseong Park, Suhan Woo, Euntai Kim
Abstract: Learning efficient representations of local features is a key challenge in feature volume-based 3D neural mapping, especially in large-scale environments. In this paper, we introduce Decomposition-based Neural Mapping (DNMap), a storage-efficient large-scale 3D mapping method that employs a discrete representation based on a decomposition strategy. This decomposition strategy aims to efficiently capture repetitive and representative patterns of shapes by decomposing each discrete embedding into component vectors that are shared across the embedding space. Our DNMap optimizes a set of component vectors, rather than entire discrete embeddings, and learns composition rather than indexing the discrete embeddings. Furthermore, to complement the mapping quality, we additionally learn low-resolution continuous embeddings that require tiny storage space. By combining these representations with a shallow neural network and an efficient octree-based feature volume, our DNMap successfully approximates signed distance functions and compresses the feature volume while preserving mapping quality. Our source code is available at https://github.com/minseong-p/dnmap.
Authors: Yang Liu, Qianqian Xu, Peisong Wen, Siran Dai, Qingming Huang
Abstract: The rapid growth of online video resources has significantly promoted the development of video retrieval methods. As a standard evaluation metric for video retrieval, Average Precision (AP) assesses the overall rankings of relevant videos at the top list, making the predicted scores a reliable reference for users. However, recent video retrieval methods utilize pair-wise losses that treat all sample pairs equally, leading to an evident gap between the training objective and evaluation metric. To effectively bridge this gap, in this work, we aim to address two primary challenges: a) The current similarity measure and AP-based loss are suboptimal for video retrieval; b) The noticeable noise from frame-to-frame matching introduces ambiguity in estimating the AP loss. In response to these challenges, we propose the Hierarchical learning framework for Average-Precision-oriented Video Retrieval (HAP-VR). For the former challenge, we develop the TopK-Chamfer Similarity and QuadLinear-AP loss to measure and optimize video-level similarities in terms of AP. For the latter challenge, we suggest constraining the frame-level similarities to achieve an accurate AP loss estimation. Experimental results present that HAP-VR outperforms existing methods on several benchmark datasets, providing a feasible solution for video retrieval tasks and thus offering potential benefits for the multi-media application.
Authors: Amir Mohammad Karimi Mamaghan, Samuele Papa, Karl Henrik Johansson, Stefan Bauer, Andrea Dittadi
Abstract: Object-centric (OC) representations, which represent the state of a visual scene by modeling it as a composition of objects, have the potential to be used in various downstream tasks to achieve systematic compositional generalization and facilitate reasoning. However, these claims have not been thoroughly analyzed yet. Recently, foundation models have demonstrated unparalleled capabilities across diverse domains from language to computer vision, marking them as a potential cornerstone of future research for a multitude of computational tasks. In this paper, we conduct an extensive empirical study on representation learning for downstream Visual Question Answering (VQA), which requires an accurate compositional understanding of the scene. We thoroughly investigate the benefits and trade-offs of OC models and alternative approaches including large pre-trained foundation models on both synthetic and real-world data, and demonstrate a viable way to achieve the best of both worlds. The extensiveness of our study, encompassing over 800 downstream VQA models and 15 different types of upstream representations, also provides several additional insights that we believe will be of interest to the community at large.
Authors: Xinji Mai, Junxiong Lin, Haoran Wang, Zeng Tao, Yan Wang, Shaoqi Yan, Xuan Tong, Jiawen Yu, Boyang Wang, Ziheng Zhou, Qing Zhao, Shuyong Gao, Wenqiang Zhang
Abstract: In the field of affective computing, fully leveraging information from a variety of sensory modalities is essential for the comprehensive understanding and processing of human emotions. Inspired by the process through which the human brain handles emotions and the theory of cross-modal plasticity, we propose UMBEnet, a brain-like unified modal affective processing network. The primary design of UMBEnet includes a Dual-Stream (DS) structure that fuses inherent prompts with a Prompt Pool and a Sparse Feature Fusion (SFF) module. The design of the Prompt Pool is aimed at integrating information from different modalities, while inherent prompts are intended to enhance the system's predictive guidance capabilities and effectively manage knowledge related to emotion classification. Moreover, considering the sparsity of effective information across different modalities, the SSF module aims to make full use of all available sensory data through the sparse integration of modality fusion prompts and inherent prompts, maintaining high adaptability and sensitivity to complex emotional states. Extensive experiments on the largest benchmark datasets in the Dynamic Facial Expression Recognition (DFER) field, including DFEW, FERV39k, and MAFW, have proven that UMBEnet consistently outperforms the current state-of-the-art methods. Notably, in scenarios of Modality Missingness and multimodal contexts, UMBEnet significantly surpasses the leading current methods, demonstrating outstanding performance and adaptability in tasks that involve complex emotional understanding with rich multimodal information.
Authors: Luca Di Giammarino, Boyang Sun, Giorgio Grisetti, Marc Pollefeys, Hermann Blum, Daniel Barath
Abstract: Accurate localization in diverse environments is a fundamental challenge in computer vision and robotics. The task involves determining a sensor's precise position and orientation, typically a camera, within a given space. Traditional localization methods often rely on passive sensing, which may struggle in scenarios with limited features or dynamic environments. In response, this paper explores the domain of active localization, emphasizing the importance of viewpoint selection to enhance localization accuracy. Our contributions involve using a data-driven approach with a simple architecture designed for real-time operation, a self-supervised data training method, and the capability to consistently integrate our map into a planning framework tailored for real-world robotics applications. Our results demonstrate that our method performs better than the existing one, targeting similar problems and generalizing on synthetic and real data. We also release an open-source implementation to benefit the community.
Authors: Thinesh Thiyakesan Ponbagavathi, Kunyu Peng, Alina Roitberg
Abstract: Foundation models (FMs) are large neural networks trained on broad datasets, excelling in downstream tasks with minimal fine-tuning. Human activity recognition in video has advanced with FMs, driven by competition among different architectures. However, high accuracies on standard benchmarks can draw an artificially rosy picture, as they often overlook real-world factors like changing camera perspectives. Popular benchmarks, mostly from YouTube or movies, offer diverse views but only coarse actions, which are insufficient for use-cases needing fine-grained, domain-specific actions. Domain-specific datasets (e.g., for industrial assembly) typically use data from limited static perspectives. This paper empirically evaluates how perspective changes affect different FMs in fine-grained human activity recognition. We compare multiple backbone architectures and design choices, including image- and video- based models, and various strategies for temporal information fusion, including commonly used score averaging and more novel attention-based temporal aggregation mechanisms. This is the first systematic study of different foundation models and specific design choices for human activity recognition from unknown views, conducted with the goal to provide guidance for backbone- and temporal- fusion scheme selection. Code and models will be made publicly available to the community.
Authors: Xiangyan Qu, Jing Yu, Keke Gai, Jiamin Zhuang, Yuanmin Tang, Gang Xiong, Gaopeng Gou, Qi Wu
Abstract: Recent work shows that documents from encyclopedias serve as helpful auxiliary information for zero-shot learning. Existing methods align the entire semantics of a document with corresponding images to transfer knowledge. However, they disregard that semantic information is not equivalent between them, resulting in a suboptimal alignment. In this work, we propose a novel network to extract multi-view semantic concepts from documents and images and align the matching rather than entire concepts. Specifically, we propose a semantic decomposition module to generate multi-view semantic embeddings from visual and textual sides, providing the basic concepts for partial alignment. To alleviate the issue of information redundancy among embeddings, we propose the local-to-semantic variance loss to capture distinct local details and multiple semantic diversity loss to enforce orthogonality among embeddings. Subsequently, two losses are introduced to partially align visual-semantic embedding pairs according to their semantic relevance at the view and word-to-patch levels. Consequently, we consistently outperform state-of-the-art methods under two document sources in three standard benchmarks for document-based zero-shot learning. Qualitatively, we show that our model learns the interpretable partial association.
Authors: Hanwei Liu, Rudong An, Zhimeng Zhang, Bowen Ma, Wei Zhang, Yan Song, Yujing Hu, Wei Chen, Yu Ding
Abstract: Facial Expression Analysis remains a challenging task due to unexpected task-irrelevant noise, such as identity, head pose, and background. To address this issue, this paper proposes a novel framework, called Norface, that is unified for both Action Unit (AU) analysis and Facial Emotion Recognition (FER) tasks. Norface consists of a normalization network and a classification network. First, the carefully designed normalization network struggles to directly remove the above task-irrelevant noise, by maintaining facial expression consistency but normalizing all original images to a common identity with consistent pose, and background. Then, these additional normalized images are fed into the classification network. Due to consistent identity and other factors (e.g. head pose, background, etc.), the normalized images enable the classification network to extract useful expression information more effectively. Additionally, the classification network incorporates a Mixture of Experts to refine the latent representation, including handling the input of facial representations and the output of multiple (AU or emotion) labels. Extensive experiments validate the carefully designed framework with the insight of identity normalization. The proposed method outperforms existing SOTA methods in multiple facial expression analysis tasks, including AU detection, AU intensity estimation, and FER tasks, as well as their cross-dataset tasks. For the normalized datasets and code please visit {https://norface-fea.github.io/}.
Authors: Nico Messikommer, Giovanni Cioffi, Mathias Gehrig, Davide Scaramuzza
Abstract: Visual Odometry (VO) is essential to downstream mobile robotics and augmented/virtual reality tasks. Despite recent advances, existing VO methods still rely on heuristic design choices that require several weeks of hyperparameter tuning by human experts, hindering generalizability and robustness. We address these challenges by reframing VO as a sequential decision-making task and applying Reinforcement Learning (RL) to adapt the VO process dynamically. Our approach introduces a neural network, operating as an agent within the VO pipeline, to make decisions such as keyframe and grid-size selection based on real-time conditions. Our method minimizes reliance on heuristic choices using a reward function based on pose error, runtime, and other metrics to guide the system. Our RL framework treats the VO system and the image sequence as an environment, with the agent receiving observations from keypoints, map statistics, and prior poses. Experimental results using classical VO methods and public benchmarks demonstrate improvements in accuracy and robustness, validating the generalizability of our RL-enhanced VO approach to different scenarios. We believe this paradigm shift advances VO technology by eliminating the need for time-intensive parameter tuning of heuristics.
Authors: Xin Ma, Yaohui Wang, Gengyu Jia, Xinyuan Chen, Yuan-Fang Li, Cunjian Chen, Yu Qiao
Abstract: Diffusion models have achieved great progress in image animation due to powerful generative capabilities. However, maintaining spatio-temporal consistency with detailed information from the input static image over time (e.g., style, background, and object of the input static image) and ensuring smoothness in animated video narratives guided by textual prompts still remains challenging. In this paper, we introduce Cinemo, a novel image animation approach towards achieving better motion controllability, as well as stronger temporal consistency and smoothness. In general, we propose three effective strategies at the training and inference stages of Cinemo to accomplish our goal. At the training stage, Cinemo focuses on learning the distribution of motion residuals, rather than directly predicting subsequent via a motion diffusion model. Additionally, a structural similarity index-based strategy is proposed to enable Cinemo to have better controllability of motion intensity. At the inference stage, a noise refinement technique based on discrete cosine transformation is introduced to mitigate sudden motion changes. Such three strategies enable Cinemo to produce highly consistent, smooth, and motion-controllable results. Compared to previous methods, Cinemo offers simpler and more precise user controllability. Extensive experiments against several state-of-the-art methods, including both commercial tools and research approaches, across multiple metrics, demonstrate the effectiveness and superiority of our proposed approach.
Authors: Daniel Jakab, Alexander Braun, Cathaoir Agnew, Reenu Mohandas, Brian Michael Deegan, Dara Molloy, Enda Ward, Tony Scanlan, Ciar\'an Eising
Abstract: Automotive simulation can potentially compensate for a lack of training data in computer vision applications. However, there has been little to no image quality evaluation of automotive simulation and the impact of optical degradations on simulation is little explored. In this work, we investigate Virtual KITTI and the impact of applying variations of Gaussian blur on image sharpness. Furthermore, we consider object detection, a common computer vision application on three different state-of-the-art models, thus allowing us to characterize the relationship between object detection and sharpness. It was found that while image sharpness (MTF50) degrades from an average of 0.245cy/px to approximately 0.119cy/px; object detection performance stays largely robust within 0.58\%(Faster RCNN), 1.45\%(YOLOF) and 1.93\%(DETR) across all respective held-out test sets.
Authors: Mengqi Guo, Chen Li, Yuyang Zhao, Gim Hee Lee
Abstract: Inferring step-wise actions to assemble 3D objects with primitive bricks from images is a challenging task due to complex constraints and the vast number of possible combinations. Recent studies have demonstrated promising results on sequential LEGO brick assembly through the utilization of LEGO-Graph modeling to predict sequential actions. However, existing approaches are class-specific and require significant computational and 3D annotation resources. In this work, we first propose a computationally efficient breadth-first search (BFS) LEGO-Tree structure to model the sequential assembly actions by considering connections between consecutive layers. Based on the LEGO-Tree structure, we then design a class-agnostic tree-transformer framework to predict the sequential assembly actions from the input multi-view images. A major challenge of the sequential brick assembly task is that the step-wise action labels are costly and tedious to obtain in practice. We mitigate this problem by leveraging synthetic-to-real transfer learning. Specifically, our model is first pre-trained on synthetic data with full supervision from the available action labels. We then circumvent the requirement for action labels in the real data by proposing an action-to-silhouette projection that replaces action labels with input image silhouettes for self-supervision. Without any annotation on the real data, our model outperforms existing methods with 3D supervision by 7.8% and 11.3% in mIoU on the MNIST and ModelNet Construction datasets, respectively.
Authors: Jiahang Tu, Wei Ji, Hanbin Zhao, Chao Zhang, Roger Zimmermann, Hui Qian
Abstract: In autonomous driving, deep models have shown remarkable performance across various visual perception tasks with the demand of high-quality and huge-diversity training datasets. Such datasets are expected to cover various driving scenarios with adverse weather, lighting conditions and diverse moving objects. However, manually collecting these data presents huge challenges and expensive cost. With the rapid development of large generative models, we propose DriveDiTFit, a novel method for efficiently generating autonomous Driving data by Fine-tuning pre-trained Diffusion Transformers (DiTs). Specifically, DriveDiTFit utilizes a gap-driven modulation technique to carefully select and efficiently fine-tune a few parameters in DiTs according to the discrepancy between the pre-trained source data and the target driving data. Additionally, DriveDiTFit develops an effective weather and lighting condition embedding module to ensure diversity in the generated data, which is initialized by a nearest-semantic-similarity initialization approach. Through progressive tuning scheme to refined the process of detail generation in early diffusion process and enlarging the weights corresponding to small objects in training loss, DriveDiTFit ensures high-quality generation of small moving objects in the generated data. Extensive experiments conducted on driving datasets confirm that our method could efficiently produce diverse real driving data. The source codes will be available at https://github.com/TtuHamg/DriveDiTFit.
Authors: Alexander Melekhin, Dmitry Yudin, Ilia Petryashin, Vitaly Bezuglyj
Abstract: Place recognition is a challenging task in computer vision, crucial for enabling autonomous vehicles and robots to navigate previously visited environments. While significant progress has been made in learnable multimodal methods that combine onboard camera images and LiDAR point clouds, the full potential of these methods remains largely unexplored in localization applications. In this paper, we study the impact of leveraging a multi-camera setup and integrating diverse data sources for multimodal place recognition, incorporating explicit visual semantics and text descriptions. Our proposed method named MSSPlace utilizes images from multiple cameras, LiDAR point clouds, semantic segmentation masks, and text annotations to generate comprehensive place descriptors. We employ a late fusion approach to integrate these modalities, providing a unified representation. Through extensive experiments on the Oxford RobotCar and NCLT datasets, we systematically analyze the impact of each data source on the overall quality of place descriptors. Our experiments demonstrate that combining data from multiple sensors significantly improves place recognition model performance compared to single modality approaches and leads to state-of-the-art quality. We also show that separate usage of visual or textual semantics (which are more compact representations of sensory data) can achieve promising results in place recognition. The code for our method is publicly available: https://github.com/alexmelekhin/MSSPlace
Authors: Gon\c{c}alo Vinagre Martins, Afonso Quinaz, Carla Viegas, Sofia Cavaco, Jo\~ao Magalh\~aes
Abstract: Sign Language Recognition has been studied and developed throughout the years to help the deaf and hard-of-hearing people in their day-to-day lives. These technologies leverage manual sign recognition algorithms, however, most of them lack the recognition of facial expressions, which are also an essential part of Sign Language as they allow the speaker to add expressiveness to their dialogue or even change the meaning of certain manual signs. SLVideo is a video moment retrieval software for Sign Language videos with a focus on both hands and facial signs. The system extracts embedding representations for the hand and face signs from video frames to capture the language signs in full. This will then allow the user to search for a specific sign language video segment with text queries, or to search by similar sign language videos. To test this system, a collection of five hours of annotated Sign Language videos is used as the dataset, and the initial results are promising in a zero-shot setting.SLVideo is shown to not only address the problem of searching sign language videos but also supports a Sign Language thesaurus with a search by similarity technique. Project web page: https://novasearch.github.io/SLVideo/
Authors: Rabbia Asghar, Wenqian Liu, Lukas Rummelhard, Anne Spalanzani, Christian Laugier
Abstract: Accurate prediction of driving scenes is essential for road safety and autonomous driving. Occupancy Grid Maps (OGMs) are commonly employed for scene prediction due to their structured spatial representation, flexibility across sensor modalities and integration of uncertainty. Recent studies have successfully combined OGMs with deep learning methods to predict the evolution of scene and learn complex behaviours. These methods, however, do not consider prediction of flow or velocity vectors in the scene. In this work, we propose a novel multi-task framework that leverages dynamic OGMs and semantic information to predict both future vehicle semantic grids and the future flow of the scene. This incorporation of semantic flow not only offers intermediate scene features but also enables the generation of warped semantic grids. Evaluation on the real-world NuScenes dataset demonstrates improved prediction capabilities and enhanced ability of the model to retain dynamic vehicles within the scene.
Authors: Zhecan Wang, Garrett Bingham, Adams Yu, Quoc Le, Thang Luong, Golnaz Ghiasi
Abstract: Hallucination has been a major problem for large language models and remains a critical challenge when it comes to multimodality in which vision-language models (VLMs) have to deal with not just textual but also visual inputs. Despite rapid progress in VLMs, resources for evaluating and addressing multimodal hallucination are limited and mostly focused on evaluation. This work introduces HaloQuest, a novel visual question answering dataset that captures various aspects of multimodal hallucination such as false premises, insufficient contexts, and visual challenges. A novel idea from HaloQuest is to leverage synthetic images, apart from real ones, to enable dataset creation at scale. With over 7.7K examples spanning across a wide variety of categories, HaloQuest was designed to be both a challenging benchmark for VLMs and a fine-tuning dataset for advancing multimodal reasoning. Our experiments reveal that current models struggle with HaloQuest, with all open-source VLMs achieving below 36% accuracy. On the other hand, fine-tuning on HaloQuest significantly reduces hallucination rates while preserving performance on standard reasoning tasks. Our results discover that benchmarking with generated images is highly correlated (r=0.97) with real images. Last but not least, we propose a novel Auto-Eval mechanism that is highly correlated with human raters (r=0.99) for evaluating VLMs. In sum, this work makes concrete strides towards understanding, evaluating, and mitigating hallucination in VLMs, serving as an important step towards more reliable multimodal AI systems in the future.
Authors: Bowen Peng, Li Liu, Tianpeng Liu, Zhen Liu, Yongxiang Liu
Abstract: Transfer-based targeted adversarial attacks against black-box deep neural networks (DNNs) have been proven to be significantly more challenging than untargeted ones. The impressive transferability of current SOTA, the generative methods, comes at the cost of requiring massive amounts of additional data and time-consuming training for each targeted label. This results in limited efficiency and flexibility, significantly hindering their deployment in practical applications. In this paper, we offer a self-universal perspective that unveils the great yet underexplored potential of input transformations in pursuing this goal. Specifically, transformations universalize gradient-based attacks with intrinsic but overlooked semantics inherent within individual images, exhibiting similar scalability and comparable results to time-consuming learning over massive additional data from diverse classes. We also contribute a surprising empirical insight that one of the most fundamental transformations, simple image scaling, is highly effective, scalable, sufficient, and necessary in enhancing targeted transferability. We further augment simple scaling with orthogonal transformations and block-wise applicability, resulting in the Simple, faSt, Self-universal yet Strong Scale Transformation (S$^4$ST) for self-universal TTA. On the ImageNet-Compatible benchmark dataset, our method achieves a 19.8% improvement in the average targeted transfer success rate against various challenging victim models over existing SOTA transformation methods while only consuming 36% time for attacking. It also outperforms resource-intensive attacks by a large margin in various challenging settings.
Authors: Jinfu Liu, Chen Chen, Mengyuan Liu
Abstract: Skeleton-based action recognition has garnered significant attention due to the utilization of concise and resilient skeletons. Nevertheless, the absence of detailed body information in skeletons restricts performance, while other multimodal methods require substantial inference resources and are inefficient when using multimodal data during both training and inference stages. To address this and fully harness the complementary multimodal features, we propose a novel multi-modality co-learning (MMCL) framework by leveraging the multimodal large language models (LLMs) as auxiliary networks for efficient skeleton-based action recognition, which engages in multi-modality co-learning during the training stage and keeps efficiency by employing only concise skeletons in inference. Our MMCL framework primarily consists of two modules. First, the Feature Alignment Module (FAM) extracts rich RGB features from video frames and aligns them with global skeleton features via contrastive learning. Second, the Feature Refinement Module (FRM) uses RGB images with temporal information and text instruction to generate instructive features based on the powerful generalization of multimodal LLMs. These instructive text features will further refine the classification scores and the refined scores will enhance the model's robustness and generalization in a manner similar to soft labels. Extensive experiments on NTU RGB+D, NTU RGB+D 120 and Northwestern-UCLA benchmarks consistently verify the effectiveness of our MMCL, which outperforms the existing skeleton-based action recognition methods. Meanwhile, experiments on UTD-MHAD and SYSU-Action datasets demonstrate the commendable generalization of our MMCL in zero-shot and domain-adaptive action recognition. Our code is publicly available at: https://github.com/liujf69/MMCL-Action.
Authors: Basit Alawode, Sajid Javed, Arif Mahmood, Jiri Matas
Abstract: We observe that the performance of SOTA visual trackers surprisingly strongly varies across different video attributes and datasets. No single tracker remains the best performer across all tracking attributes and datasets. To bridge this gap, for a given video sequence, we predict the "Best of the N Trackers", called the BofN meta-tracker. At its core, a Tracking Performance Prediction Network (TP2N) selects a predicted best performing visual tracker for the given video sequence using only a few initial frames. We also introduce a frame-level BofN meta-tracker which keeps predicting best performer after regular temporal intervals. The TP2N is based on self-supervised learning architectures MocoV2, SwAv, BT, and DINO; experiments show that the DINO with ViT-S as a backbone performs the best. The video-level BofN meta-tracker outperforms, by a large margin, existing SOTA trackers on nine standard benchmarks - LaSOT, TrackingNet, GOT-10K, VOT2019, VOT2021, VOT2022, UAV123, OTB100, and WebUAV-3M. Further improvement is achieved by the frame-level BofN meta-tracker effectively handling variations in the tracking scenarios within long sequences. For instance, on GOT-10k, BofN meta-tracker average overlap is 88.7% and 91.1% with video and frame-level settings respectively. The best performing tracker, RTS, achieves 85.20% AO. On VOT2022, BofN expected average overlap is 67.88% and 70.98% with video and frame level settings, compared to the best performing ARTrack, 64.12%. This work also presents an extensive evaluation of competitive tracking methods on all commonly used benchmarks, following their protocols. The code, the trained models, and the results will soon be made publicly available on https://github.com/BasitAlawode/Best_of_N_Trackers.
Authors: Liangyan Jiang, Chuang Zhu, Yanxu Chen
Abstract: The spike camera, with its high temporal resolution, low latency, and high dynamic range, addresses high-speed imaging challenges like motion blur. It captures photons at each pixel independently, creating binary spike streams rich in temporal information but challenging for image reconstruction. Current algorithms, both traditional and deep learning-based, still need to be improved in the utilization of the rich temporal detail and the restoration of the details of the reconstructed image. To overcome this, we introduce Swin Spikeformer (SwinSF), a novel model for dynamic scene reconstruction from spike streams. SwinSF is composed of Spike Feature Extraction, Spatial-Temporal Feature Extraction, and Final Reconstruction Module. It combines shifted window self-attention and proposed temporal spike attention, ensuring a comprehensive feature extraction that encapsulates both spatial and temporal dynamics, leading to a more robust and accurate reconstruction of spike streams. Furthermore, we build a new synthesized dataset for spike image reconstruction which matches the resolution of the latest spike camera, ensuring its relevance and applicability to the latest developments in spike camera imaging. Experimental results demonstrate that the proposed network SwinSF sets a new benchmark, achieving state-of-the-art performance across a series of datasets, including both real-world and synthesized data across various resolutions. Our codes and proposed dataset will be available soon.
Authors: Zhili He, Yu-Hsing Wang
Abstract: Cracks pose safety risks to infrastructure and cannot be overlooked. The prevailing structures in existing crack segmentation networks predominantly consist of CNNs or Transformers. However, CNNs exhibit a deficiency in global modeling capability, hindering the representation to entire crack features. Transformers can capture long-range dependencies but suffer from high and quadratic complexity. Recently, Mamba has garnered extensive attention due to its linear spatial and computational complexity and its powerful global perception. This study explores the representation capabilities of Mamba to crack features. Specifically, this paper uncovers the connection between Mamba and the attention mechanism, providing a profound insight, an attention perspective, into interpreting Mamba and devising a novel Mamba module following the principles of attention blocks, namely CrackMamba. We compare CrackMamba with the most prominent visual Mamba modules, Vim and Vmamba, on two datasets comprising asphalt pavement and concrete pavement cracks, and steel cracks, respectively. The quantitative results show that CrackMamba stands out as the sole Mamba block consistently enhancing the baseline model's performance across all evaluation measures, while reducing its parameters and computational costs. Moreover, this paper substantiates that Mamba can achieve global receptive fields through both theoretical analysis and visual interpretability. The discoveries of this study offer a dual contribution. First, as a plug-and-play and simple yet effective Mamba module, CrackMamba exhibits immense potential for integration into various crack segmentation models. Second, the proposed innovative Mamba design concept, integrating Mamba with the attention mechanism, holds significant reference value for all Mamba-based computer vision models, not limited to crack segmentation networks, as investigated in this study.
Authors: Farzad Beizaee, Gregory A. Lodygensky, Chris L. Adamson, Deanne K. Thompso, Jeanie L. Y. Cheon, Alicia J. Spittl. Peter J. Anderso, Christian Desrosier, Jose Dolz
Abstract: Lack of standardization and various intrinsic parameters for magnetic resonance (MR) image acquisition results in heterogeneous images across different sites and devices, which adversely affects the generalization of deep neural networks. To alleviate this issue, this work proposes a novel unsupervised harmonization framework that leverages normalizing flows to align MR images, thereby emulating the distribution of a source domain. The proposed strategy comprises three key steps. Initially, a normalizing flow network is trained to capture the distribution characteristics of the source domain. Then, we train a shallow harmonizer network to reconstruct images from the source domain via their augmented counterparts. Finally, during inference, the harmonizer network is updated to ensure that the output images conform to the learned source domain distribution, as modeled by the normalizing flow network. Our approach, which is unsupervised, source-free, and task-agnostic is assessed in the context of both adults and neonatal cross-domain brain MRI segmentation, as well as neonatal brain age estimation, demonstrating its generalizability across tasks and population demographics. The results underscore its superior performance compared to existing methodologies. The code is available at https://github.com/farzad-bz/Harmonizing-Flows
Authors: Zhaojie Fang, Shenghao Zhu, Yifei Chen, Binfeng Zou, Fan Jia, Linwei Qiu, Chang Liu, Yiyu Huang, Xiang Feng, Feiwei Qin, Changmiao Wang, Yeru Wang, Jin Fan, Changbiao Chu, Wan-Zhen Wu, Hu Zhao
Abstract: Alzheimer's Disease (AD) is an irreversible neurodegenerative disorder that often progresses from Mild Cognitive Impairment (MCI), leading to memory loss and significantly impacting patients' lives. Clinical trials indicate that early targeted interventions for MCI patients can potentially slow or halt the development and progression of AD. Previous research has shown that accurate medical classification requires the inclusion of extensive multimodal data, such as assessment scales and various neuroimaging techniques like Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET). However, consistently tracking the diagnosis of the same individual over time and simultaneously collecting multimodal data poses significant challenges. To address this issue, we introduce GFE-Mamba, a classifier based on Generative Feature Extraction (GFE). This classifier effectively integrates data from assessment scales, MRI, and PET, enabling deeper multimodal fusion. It efficiently extracts both long and short sequence information and incorporates additional information beyond the pixel space. This approach not only improves classification accuracy but also enhances the interpretability and stability of the model. We constructed datasets of over 3000 samples based on the Alzheimer's Disease Neuroimaging Initiative (ADNI) for a two-step training process. Our experimental results demonstrate that the GFE-Mamba model is effective in predicting the conversion from MCI to AD and outperforms several state-of-the-art methods. Our source code and ADNI dataset processing code are available at https://github.com/Tinysqua/GFE-Mamba.
Authors: Josiah Couch, Ramy Arnaout, Rima Arnaout
Abstract: In deep learning, achieving high performance on image classification tasks requires diverse training sets. However, dataset diversity is incompletely understood. The current best practice is to try to maximize dataset size and class balance. Yet large, class-balanced datasets are not guaranteed to be diverse: images can still be arbitrarily similar. We hypothesized that, for a given model architecture, better model performance can be achieved by maximizing dataset diversity more directly. This could open a path for performance improvement without additional computational resources or architectural advances. To test this hypothesis, we introduce a comprehensive framework of diversity measures, developed in ecology, that generalizes familiar quantities like Shannon entropy by accounting for similarities and differences among images. (Dataset size and class balance emerge from this framework as special cases.) By analyzing thousands of subsets from seven medical datasets representing ultrasound, X-ray, CT, and pathology images, we found that the best correlates of performance were not size or class balance but $A$ -- ``big alpha'' -- a set of generalized entropy measures interpreted as the effective number of image-class pairs in the dataset, after accounting for similarities among images. One of these, $A_0$, explained 67\% of the variance in balanced accuracy across all subsets, vs. 54\% for class balance and just 39\% for size. The best pair was size and $A_1$ (79\%), which outperformed size and class balance (74\%). $A$ performed best for subsets from individual datasets as well as across datasets, supporting the generality of these results. We propose maximizing $A$ as a potential new way to improve the performance of deep learning in medical imaging.
Authors: Laura Niss, Kevin Vogt-Lowell, Theodoros Tsiligkaridis
Abstract: Despite the proliferation of large vision-language foundation models, estimation of the learning and forgetting outcomes following fine-tuning of these models remains largely unexplored. Inspired by work highlighting the significance of the modality gap in contrastive dual-encoders, we propose the Inter-Intra Modal Measure (IIMM). Combining terms quantifying the similarity between image embeddings and the similarity between incorrect image and label embedding pairs, the IIMM functions as a strong predictor of performance changes with fine-tuning. Our extensive empirical analysis across four state-of-the-art vision-language models (CLIP, SigLIP, CoCa, EVA-02-CLIP) and five fine-tuning techniques (full fine-tuning, BitFit, attention-weight tuning, LoRA, CLIP-Adapter) demonstrates a strong, statistically significant linear relationship: fine-tuning on tasks with higher IIMM scores produces greater in-domain performance gains but also induces more severe out-of-domain performance degradation, with some parameter-efficient fine-tuning (PEFT) methods showing extreme forgetting. We compare our measure against transfer scores from state-of-the-art model selection methods and show that the IIMM is significantly more predictive of accuracy gains. With only a single forward pass of the target data, practitioners can leverage this key insight to heuristically evaluate the degree to which a model can be expected to improve following fine-tuning. Given additional knowledge about the model's performance on a few diverse tasks, this heuristic further evolves into a strong predictor of expected performance changes when training for new tasks.
Authors: Silvio Galesso, Philipp Schr\"oppel, Hssan Driss, Thomas Brox
Abstract: In recent years, research on out-of-distribution (OoD) detection for semantic segmentation has mainly focused on road scenes -- a domain with a constrained amount of semantic diversity. In this work, we challenge this constraint and extend the domain of this task to general natural images. To this end, we introduce: 1. the ADE-OoD benchmark, which is based on the ADE20k dataset and includes images from diverse domains with a high semantic diversity, and 2. a novel approach that uses Diffusion score matching for OoD detection (DOoD) and is robust to the increased semantic diversity. ADE-OoD features indoor and outdoor images, defines 150 semantic categories as in-distribution, and contains a variety of OoD objects. For DOoD, we train a diffusion model with an MLP architecture on semantic in-distribution embeddings and build on the score matching interpretation to compute pixel-wise OoD scores at inference time. On common road scene OoD benchmarks, DOoD performs on par or better than the state of the art, without using outliers for training or making assumptions about the data domain. On ADE-OoD, DOoD outperforms previous approaches, but leaves much room for future improvements.
Authors: Haoning Wu, Dongxu Li, Bei Chen, Junnan Li
Abstract: Large multimodal models (LMMs) are processing increasingly longer and richer inputs. Albeit the progress, few public benchmark is available to measure such development. To mitigate this gap, we introduce LongVideoBench, a question-answering benchmark that features video-language interleaved inputs up to an hour long. Our benchmark includes 3,763 varying-length web-collected videos with their subtitles across diverse themes, designed to comprehensively evaluate LMMs on long-term multimodal understanding. To achieve this, we interpret the primary challenge as to accurately retrieve and reason over detailed multimodal information from long inputs. As such, we formulate a novel video question-answering task termed referring reasoning. Specifically, as part of the question, it contains a referring query that references related video contexts, called referred context. The model is then required to reason over relevant video details from the referred context. Following the paradigm of referring reasoning, we curate 6,678 human-annotated multiple-choice questions in 17 fine-grained categories, establishing one of the most comprehensive benchmarks for long-form video understanding. Evaluations suggest that the LongVideoBench presents significant challenges even for the most advanced proprietary models (e.g. GPT-4o, Gemini-1.5-Pro, GPT-4-Turbo), while their open-source counterparts show an even larger performance gap. In addition, our results indicate that model performance on the benchmark improves only when they are capable of processing more frames, positioning LongVideoBench as a valuable benchmark for evaluating future-generation long-context LMMs.
Authors: Brian K. S. Isaac-Medina, Yona Falinie A. Gaus, Neelanjan Bhowmik, Toby P. Breckon
Abstract: Object detection is a pivotal task in computer vision that has received significant attention in previous years. Nonetheless, the capability of a detector to localise objects out of the training distribution remains unexplored. Whilst recent approaches in object-level out-of-distribution (OoD) detection heavily rely on class labels, such approaches contradict truly open-world scenarios where the class distribution is often unknown. In this context, anomaly detection focuses on detecting unseen instances rather than classifying detections as OoD. This work aims to bridge this gap by leveraging an open-world object detector and an OoD detector via virtual outlier synthesis. This is achieved by using the detector backbone features to first learn object pseudo-classes via self-supervision. These pseudo-classes serve as the basis for class-conditional virtual outlier sampling of anomalous features that are classified by an OoD head. Our approach empowers our overall object detector architecture to learn anomaly-aware feature representations without relying on class labels, hence enabling truly open-world object anomaly detection. Empirical validation of our approach demonstrates its effectiveness across diverse datasets encompassing various imaging modalities (visible, infrared, and X-ray). Moreover, our method establishes state-of-the-art performance on object-level anomaly detection, achieving an average recall score improvement of over 5.4% for natural images and 23.5% for a security X-ray dataset compared to the current approaches. In addition, our method detects anomalies in datasets where current approaches fail. Code available at https://github.com/KostadinovShalon/oln-ssos.
Authors: Yike Zhang, Dingjie Su, Eduardo Davalos, Jack H. Noble
Abstract: Cochlear Implant (CI) procedures involve inserting an array of electrodes into the cochlea located inside the inner ear. Mastoidectomy is a surgical procedure that uses a high-speed drill to remove part of the mastoid region of the temporal bone, providing safe access to the cochlea through the middle and inner ear. We aim to develop an intraoperative navigation system that registers plans created using 3D preoperative Computerized Tomography (CT) volumes with the 2D surgical microscope view. Herein, we propose a method to synthesize the mastoidectomy volume using only the preoperative CT scan, where the mastoid is intact. We introduce an unsupervised learning framework designed to synthesize mastoidectomy. For model training purposes, this method uses postoperative CT scans to avoid manual data cleaning or labeling, even when the region removed during mastoidectomy is visible but affected by metal artifacts, low signal-to-noise ratio, or electrode wiring. Our approach estimates mastoidectomy regions with a mean dice score of 70.0%. This approach represents a major step forward for CI intraoperative navigation by predicting realistic mastoidectomy-removed regions in preoperative planning that can be used to register the pre-surgery plan to intraoperative microscopy.
Authors: Jingtai He, Gehao Zhang, Tingting Liu, Songlin Du
Abstract: Recent advancements in keypoint detection and descriptor extraction have shown impressive performance in local feature learning tasks. However, existing methods generally exhibit suboptimal performance under extreme conditions such as significant appearance changes and domain shifts. In this study, we introduce a multi-level feature aggregation network that incorporates two pivotal components to facilitate the learning of robust and accurate features with domain adaptation. First, we employ domain adaptation supervision to align high-level feature distributions across different domains to achieve invariant domain representations. Second, we propose a Transformer-based booster that enhances descriptor robustness by integrating visual and geometric information through wave position encoding concepts, effectively handling complex conditions. To ensure the accuracy and robustness of features, we adopt a hierarchical architecture to capture comprehensive information and apply meticulous targeted supervision to keypoint detection, descriptor extraction, and their coupled processing. Extensive experiments demonstrate that our method, RADA, achieves excellent results in image matching, camera pose estimation, and visual localization tasks.
Authors: Emanuele Frascaroli, Aniello Panariello, Pietro Buzzega, Lorenzo Bonicelli, Angelo Porrello, Simone Calderara
Abstract: With the emergence of Transformers and Vision-Language Models (VLMs) such as CLIP, large pre-trained models have become a common strategy to enhance performance in Continual Learning scenarios. This led to the development of numerous prompting strategies to effectively fine-tune transformer-based models without succumbing to catastrophic forgetting. However, these methods struggle to specialize the model on domains significantly deviating from the pre-training and preserving its zero-shot capabilities. In this work, we propose Continual Generative training for Incremental prompt-Learning, a novel approach to mitigate forgetting while adapting a VLM, which exploits generative replay to align prompts to tasks. We also introduce a new metric to evaluate zero-shot capabilities within CL benchmarks. Through extensive experiments on different domains, we demonstrate the effectiveness of our framework in adapting to new tasks while improving zero-shot capabilities. Further analysis reveals that our approach can bridge the gap with joint prompt tuning. The codebase is available at https://github.com/aimagelab/mammoth.
Authors: Guiqiu Liao, Matjaz Jogan, Sai Koushik, Eric Eaton, Daniel A. Hashimoto
Abstract: Weakly supervised video object segmentation (WSVOS) enables the identification of segmentation maps without requiring an extensive training dataset of object masks, relying instead on coarse video labels indicating object presence. Current state-of-the-art methods either require multiple independent stages of processing that employ motion cues or, in the case of end-to-end trainable networks, lack in segmentation accuracy, in part due to the difficulty of learning segmentation maps from videos with transient object presence. This limits the application of WSVOS for semantic annotation of surgical videos where multiple surgical tools frequently move in and out of the field of view, a problem that is more difficult than typically encountered in WSVOS. This paper introduces Video Spatio-Temporal Disentanglement Networks (VDST-Net), a framework to disentangle spatiotemporal information using semi-decoupled knowledge distillation to predict high-quality class activation maps (CAMs). A teacher network designed to resolve temporal conflicts when specifics about object location and timing in the video are not provided works with a student network that integrates information over time by leveraging temporal dependencies. We demonstrate the efficacy of our framework on a public reference dataset and on a more challenging surgical video dataset where objects are, on average, present in less than 60\% of annotated frames. Our method outperforms state-of-the-art techniques and generates superior segmentation masks under video-level weak supervision.
Authors: Yunkang Cao, Jiangning Zhang, Luca Frittoli, Yuqi Cheng, Weiming Shen, Giacomo Boracchi
Abstract: Zero-shot anomaly detection (ZSAD) targets the identification of anomalies within images from arbitrary novel categories. This study introduces AdaCLIP for the ZSAD task, leveraging a pre-trained vision-language model (VLM), CLIP. AdaCLIP incorporates learnable prompts into CLIP and optimizes them through training on auxiliary annotated anomaly detection data. Two types of learnable prompts are proposed: static and dynamic. Static prompts are shared across all images, serving to preliminarily adapt CLIP for ZSAD. In contrast, dynamic prompts are generated for each test image, providing CLIP with dynamic adaptation capabilities. The combination of static and dynamic prompts is referred to as hybrid prompts, and yields enhanced ZSAD performance. Extensive experiments conducted across 14 real-world anomaly detection datasets from industrial and medical domains indicate that AdaCLIP outperforms other ZSAD methods and can generalize better to different categories and even domains. Finally, our analysis highlights the importance of diverse auxiliary data and optimized prompts for enhanced generalization capacity. Code is available at https://github.com/caoyunkang/AdaCLIP.
Authors: Nermin Samet, Gilles Puy, Oriane Sim\'eoni, Renaud Marlet
Abstract: Annotating lidar point clouds for autonomous driving is a notoriously expensive and time-consuming task. In this work, we show that the quality of recent self-supervised lidar scan representations allows a great reduction of the annotation cost. Our method has two main steps. First, we show that self-supervised representations allow a simple and direct selection of highly informative lidar scans to annotate: training a network on these selected scans leads to much better results than a random selection of scans and, more interestingly, to results on par with selections made by SOTA active learning methods. In a second step, we leverage the same self-supervised representations to cluster points in our selected scans. Asking the annotator to classify each cluster, with a single click per cluster, then permits us to close the gap with fully-annotated training sets, while only requiring one thousandth of the point labels.
Authors: Guanyu Hu, Jie Wei, Siyang Song, Dimitrios Kollias, Xinyu Yang, Zhonglin Sun, Odysseus Kaloidas
Abstract: The objective of the Multiple Appropriate Facial Reaction Generation (MAFRG) task is to produce contextually appropriate and diverse listener facial behavioural responses based on the multimodal behavioural data of the conversational partner (i.e., the speaker). Current methodologies typically assume continuous availability of speech and facial modality data, neglecting real-world scenarios where these data may be intermittently unavailable, which often results in model failures. Furthermore, despite utilising advanced deep learning models to extract information from the speaker's multimodal inputs, these models fail to adequately leverage the speaker's emotional context, which is vital for eliciting appropriate facial reactions from human listeners. To address these limitations, we propose an Emotion-aware Modality Compensatory (EMC) framework. This versatile solution can be seamlessly integrated into existing models, thereby preserving their advantages while significantly enhancing performance and robustness in scenarios with missing modalities. Our framework ensures resilience when faced with missing modality data through the Compensatory Modality Alignment (CMA) module. It also generates more appropriate emotion-aware reactions via the Emotion-aware Attention (EA) module, which incorporates the speaker's emotional information throughout the entire encoding and decoding process. Experimental results demonstrate that our framework improves the appropriateness metric FRCorr by an average of 57.2\% compared to the original model structure. In scenarios where speech modality data is missing, the performance of appropriate generation shows an improvement, and when facial data is missing, it only exhibits minimal degradation.
Authors: Manfred Georg, Garrett Tanzer, Saad Hassan, Maximus Shengelia, Esha Uboweja, Sam Sepah, Sean Forbes, Thad Starner
Abstract: Progress in machine understanding of sign languages has been slow and hampered by limited data. In this paper, we present FSboard, an American Sign Language fingerspelling dataset situated in a mobile text entry use case, collected from 147 paid and consenting Deaf signers using Pixel 4A selfie cameras in a variety of environments. Fingerspelling recognition is an incomplete solution that is only one small part of sign language translation, but it could provide some immediate benefit to Deaf/Hard of Hearing signers as more broadly capable technology develops. At >3 million characters in length and >250 hours in duration, FSboard is the largest fingerspelling recognition dataset to date by a factor of >10x. As a simple baseline, we finetune 30 Hz MediaPipe Holistic landmark inputs into ByT5-Small and achieve 11.1% Character Error Rate (CER) on a test set with unique phrases and signers. This quality degrades gracefully when decreasing frame rate and excluding face/body landmarks: plausible optimizations to help models run on device in real time.
Authors: Siddharth D Jaiswal, Animesh Ganai, Abhisek Dash, Saptarshi Ghosh, Animesh Mukherjee
Abstract: Facial Recognition Systems (FRSs) are being developed and deployed globally at unprecedented rates. Most platforms are designed in a limited set of countries but deployed in worldwide, without adequate checkpoints. This is especially problematic for Global South countries which lack strong legislation to safeguard persons facing disparate performance of these systems. A combination of unavailability of datasets, lack of understanding of FRS functionality and low-resource bias mitigation measures accentuate the problem. In this work, we propose a new face dataset composed of 6,579 unique male and female sportspersons from eight countries around the world. More than 50% of the dataset comprises individuals from the Global South countries and is demographically diverse. To aid adversarial audits and robust model training, each image has four adversarial variants, totaling over 40,000 images. We also benchmark five popular FRSs, both commercial and open-source, for the task of gender prediction (and country prediction for one of the open-source models as an example of red-teaming). Experiments on industrial FRSs reveal accuracies ranging from 98.2%--38.1%, with a large disparity between males and females in the Global South (max difference of 38.5%). Biases are also observed in all FRSs between females of the Global North and South (max difference of ~50%). Grad-CAM analysis identifies the nose, forehead and mouth as the regions of interest on one of the open-source FRSs. Utilizing this insight, we design simple, low-resource bias mitigation solutions using few-shot and novel contrastive learning techniques significantly improving the accuracy with disparity between males and females reducing from 50% to 1.5% in one of the settings. In the red-teaming experiment with the open-source Deepface model, contrastive learning proves more effective than simple fine-tuning.
Authors: Vikash Sehwag, Xianghao Kong, Jingtao Li, Michael Spranger, Lingjuan Lyu
Abstract: As scaling laws in generative AI push performance, they also simultaneously concentrate the development of these models among actors with large computational resources. With a focus on text-to-image (T2I) generative models, we aim to address this bottleneck by demonstrating very low-cost training of large-scale T2I diffusion transformer models. As the computational cost of transformers increases with the number of patches in each image, we propose to randomly mask up to 75% of the image patches during training. We propose a deferred masking strategy that preprocesses all patches using a patch-mixer before masking, thus significantly reducing the performance degradation with masking, making it superior to model downscaling in reducing computational cost. We also incorporate the latest improvements in transformer architecture, such as the use of mixture-of-experts layers, to improve performance and further identify the critical benefit of using synthetic images in micro-budget training. Finally, using only 37M publicly available real and synthetic images, we train a 1.16 billion parameter sparse transformer with only \$1,890 economical cost and achieve a 12.7 FID in zero-shot generation on the COCO dataset. Notably, our model achieves competitive FID and high-quality generations while incurring 118$\times$ lower cost than stable diffusion models and 14$\times$ lower cost than the current state-of-the-art approach that costs \$28,400. We aim to release our end-to-end training pipeline to further democratize the training of large-scale diffusion models on micro-budgets.
Authors: Kshitij Ingale, Sun Hae Hong, Qiyuan Hu, Renyu Zhang, Bo Osinski, Mina Khoshdeli, Josh Och, Kunal Nagpal, Martin C. Stumpe, Rohan P. Joshi
Abstract: Molecular testing of tumor samples for targetable biomarkers is restricted by a lack of standardization, turnaround-time, cost, and tissue availability across cancer types. Additionally, targetable alterations of low prevalence may not be tested in routine workflows. Algorithms that predict DNA alterations from routinely generated hematoxylin and eosin (H&E)-stained images could prioritize samples for confirmatory molecular testing. Costs and the necessity of a large number of samples containing mutations limit approaches that train individual algorithms for each alteration. In this work, models were trained for simultaneous prediction of multiple DNA alterations from H&E images using a multi-task approach. Compared to biomarker-specific models, this approach performed better on average, with pronounced gains for rare mutations. The models reasonably generalized to independent temporal-holdout, externally-stained, and multi-site TCGA test sets. Additionally, whole slide image embeddings derived using multi-task models demonstrated strong performance in downstream tasks that were not a part of training. Overall, this is a promising approach to develop clinically useful algorithms that provide multiple actionable predictions from a single slide.
Authors: Ziyuan Huang, Kaixiang Ji, Biao Gong, Zhiwu Qing, Qinglong Zhang, Kecheng Zheng, Jian Wang, Jingdong Chen, Ming Yang
Abstract: This paper introduces Chain-of-Sight, a vision-language bridge module that accelerates the pre-training of Multimodal Large Language Models (MLLMs). Our approach employs a sequence of visual resamplers that capture visual details at various spacial scales. This architecture not only leverages global and local visual contexts effectively, but also facilitates the flexible extension of visual tokens through a compound token scaling strategy, allowing up to a 16x increase in the token count post pre-training. Consequently, Chain-of-Sight requires significantly fewer visual tokens in the pre-training phase compared to the fine-tuning phase. This intentional reduction of visual tokens during pre-training notably accelerates the pre-training process, cutting down the wall-clock training time by ~73%. Empirical results on a series of vision-language benchmarks reveal that the pre-train acceleration through Chain-of-Sight is achieved without sacrificing performance, matching or surpassing the standard pipeline of utilizing all visual tokens throughout the entire training process. Further scaling up the number of visual tokens for pre-training leads to stronger performances, competitive to existing approaches in a series of benchmarks.
Authors: Yibing Wei, Abhinav Gupta, Pedro Morgado
Abstract: Masked Image Modeling (MIM) has emerged as a promising method for deriving visual representations from unlabeled image data by predicting missing pixels from masked portions of images. It excels in region-aware learning and provides strong initializations for various tasks, but struggles to capture high-level semantics without further supervised fine-tuning, likely due to the low-level nature of its pixel reconstruction objective. A promising yet unrealized framework is learning representations through masked reconstruction in latent space, combining the locality of MIM with the high-level targets. However, this approach poses significant training challenges as the reconstruction targets are learned in conjunction with the model, potentially leading to trivial or suboptimal solutions.Our study is among the first to thoroughly analyze and address the challenges of such framework, which we refer to as Latent MIM. Through a series of carefully designed experiments and extensive analysis, we identify the source of these challenges, including representation collapsing for joint online/target optimization, learning objectives, the high region correlation in latent space and decoding conditioning. By sequentially addressing these issues, we demonstrate that Latent MIM can indeed learn high-level representations while retaining the benefits of MIM models.
Authors: Yangzhou Liu, Yue Cao, Zhangwei Gao, Weiyun Wang, Zhe Chen, Wenhai Wang, Hao Tian, Lewei Lu, Xizhou Zhu, Tong Lu, Yu Qiao, Jifeng Dai
Abstract: Despite the effectiveness of vision-language supervised fine-tuning in enhancing the performance of Vision Large Language Models (VLLMs). However, existing visual instruction tuning datasets include the following limitations: (1) Instruction annotation quality: despite existing VLLMs exhibiting strong performance, instructions generated by those advanced VLLMs may still suffer from inaccuracies, such as hallucinations. (2) Instructions and image diversity: the limited range of instruction types and the lack of diversity in image data may impact the model's ability to generate diversified and closer to real-world scenarios outputs. To address these challenges, we construct a high-quality, diverse visual instruction tuning dataset MMInstruct, which consists of 973K instructions from 24 domains. There are four instruction types: Judgement, Multiple-Choice, Long Visual Question Answering and Short Visual Question Answering. To construct MMInstruct, we propose an instruction generation data engine that leverages GPT-4V, GPT-3.5, and manual correction. Our instruction generation engine enables semi-automatic, low-cost, and multi-domain instruction generation at 1/6 the cost of manual construction. Through extensive experiment validation and ablation experiments, we demonstrate that MMInstruct could significantly improve the performance of VLLMs, e.g., the model fine-tuning on MMInstruct achieves new state-of-the-art performance on 10 out of 12 benchmarks. The code and data shall be available at https://github.com/yuecao0119/MMInstruct.
Authors: Mingze Xu, Mingfei Gao, Zhe Gan, Hong-You Chen, Zhengfeng Lai, Haiming Gang, Kai Kang, Afshin Dehghan
Abstract: We propose SlowFast-LLaVA (or SF-LLaVA for short), a training-free video large language model (LLM) that can jointly capture the detailed spatial semantics and long-range temporal context without exceeding the token budget of commonly used LLMs. This is realized by using a two-stream SlowFast design of inputs for Video LLMs to aggregate features from sampled video frames in an effective way. Specifically, the Slow pathway extracts features at a low frame rate while keeping as many spatial details as possible (e.g., with 24x24 tokens), and the Fast pathway operates on a high frame rate but uses a larger spatial pooling stride (e.g., downsampling 6x) to focus on the motion cues. As a result, this design allows us to adequately capture both spatial and temporal features that are beneficial for understanding details along the video. Experimental results show that SF-LLaVA outperforms existing training-free methods on a wide range of video tasks. On some benchmarks, it achieves comparable or even better performance compared to state-of-the-art Video LLMs that are fine-tuned on video datasets.
Authors: Ruixiang Jiang, Changwen Chen
Abstract: Diffusion models entangle content and style generation during the denoising process, leading to undesired content modification when directly applied to stylization tasks. Existing methods struggle to effectively control the diffusion model to meet the aesthetic-level requirements for stylization. In this paper, we introduce \textbf{Artist}, a training-free approach that aesthetically controls the content and style generation of a pretrained diffusion model for text-driven stylization. Our key insight is to disentangle the denoising of content and style into separate diffusion processes while sharing information between them. We propose simple yet effective content and style control methods that suppress style-irrelevant content generation, resulting in harmonious stylization results. Extensive experiments demonstrate that our method excels at achieving aesthetic-level stylization requirements, preserving intricate details in the content image and aligning well with the style prompt. Furthermore, we showcase the highly controllability of the stylization strength from various perspectives. Code will be released, project home page: https://DiffusionArtist.github.io
Authors: Shadi Hamdan, Fatma G\"uney
Abstract: The choice of representation plays a key role in self-driving. Bird's eye view (BEV) representations have shown remarkable performance in recent years. In this paper, we propose to learn object-centric representations in BEV to distill a complex scene into more actionable information for self-driving. We first learn to place objects into slots with a slot attention model on BEV sequences. Based on these object-centric representations, we then train a transformer to learn to drive as well as reason about the future of other vehicles. We found that object-centric slot representations outperform both scene-level and object-level approaches that use the exact attributes of objects. Slot representations naturally incorporate information about objects from their spatial and temporal context such as position, heading, and speed without explicitly providing it. Our model with slots achieves an increased completion rate of the provided routes and, consequently, a higher driving score, with a lower variance across multiple runs, affirming slots as a reliable alternative in object-centric approaches. Additionally, we validate our model's performance as a world model through forecasting experiments, demonstrating its capability to predict future slot representations accurately. The code and the pre-trained models can be found at https://kuis-ai.github.io/CarFormer/.
Authors: Eugene Valassakis, Guillermo Garcia-Hernando
Abstract: Predicting camera-space hand meshes from single RGB images is crucial for enabling realistic hand interactions in 3D virtual and augmented worlds. Previous work typically divided the task into two stages: given a cropped image of the hand, predict meshes in relative coordinates, followed by lifting these predictions into camera space in a separate and independent stage, often resulting in the loss of valuable contextual and scale information. To prevent the loss of these cues, we propose unifying these two stages into an end-to-end solution that addresses the 2D-3D correspondence problem. This solution enables back-propagation from camera space outputs to the rest of the network through a new differentiable global positioning module. We also introduce an image rectification step that harmonizes both the training dataset and the input image as if they were acquired with the same camera, helping to alleviate the inherent scale-depth ambiguity of the problem. We validate the effectiveness of our framework in evaluations against several baselines and state-of-the-art approaches across three public benchmarks.
Authors: Chih-Hai Su, Chih-Yao Hu, Shr-Ruei Tsai, Jie-Ying Lee, Chin-Yang Lin, Yu-Lun Liu
Abstract: While Neural Radiance Fields (NeRFs) have demonstrated exceptional quality, their protracted training duration remains a limitation. Generalizable and MVS-based NeRFs, although capable of mitigating training time, often incur tradeoffs in quality. This paper presents a novel approach called BoostMVSNeRFs to enhance the rendering quality of MVS-based NeRFs in large-scale scenes. We first identify limitations in MVS-based NeRF methods, such as restricted viewport coverage and artifacts due to limited input views. Then, we address these limitations by proposing a new method that selects and combines multiple cost volumes during volume rendering. Our method does not require training and can adapt to any MVS-based NeRF methods in a feed-forward fashion to improve rendering quality. Furthermore, our approach is also end-to-end trainable, allowing fine-tuning on specific scenes. We demonstrate the effectiveness of our method through experiments on large-scale datasets, showing significant rendering quality improvements in large-scale scenes and unbounded outdoor scenarios. We release the source code of BoostMVSNeRFs at https://su-terry.github.io/BoostMVSNeRFs/.
Authors: Junyu Xie, Tengda Han, Max Bain, Arsha Nagrani, G\"ul Varol, Weidi Xie, Andrew Zisserman
Abstract: Our objective is to generate Audio Descriptions (ADs) for both movies and TV series in a training-free manner. We use the power of off-the-shelf Visual-Language Models (VLMs) and Large Language Models (LLMs), and develop visual and text prompting strategies for this task. Our contributions are three-fold: (i) We demonstrate that a VLM can successfully name and refer to characters if directly prompted with character information through visual indications without requiring any fine-tuning; (ii) A two-stage process is developed to generate ADs, with the first stage asking the VLM to comprehensively describe the video, followed by a second stage utilising a LLM to summarise dense textual information into one succinct AD sentence; (iii) A new dataset for TV audio description is formulated. Our approach, named AutoAD-Zero, demonstrates outstanding performance (even competitive with some models fine-tuned on ground truth ADs) in AD generation for both movies and TV series, achieving state-of-the-art CRITIC scores.
Authors: Jayadeva, Himanshu Pant, Mayank Sharma, Abhimanyu Dubey, Sumit Soman, Suraj Tripathi, Sai Guruju, Nihal Goalla
Abstract: Modern neural network architectures for large-scale learning tasks have substantially higher model complexities, which makes understanding, visualizing and training these architectures difficult. Recent contributions to deep learning techniques have focused on architectural modifications to improve parameter efficiency and performance. In this paper, we derive a continuous and differentiable error functional for a neural network that minimizes its empirical error as well as a measure of the model complexity. The latter measure is obtained by deriving a differentiable upper bound on the Vapnik-Chervonenkis (VC) dimension of the classifier layer of a class of deep networks. Using standard backpropagation, we realize a training rule that tries to minimize the error on training samples, while improving generalization by keeping the model complexity low. We demonstrate the effectiveness of our formulation (the Low Complexity Neural Network - LCNN) across several deep learning algorithms, and a variety of large benchmark datasets. We show that hidden layer neurons in the resultant networks learn features that are crisp, and in the case of image datasets, quantitatively sharper. Our proposed approach yields benefits across a wide range of architectures, in comparison to and in conjunction with methods such as Dropout and Batch Normalization, and our results strongly suggest that deep learning techniques can benefit from model complexity control methods such as the LCNN learning rule.
Authors: Yi Sheng, Hanchen Wang, Yipei Liu, Junhuan Yang, Weiwen Jiang, Youzuo Lin, Lei Yang
Abstract: Ultrasound computed tomography (USCT) is a promising technique that achieves superior medical imaging reconstruction resolution by fully leveraging waveform information, outperforming conventional ultrasound methods. Despite its advantages, high-quality USCT reconstruction relies on extensive data acquisition by a large number of transducers, leading to increased costs, computational demands, extended patient scanning times, and manufacturing complexities. To mitigate these issues, we propose a new USCT method called APS-USCT, which facilitates imaging with sparse data, substantially reducing dependence on high-cost dense data acquisition. Our APS-USCT method consists of two primary components: APS-wave and APS-FWI. The APS-wave component, an encoder-decoder system, preprocesses the waveform data, converting sparse data into dense waveforms to augment sample density prior to reconstruction. The APS-FWI component, utilizing the InversionNet, directly reconstructs the speed of sound (SOS) from the ultrasound waveform data. We further improve the model's performance by incorporating Squeeze-and-Excitation (SE) Blocks and source encoding techniques. Testing our method on a breast cancer dataset yielded promising results. It demonstrated outstanding performance with an average Structural Similarity Index (SSIM) of 0.8431. Notably, over 82% of samples achieved an SSIM above 0.8, with nearly 61% exceeding 0.85, highlighting the significant potential of our approach in improving USCT image reconstruction by efficiently utilizing sparse data.
Authors: D. Denipitiyage, B. Silva, K. Gunathilaka, S. Seneviratne, A. Mahanti, A. Seneviratne, S. Chawla
Abstract: App markets have evolved into highly competitive and dynamic environments for developers. While the traditional app life cycle involves incremental updates for feature enhancements and issue resolution, some apps deviate from this norm by undergoing significant transformations in their use cases or market positioning. We define this previously unstudied phenomenon as 'app metamorphosis'. In this paper, we propose a novel and efficient multi-modal search methodology to identify apps undergoing metamorphosis and apply it to analyse two snapshots of the Google Play Store taken five years apart. Our methodology uncovers various metamorphosis scenarios, including re-births, re-branding, re-purposing, and others, enabling comprehensive characterisation. Although these transformations may register as successful for app developers based on our defined success score metric (e.g., re-branded apps performing approximately 11.3% better than an average top app), we shed light on the concealed security and privacy risks that lurk within, potentially impacting even tech-savvy end-users.
Authors: Yiying Wang, Abhirup Banerjee, Robin P. Choudhury, Vicente Grau
Abstract: Cardiovascular diseases (CVDs) are the most common cause of death worldwide. Invasive x-ray coronary angiography (ICA) is one of the most important imaging modalities for the diagnosis of CVDs. ICA typically acquires only two 2D projections, which makes the 3D geometry of coronary vessels difficult to interpret, thus requiring 3D coronary tree reconstruction from two projections. State-of-the-art approaches require significant manual interactions and cannot correct the non-rigid cardiac and respiratory motions between non-simultaneous projections. In this study, we propose a novel deep learning pipeline. We leverage the Wasserstein conditional generative adversarial network with gradient penalty, latent convolutional transformer layers, and a dynamic snake convolutional critic to implicitly compensate for the non-rigid motion and provide 3D coronary tree reconstruction. Through simulating projections from coronary computed tomography angiography (CCTA), we achieve the generalisation of 3D coronary tree reconstruction on real non-simultaneous ICA projections. We incorporate an application-specific evaluation metric to validate our proposed model on both a CCTA dataset and a real ICA dataset, together with Chamfer L1 distance. The results demonstrate the good performance of our model in vessel topology preservation, recovery of missing features, and generalisation ability to real ICA data. To the best of our knowledge, this is the first study that leverages deep learning to achieve 3D coronary tree reconstruction from two real non-simultaneous x-ray angiography projections.
Authors: Yuetan Chu, Yilan Zhang, Zhongyi Han, Changchun Yang, Longxi Zhou, Gongning Luo, Xin Gao
Abstract: Foundation models have recently attracted significant attention for their impressive generalizability across diverse downstream tasks. However, these models are demonstrated to exhibit great limitations in representing high-frequency components and fine-grained details. In many medical imaging tasks, the precise representation of such information is crucial due to the inherently intricate anatomical structures, sub-visual features, and complex boundaries involved. Consequently, the limited representation of prevalent foundation models can result in significant performance degradation or even failure in these tasks. To address these challenges, we propose a novel pretraining strategy, named Frequency-advanced Representation Autoencoder (Frepa). Through high-frequency masking and low-frequency perturbation combined with adversarial learning, Frepa encourages the encoder to effectively represent and preserve high-frequency components in the image embeddings. Additionally, we introduce an innovative histogram-equalized image masking strategy, extending the Masked Autoencoder approach beyond ViT to other architectures such as Swin Transformer and convolutional networks. We develop Frepa across nine medical modalities and validate it on 32 downstream tasks for both 2D images and 3D volume data. Without fine-tuning, Frepa can outperform other self-supervised pretraining methods and, in some cases, even surpasses task-specific trained models. This improvement is particularly significant for tasks involving fine-grained details, such as achieving up to a +15% increase in DSC for retina vessel segmentation and a +7% increase in IoU for lung nodule detection. Further experiments quantitatively reveal that Frepa enables superior high-frequency representations and preservation in the embeddings, underscoring its potential for developing more generalized and universal medical image foundation models.
Authors: Marcelo Keese Albertini, Andr\'e Ricardo Backes
Abstract: Productivity in Research (PQ) is a scholarship granted by CNPq (Brazilian National Council for Scientific and Technological Development). This scholarship aims to recognize a few selected faculty researchers for their scientific production, outstanding technology and innovation in their respective areas of knowledge. In the present study, we evaluated the scientific production of the 185 researchers in the Computer Science area granted with PQ scholarship in the last PQ selection notice. To evaluate the productivity of each professor, we considered papers published in scientific journals and conferences (complete works) in a five years period (from 2017 to 2021). We analyzed the productivity in terms of both quantity and quality. We also evaluated its distribution over the country, universities and research facilities, as well as, the co-authorship network produced.
Authors: Ahmed Radwan, Islam Osman, Mohamed S. Shehata
Abstract: Achieving domain generalization in medical imaging poses a significant challenge, primarily due to the limited availability of publicly labeled datasets in this domain. This limitation arises from concerns related to data privacy and the necessity for medical expertise to accurately label the data. In this paper, we propose a federated learning approach to transfer knowledge from multiple local models to a global model, eliminating the need for direct access to the local datasets used to train each model. The primary objective is to train a global model capable of performing a wide variety of medical imaging tasks. This is done while ensuring the confidentiality of the private datasets utilized during the training of these models. To validate the effectiveness of our approach, extensive experiments were conducted on eight datasets, each corresponding to a different medical imaging application. The client's data distribution in our experiments varies significantly as they originate from diverse domains. Despite this variation, we demonstrate a statistically significant improvement over a state-of-the-art baseline utilizing masked image modeling over a diverse pre-training dataset that spans different body parts and scanning types. This improvement is achieved by curating information learned from clients without accessing any labeled dataset on the server.
Authors: Jayneel Vora, Nader Bouacida, Aditya Krishnan, Prasant Mohapatra
Abstract: We introduce FedDM, a novel training framework designed for the federated training of diffusion models. Our theoretical analysis establishes the convergence of diffusion models when trained in a federated setting, presenting the specific conditions under which this convergence is guaranteed. We propose a suite of training algorithms that leverage the U-Net architecture as the backbone for our diffusion models. These include a basic Federated Averaging variant, FedDM-vanilla, FedDM-prox to handle data heterogeneity among clients, and FedDM-quant, which incorporates a quantization module to reduce the model update size, thereby enhancing communication efficiency across the federated network. We evaluate our algorithms on FashionMNIST (28x28 resolution), CIFAR-10 (32x32 resolution), and CelebA (64x64 resolution) for DDPMs, as well as LSUN Church Outdoors (256x256 resolution) for LDMs, focusing exclusively on the imaging modality. Our evaluation results demonstrate that FedDM algorithms maintain high generation quality across image resolutions. At the same time, the use of quantized updates and proximal terms in the local training objective significantly enhances communication efficiency (up to 4x) and model convergence, particularly in non-IID data settings, at the cost of increased FID scores (up to 1.75x).
Authors: Fan Zhao, You Chen
Abstract: Deep learning-based methods for Time Series Classification (TSC) typically utilize deep networks to extract features, which are then processed through a combination of a Fully Connected (FC) layer and a SoftMax function. However, we have observed the phenomenon of inter-class similarity and intra-class inconsistency in the datasets from the UCR archive and further analyzed how this phenomenon adversely affects the "FC+SoftMax" paradigm. To address the issue, we introduce ECR, which, for the first time to our knowledge, applies deep learning-based retrieval algorithm to the TSC problem and integrates classification and retrieval models. Experimental results on 112 UCR datasets demonstrate that ECR is state-of-the-art(sota) compared to existing deep learning-based methods. Furthermore, we have developed a more precise classifier, ECRTime, which is an ensemble of ECR. ECRTime surpasses the currently most accurate deep learning classifier, InceptionTime, in terms of accuracy, achieving this with reduced training time and comparable scalability.
Authors: Jiaxing Huang, Yanfeng Zhou, Yaoru Luo, Guole Liu, Heng Guo, Ge Yang
Abstract: Accurate segmentation of long and thin tubular structures is required in a wide variety of areas such as biology, medicine, and remote sensing. The complex topology and geometry of such structures often pose significant technical challenges. A fundamental property of such structures is their topological self-similarity, which can be quantified by fractal features such as fractal dimension (FD). In this study, we incorporate fractal features into a deep learning model by extending FD to the pixel-level using a sliding window technique. The resulting fractal feature maps (FFMs) are then incorporated as additional input to the model and additional weight in the loss function to enhance segmentation performance by utilizing the topological self-similarity. Moreover, we extend the U-Net architecture by incorporating an edge decoder and a skeleton decoder to improve boundary accuracy and skeletal continuity of segmentation, respectively. Extensive experiments on five tubular structure datasets validate the effectiveness and robustness of our approach. Furthermore, the integration of FFMs with other popular segmentation models such as HR-Net also yields performance enhancement, suggesting FFM can be incorporated as a plug-in module with different model architectures. Code and data are openly accessible at https://github.com/cbmi-group/FFM-Multi-Decoder-Network.
URLs: https://github.com/cbmi-group/FFM-Multi-Decoder-Network.
Authors: Anubhav Gupta, Islam Osman, Mohamed S. Shehata, John W. Braun
Abstract: Medical imaging tasks are very challenging due to the lack of publicly available labeled datasets. Hence, it is difficult to achieve high performance with existing deep-learning models as they require a massive labeled dataset to be trained effectively. An alternative solution is to use pre-trained models and fine-tune them using the medical imaging dataset. However, all existing models are pre-trained using natural images, which is a completely different domain from that of medical imaging, which leads to poor performance due to domain shift. To overcome these problems, we propose a large-scale unlabeled dataset of medical images and a backbone pre-trained using the proposed dataset with a self-supervised learning technique called Masked autoencoder. This backbone can be used as a pre-trained model for any medical imaging task, as it is trained to learn a visual representation of different types of medical images. To evaluate the performance of the proposed backbone, we used four different medical imaging tasks. The results are compared with existing pre-trained models. These experiments show the superiority of our proposed backbone in medical imaging tasks.
Authors: Quentin Malartic, Nilabhra Roy Chowdhury, Ruxandra Cojocaru, Mugariya Farooq, Giulia Campesan, Yasser Abdelaziz Dahou Djilali, Sanath Narayan, Ankit Singh, Maksim Velikanov, Basma El Amel Boussaha, Mohammed Al-Yafeai, Hamza Alobeidli, Leen Al Qadi, Mohamed El Amine Seddik, Kirill Fedyanin, Reda Alami, Hakim Hacid
Abstract: We introduce Falcon2-11B, a foundation model trained on over five trillion tokens, and its multimodal counterpart, Falcon2-11B-vlm, which is a vision-to-text model. We report our findings during the training of the Falcon2-11B which follows a multi-stage approach where the early stages are distinguished by their context length and a final stage where we use a curated, high-quality dataset. Additionally, we report the effect of doubling the batch size mid-training and how training loss spikes are affected by the learning rate. The downstream performance of the foundation model is evaluated on established benchmarks, including multilingual and code datasets. The foundation model shows strong generalization across all the tasks which makes it suitable for downstream finetuning use cases. For the vision language model, we report the performance on several benchmarks and show that our model achieves a higher average score compared to open-source models of similar size. The model weights and code of both Falcon2-11B and Falcon2-11B-vlm are made available under a permissive license.
Authors: Yuening Li, Xiao Fu, Junbin Liu, Wing-Kin Ma
Abstract: This work proposes a variational inference (VI) framework for hyperspectral unmixing in the presence of endmember variability (HU-EV). An EV-accounted noisy linear mixture model (LMM) is considered, and the presence of outliers is also incorporated into the model. Following the marginalized maximum likelihood (MML) principle, a VI algorithmic structure is designed for probabilistic inference for HU-EV. Specifically, a patch-wise static endmember assumption is employed to exploit spatial smoothness and to try to overcome the ill-posed nature of the HU-EV problem. The design facilitates lightweight, continuous optimization-based updates under a variety of endmember priors. Some of the priors, such as the Beta prior, were previously used under computationally heavy, sampling-based probabilistic HU-EV methods. The effectiveness of the proposed framework is demonstrated through synthetic, semi-real, and real-data experiments.
Authors: Chen Shen, Chunfeng Lian, Wanqing Zhang, Fan Wang, Jianhua Zhang, Shuanliang Fan, Xin Wei, Gongji Wang, Kehan Li, Hongshu Mu, Hao Wu, Xinggong Liang, Jianhua Ma, Zhenyuan Wang
Abstract: Forensic pathology is critical in determining the cause and manner of death through post-mortem examinations, both macroscopic and microscopic. The field, however, grapples with issues such as outcome variability, laborious processes, and a scarcity of trained professionals. This paper presents SongCi, an innovative visual-language model (VLM) designed specifically for forensic pathology. SongCi utilizes advanced prototypical cross-modal self-supervised contrastive learning to enhance the accuracy, efficiency, and generalizability of forensic analyses. It was pre-trained and evaluated on a comprehensive multi-center dataset, which includes over 16 million high-resolution image patches, 2,228 vision-language pairs of post-mortem whole slide images (WSIs), and corresponding gross key findings, along with 471 distinct diagnostic outcomes. Our findings indicate that SongCi surpasses existing multi-modal AI models in many forensic pathology tasks, performs comparably to experienced forensic pathologists and significantly better than less experienced ones, and provides detailed multi-modal explainability, offering critical assistance in forensic investigations. To the best of our knowledge, SongCi is the first VLM specifically developed for forensic pathological analysis and the first large-vocabulary computational pathology (CPath) model that directly processes gigapixel WSIs in forensic science.
Authors: Chen Wei, Jiachen Zou, Dietmar Heinke, Quanying Liu
Abstract: Humans interpret complex visual stimuli using abstract concepts that facilitate decision-making tasks such as food selection and risk avoidance. Similarity judgment tasks are effective for exploring these concepts. However, methods for controllable image generation in concept space are underdeveloped. In this study, we present a novel framework called CoCoG-2, which integrates generated visual stimuli into similarity judgment tasks. CoCoG-2 utilizes a training-free guidance algorithm to enhance generation flexibility. CoCoG-2 framework is versatile for creating experimental stimuli based on human concepts, supporting various strategies for guiding visual stimuli generation, and demonstrating how these stimuli can validate various experimental hypotheses. CoCoG-2 will advance our understanding of the causal relationship between concept representations and behaviors by generating visual stimuli. The code is available at \url{https://github.com/ncclab-sustech/CoCoG-2}.
Authors: Yijie Yuan, Grace J. Gang, J. Webster Stayman
Abstract: The restoration of images affected by blur and noise has been widely studied and has broad potential for applications including in medical imaging modalities like computed tomography (CT). Although the blur and noise in CT images can be attributed to a variety of system factors, these image properties can often be modeled and predicted accurately and used in classical restoration approaches for deconvolution and denoising. In classical approaches, simultaneous deconvolution and denoising can be challenging and often represent competing goals. Recently, deep learning approaches have demonstrated the potential to enhance image quality beyond classic limits; however, most deep learning models attempt a blind restoration problem and base their restoration on image inputs alone without direct knowledge of the image noise and blur properties. In this work, we present a method that leverages both degraded image inputs and a characterization of the system blur and noise to combine modeling and deep learning approaches. Different methods to integrate these auxiliary inputs are presented. Namely, an input-variant and a weight-variant approach wherein the auxiliary inputs are incorporated as a parameter vector before and after the convolutional block, respectively, allowing easy integration into any CNN architecture. The proposed model shows superior performance compared to baseline models lacking auxiliary inputs. Evaluations are based on the average Peak Signal-to-Noise Ratio (PSNR), selected examples of good and poor performance for varying approaches, and an input space analysis to assess the effect of different noise and blur on performance. Results demonstrate the efficacy of providing a deep learning model with auxiliary inputs, representing system blur and noise characteristics, to enhance the performance of the model in image restoration tasks.
Authors: Karl Van Eeden Risager, Torkan Gholamalizadeh, Mostafa Mehdipour Ghazi
Abstract: Generating high-quality synthetic data is crucial for addressing challenges in medical imaging, such as domain adaptation, data scarcity, and privacy concerns. Existing image quality metrics often rely on reference images, are tailored for group comparisons, or are intended for 2D natural images, limiting their efficacy in complex domains like medical imaging. This study introduces a novel deep learning-based non-reference approach to assess brain MRI quality by training a 3D ResNet. The network is designed to estimate quality across six distinct artifacts commonly encountered in MRI scans. Additionally, a diffusion model is trained on diverse datasets to generate synthetic 3D images of high fidelity. The approach leverages several datasets for training and comprehensive quality assessment, benchmarking against state-of-the-art metrics for real and synthetic images. Results demonstrate superior performance in accurately estimating distortions and reflecting image quality from multiple perspectives. Notably, the method operates without reference images, indicating its applicability for evaluating deep generative models. Besides, the quality scores in the [0, 1] range provide an intuitive assessment of image quality across heterogeneous datasets. Evaluation of generated images offers detailed insights into specific artifacts, guiding strategies for improving generative models to produce high-quality synthetic images. This study presents the first comprehensive method for assessing the quality of real and synthetic 3D medical images in MRI contexts without reliance on reference images.
Authors: Mengru Wang, Yunzhi Yao, Ziwen Xu, Shuofei Qiao, Shumin Deng, Peng Wang, Xiang Chen, Jia-Chen Gu, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen, Ningyu Zhang
Abstract: Understanding knowledge mechanisms in Large Language Models (LLMs) is crucial for advancing towards trustworthy AGI. This paper reviews knowledge mechanism analysis from a novel taxonomy including knowledge utilization and evolution. Knowledge utilization delves into the mechanism of memorization, comprehension and application, and creation. Knowledge evolution focuses on the dynamic progression of knowledge within individual and group LLMs. Moreover, we discuss what knowledge LLMs have learned, the reasons for the fragility of parametric knowledge, and the potential dark knowledge (hypothesis) that will be challenging to address. We hope this work can help understand knowledge in LLMs and provide insights for future research.
Authors: Beibei Li, Yiyuan Zheng, Beihong Jin, Tao Xiang, Haobo Wang, Lei Feng
Abstract: Partial-Label Learning (PLL) is a typical problem of weakly supervised learning, where each training instance is annotated with a set of candidate labels. Self-training PLL models achieve state-of-the-art performance but suffer from error accumulation problem caused by mistakenly disambiguated instances. Although co-training can alleviate this issue by training two networks simultaneously and allowing them to interact with each other, most existing co-training methods train two structurally identical networks with the same task, i.e., are symmetric, rendering it insufficient for them to correct each other due to their similar limitations. Therefore, in this paper, we propose an asymmetric dual-task co-training PLL model called AsyCo, which forces its two networks, i.e., a disambiguation network and an auxiliary network, to learn from different views explicitly by optimizing distinct tasks. Specifically, the disambiguation network is trained with self-training PLL task to learn label confidence, while the auxiliary network is trained in a supervised learning paradigm to learn from the noisy pairwise similarity labels that are constructed according to the learned label confidence. Finally, the error accumulation problem is mitigated via information distillation and confidence refinement. Extensive experiments on both uniform and instance-dependent partially labeled datasets demonstrate the effectiveness of AsyCo. The code is available at https://github.com/libeibeics/AsyCo.
Authors: Fred Grabovski, Lior Yasur, Guy Amit, Yuval Elovici, Yisroel Mirsky
Abstract: Recent progress in generative models has made it easier for a wide audience to edit and create image content, raising concerns about the proliferation of deepfakes, especially in healthcare. Despite the availability of numerous techniques for detecting manipulated images captured by conventional cameras, their applicability to medical images is limited. This limitation stems from the distinctive forensic characteristics of medical images, a result of their imaging process. In this work we propose a novel anomaly detector for medical imagery based on diffusion models. Normally, diffusion models are used to generate images. However, we show how a similar process can be used to detect synthetic content by making a model reverse the diffusion on a suspected image. We evaluate our method on the task of detecting fake tumors injected and removed from CT and MRI scans. Our method significantly outperforms other state of the art unsupervised detectors with an increased AUC of 0.9 from 0.79 for injection and of 0.96 from 0.91 for removal on average.
Authors: Rylan Schaeffer, Dan Valentine, Luke Bailey, James Chua, Crist\'obal Eyzaguirre, Zane Durante, Joe Benton, Brando Miranda, Henry Sleight, John Hughes, Rajashree Agrawal, Mrinank Sharma, Scott Emmons, Sanmi Koyejo, Ethan Perez
Abstract: The integration of new modalities into frontier AI systems offers exciting capabilities, but also increases the possibility such systems can be adversarially manipulated in undesirable ways. In this work, we focus on a popular class of vision-language models (VLMs) that generate text outputs conditioned on visual and textual inputs. We conducted a large-scale empirical study to assess the transferability of gradient-based universal image "jailbreaks" using a diverse set of over 40 open-parameter VLMs, including 18 new VLMs that we publicly release. Overall, we find that transferable gradient-based image jailbreaks are extremely difficult to obtain. When an image jailbreak is optimized against a single VLM or against an ensemble of VLMs, the jailbreak successfully jailbreaks the attacked VLM(s), but exhibits little-to-no transfer to any other VLMs; transfer is not affected by whether the attacked and target VLMs possess matching vision backbones or language models, whether the language model underwent instruction-following and/or safety-alignment training, or many other factors. Only two settings display partially successful transfer: between identically-pretrained and identically-initialized VLMs with slightly different VLM training data, and between different training checkpoints of a single VLM. Leveraging these results, we then demonstrate that transfer can be significantly improved against a specific target VLM by attacking larger ensembles of "highly-similar" VLMs. These results stand in stark contrast to existing evidence of universal and transferable text jailbreaks against language models and transferable adversarial attacks against image classifiers, suggesting that VLMs may be more robust to gradient-based transfer attacks.
Authors: Malek Ben Alaya, Daniel M. Lang, Benedikt Wiestler, Julia A. Schnabel, Cosmin I. Bercea
Abstract: Denoising diffusion probabilistic models enable high-fidelity image synthesis and editing. In biomedicine, these models facilitate counterfactual image editing, producing pairs of images where one is edited to simulate hypothetical conditions. For example, they can model the progression of specific diseases, such as stroke lesions. However, current image editing techniques often fail to generate realistic biomedical counterfactuals, either by inadequately modeling indirect pathological effects like brain atrophy or by excessively altering the scan, which disrupts correspondence to the original images. Here, we propose MedEdit, a conditional diffusion model for medical image editing. MedEdit induces pathology in specific areas while balancing the modeling of disease effects and preserving the integrity of the original scan. We evaluated MedEdit on the Atlas v2.0 stroke dataset using Frechet Inception Distance and Dice scores, outperforming state-of-the-art diffusion-based methods such as Palette (by 45%) and SDEdit (by 61%). Additionally, clinical evaluations by a board-certified neuroradiologist confirmed that MedEdit generated realistic stroke scans indistinguishable from real ones. We believe this work will enable counterfactual image editing research to further advance the development of realistic and clinically useful imaging tools.
Authors: Mathieu Labb\'e, Fran\c{c}ois Michaud
Abstract: In appearance-based localization and mapping, loop closure detection is the process used to determinate if the current observation comes from a previously visited location or a new one. As the size of the internal map increases, so does the time required to compare new observations with all stored locations, eventually limiting online processing. This paper presents an online loop closure detection approach for large-scale and long-term operation. The approach is based on a memory management method, which limits the number of locations used for loop closure detection so that the computation time remains under real-time constraints. The idea consists of keeping the most recent and frequently observed locations in a Working Memory (WM) used for loop closure detection, and transferring the others into a Long-Term Memory (LTM). When a match is found between the current location and one stored in WM, associated locations stored in LTM can be updated and remembered for additional loop closure detections. Results demonstrate the approach's adaptability and scalability using ten standard data sets from other appearance-based loop closure approaches, one custom data set using real images taken over a 2 km loop of our university campus, and one custom data set (7 hours) using virtual images from the racing video game ``Need for Speed: Most Wanted''.
Authors: Luciano Carvalho Ayres, S\'ergio Jos\'e Melo de Almeida, Jos\'e Carlos Moreira Bermudez, Ricardo Augusto Borsoi
Abstract: Hyperspectral image (HI) analysis approaches have recently become increasingly complex and sophisticated. Recently, the combination of spectral-spatial information and superpixel techniques have addressed some hyperspectral data issues, such as the higher spatial variability of spectral signatures and dimensionality of the data. However, most existing superpixel approaches do not account for specific HI characteristics resulting from its high spectral dimension. In this work, we propose a multiscale superpixel method that is computationally efficient for processing hyperspectral data. The Simple Linear Iterative Clustering (SLIC) oversegmentation algorithm, on which the technique is based, has been extended hierarchically. Using a novel robust homogeneity testing, the proposed hierarchical approach leads to superpixels of variable sizes but with higher spectral homogeneity when compared to the classical SLIC segmentation. For validation, the proposed homogeneity-based hierarchical method was applied as a preprocessing step in the spectral unmixing and classification tasks carried out using, respectively, the Multiscale sparse Unmixing Algorithm (MUA) and the CNN-Enhanced Graph Convolutional Network (CEGCN) methods. Simulation results with both synthetic and real data show that the technique is competitive with state-of-the-art solutions.
Authors: Zeke Zexi Hu, Haodong Chen, Yuk Ying Chung, Xiaoming Chen
Abstract: This paper presents the Multi-scale Disparity Transformer (MDT), a novel Transformer tailored for light field image super-resolution (LFSR) that addresses the issues of computational redundancy and disparity entanglement caused by the indiscriminate processing of sub-aperture images inherent in conventional methods. MDT features a multi-branch structure, with each branch utilising independent disparity self-attention (DSA) to target specific disparity ranges, effectively reducing computational complexity and disentangling disparities. Building on this architecture, we present LF-MDTNet, an efficient LFSR network. Experimental results demonstrate that LF-MDTNet outperforms existing state-of-the-art methods by 0.37 dB and 0.41 dB PSNR at the 2x and 4x scales, achieving superior performance with fewer parameters and higher speed.
Authors: Melissa Dell
Abstract: Deep learning provides powerful methods to impute structured information from large-scale, unstructured text and image datasets. For example, economists might wish to detect the presence of economic activity in satellite images, or to measure the topics or entities mentioned in social media, the congressional record, or firm filings. This review introduces deep neural networks, covering methods such as classifiers, regression models, generative AI, and embedding models. Applications include classification, document digitization, record linkage, and methods for data exploration in massive scale text and image corpora. When suitable methods are used, deep learning models can be cheap to tune and can scale affordably to problems involving millions or billions of data points.. The review is accompanied by a companion website, EconDL, with user-friendly demo notebooks, software resources, and a knowledge base that provides technical details and additional applications.
Authors: Ligen Shi, Chang Liu, Xing Zhao, Jun Qiu
Abstract: This study proposes a neural disparity field (NDF) that establishes an implicit, continuous representation of scene disparity based on a neural field and an iterative approach to address the inverse problem of NDF reconstruction from light-field data. NDF enables seamless and precise characterization of disparity variations in three-dimensional scenes and can discretize disparity at any arbitrary resolution, overcoming the limitations of traditional disparity maps that are prone to sampling errors and interpolation inaccuracies. The proposed NDF network architecture utilizes hash encoding combined with multilayer perceptrons to capture detailed disparities in texture levels, thereby enhancing its ability to represent the geometric information of complex scenes. By leveraging the spatial-angular consistency inherent in light-field data, a differentiable forward model to generate a central view image from the light-field data is developed. Based on the forward model, an optimization scheme for the inverse problem of NDF reconstruction using differentiable propagation operators is established. Furthermore, an iterative solution method is adopted to reconstruct the NDF in the optimization scheme, which does not require training datasets and applies to light-field data captured by various acquisition methods. Experimental results demonstrate that high-quality NDF can be reconstructed from light-field data using the proposed method. High-resolution disparity can be effectively recovered by NDF, demonstrating its capability for the implicit, continuous representation of scene disparities.
Authors: Tom\'as Lima, Igor Varga, Eduard Bak\v{s}tein, Daniel Nov\'ak, Victor Alves
Abstract: Deep Brain Stimulation (DBS) is one of the most successful methods to diminish late-stage Parkinson's Disease (PD) symptoms. It is a delicate surgical procedure which requires detailed pre-surgical patient's study. High-field Magnetic Resonance Imaging (MRI) has proven its improved capacity of capturing the Subthalamic Nucleus (STN) - the main target of DBS in PD - in greater detail than lower field images. Here, we present a comparison between the performance of two different Deep Learning (DL) automatic segmentation architectures, one based in the registration to a brain template and the other performing the segmentation in in the MRI acquisition native space. The study was based on publicly available high-field 7 Tesla (T) brain MRI datasets of T1-weighted and T2-weighted sequences. nnUNet was used on the segmentation step of both architectures, while the data pre and post-processing pipelines diverged. The evaluation metrics showed that the performance of the segmentation directly in the native space yielded better results for the STN segmentation, despite not showing any advantage over the template-based method for the to other analysed structures: the Red Nucleus (RN) and the Substantia Nigra (SN).
Authors: Francisco Mena, Diego Arenas, Andreas Dengel
Abstract: Multi-sensor ML models for EO aim to enhance prediction accuracy by integrating data from various sources. However, the presence of missing data poses a significant challenge, particularly in non-persistent sensors that can be affected by external factors. Existing literature has explored strategies like temporal dropout and sensor-invariant models to address the generalization to missing data issues. Inspired by these works, we study two novel methods tailored for multi-sensor scenarios, namely Input Sensor Dropout (ISensD) and Ensemble Sensor Invariant (ESensI). Through experimentation on three multi-sensor temporal EO datasets, we demonstrate that these methods effectively increase the robustness of model predictions to missing sensors. Particularly, we focus on how the predictive performance of models drops when sensors are missing at different levels. We observe that ensemble multi-sensor models are the most robust to the lack of sensors. In addition, the sensor dropout component in ISensD shows promising robustness results.
Authors: Karim Kadry, Shreya Gupta, Jonas Sogbadji, Michiel Schaap, Kersten Petersen, Takuya Mizukami, Carlos Collet, Farhad R. Nezami, Elazer R. Edelman
Abstract: Virtual interventions enable the physics-based simulation of device deployment within coronary arteries. This framework allows for counterfactual reasoning by deploying the same device in different arterial anatomies. However, current methods to create such counterfactual arteries face a trade-off between controllability and realism. In this study, we investigate how Latent Diffusion Models (LDMs) can custom synthesize coronary anatomy for virtual intervention studies based on mid-level anatomic constraints such as topological validity, local morphological shape, and global skeletal structure. We also extend diffusion model guidance strategies to the context of morpho-skeletal conditioning and propose a novel guidance method for continuous attributes that adaptively updates the negative guiding condition throughout sampling. Our framework enables the generation and editing of coronary anatomy in a controllable manner, allowing device designers to derive mechanistic insights regarding anatomic variation and simulated device deployment.
Authors: Daxuan Ren, Haiyi Mei, Hezi Shi, Jianmin Zheng, Jianfei Cai, Lei Yang
Abstract: This paper presents a novel approach for the differentiable rendering of convex polyhedra, addressing the limitations of recent methods that rely on implicit field supervision. Our technique introduces a strategy that combines non-differentiable computation of hyperplane intersection through duality transform with differentiable optimization for vertex positioning with three-plane intersection, enabling gradient-based optimization without the need for 3D implicit fields. This allows for efficient shape representation across a range of applications, from shape parsing to compact mesh reconstruction. This work not only overcomes the challenges of previous approaches but also sets a new standard for representing shapes with convex polyhedra.
Authors: Ammar Ahmed, Abdul Manaf
Abstract: Wrist fractures are highly prevalent among children and can significantly impact their daily activities, such as attending school, participating in sports, and performing basic self-care tasks. If not treated properly, these fractures can result in chronic pain, reduced wrist functionality, and other long-term complications. Recently, advancements in object detection have shown promise in enhancing fracture detection, with systems achieving accuracy comparable to, or even surpassing, that of human radiologists. The YOLO series, in particular, has demonstrated notable success in this domain. This study is the first to provide a thorough evaluation of various YOLOv10 variants to assess their performance in detecting pediatric wrist fractures using the GRAZPEDWRI-DX dataset. It investigates how changes in model complexity, scaling the architecture, and implementing a dual-label assignment strategy can enhance detection performance. Experimental results indicate that our trained model achieved mean average precision (mAP@50-95) of 51.9\% surpassing the current YOLOv9 benchmark of 43.3\% on this dataset. This represents an improvement of 8.6\%. The implementation code is publicly available at https://github.com/ammarlodhi255/YOLOv10-Fracture-Detection
URLs: https://github.com/ammarlodhi255/YOLOv10-Fracture-Detection
Authors: Dimitrios Kollias, Anastasios Arsenos, James Wingate, Stefanos Kollias
Abstract: This paper presents a new approach for effective segmentation of images that can be integrated into any model and methodology; the paradigm that we choose is classification of medical images (3-D chest CT scans) for Covid-19 detection. Our approach includes a combination of vision-language models that segment the CT scans, which are then fed to a deep neural architecture, named RACNet, for Covid-19 detection. In particular, a novel framework, named SAM2CLIP2SAM, is introduced for segmentation that leverages the strengths of both Segment Anything Model (SAM) and Contrastive Language-Image Pre-Training (CLIP) to accurately segment the right and left lungs in CT scans, subsequently feeding these segmented outputs into RACNet for classification of COVID-19 and non-COVID-19 cases. At first, SAM produces multiple part-based segmentation masks for each slice in the CT scan; then CLIP selects only the masks that are associated with the regions of interest (ROIs), i.e., the right and left lungs; finally SAM is given these ROIs as prompts and generates the final segmentation mask for the lungs. Experiments are presented across two Covid-19 annotated databases which illustrate the improved performance obtained when our method has been used for segmentation of the CT scans.
Authors: Atefeh Khoshkhahtinat, Ali Zafari, Piyush M. Mehta, Nasser M. Nasrabadi, Barbara J. Thompson, Michael S. F. Kirk, Daniel da Silva
Abstract: NASA's Solar Dynamics Observatory (SDO) mission collects extensive data to monitor the Sun's daily activity. In the realm of space mission design, data compression plays a crucial role in addressing the challenges posed by limited telemetry rates. The primary objective of data compression is to facilitate efficient data management and transmission to work within the constrained bandwidth, thereby ensuring that essential information is captured while optimizing the utilization of available resources. This paper introduces a neural video compression technique that achieves a high compression ratio for the SDO's image data collection. The proposed approach focuses on leveraging both temporal and spatial redundancies in the data, leading to a more efficient compression. In this work, we introduce an architecture based on the Transformer model, which is specifically designed to capture both local and global information from input images in an effective and efficient manner. Additionally, our network is equipped with an entropy model that can accurately model the probability distribution of the latent representations and improves the speed of the entropy decoding step. The entropy model leverages a channel-dependent approach and utilizes checkerboard-shaped local and global spatial contexts. By combining the Transformer-based video compression network with our entropy model, the proposed compression algorithm demonstrates superior performance over traditional video codecs like H.264 and H.265, as confirmed by our experimental results.
Authors: Yongcan Yu, Lijun Sheng, Ran He, Jian Liang
Abstract: Test-time adaptation (TTA) aims to address the distribution shift between the training and test data with only unlabeled data at test time. Existing TTA methods often focus on improving recognition performance specifically for test data associated with classes in the training set. However, during the open-world inference process, there are inevitably test data instances from unknown classes, commonly referred to as outliers. This paper pays attention to the problem that conducts both sample recognition and outlier rejection during inference while outliers exist. To address this problem, we propose a new approach called STAble Memory rePlay (STAMP), which performs optimization over a stable memory bank instead of the risky mini-batch. In particular, the memory bank is dynamically updated by selecting low-entropy and label-consistent samples in a class-balanced manner. In addition, we develop a self-weighted entropy minimization strategy that assigns higher weight to low-entropy samples. Extensive results demonstrate that STAMP outperforms existing TTA methods in terms of both recognition and outlier detection performance. The code is released at https://github.com/yuyongcan/STAMP.
Authors: Reeshad Khan, Dr. John Gauch, Dr. Ukash Nakarmi
Abstract: The application of Deep Neural Networks (DNNs) to image denoising has notably challenged traditional denoising methods, particularly within complex noise scenarios prevalent in medical imaging. Despite the effectiveness of traditional and some DNN-based methods, their reliance on high-quality, noiseless ground truth images limits their practical utility. In response to this, our work introduces and benchmarks innovative unsupervised learning strategies, notably Stein's Unbiased Risk Estimator (SURE), its extension (eSURE), and our novel implementation, the Extended Poisson Unbiased Risk Estimator (ePURE), within medical imaging frameworks. This paper presents a comprehensive evaluation of these methods on MRI data afflicted with Gaussian and Poisson noise types, a scenario typical in medical imaging but challenging for most denoising algorithms. Our main contribution lies in the effective adaptation and implementation of the SURE, eSURE, and particularly the ePURE frameworks for medical images, showcasing their robustness and efficacy in environments where traditional noiseless ground truth cannot be obtained.
Authors: Zhecheng Yuan, Tianming Wei, Shuiqi Cheng, Gu Zhang, Yuanpei Chen, Huazhe Xu
Abstract: Can we endow visuomotor robots with generalization capabilities to operate in diverse open-world scenarios? In this paper, we propose \textbf{Maniwhere}, a generalizable framework tailored for visual reinforcement learning, enabling the trained robot policies to generalize across a combination of multiple visual disturbance types. Specifically, we introduce a multi-view representation learning approach fused with Spatial Transformer Network (STN) module to capture shared semantic information and correspondences among different viewpoints. In addition, we employ a curriculum-based randomization and augmentation approach to stabilize the RL training process and strengthen the visual generalization ability. To exhibit the effectiveness of Maniwhere, we meticulously design 8 tasks encompassing articulate objects, bi-manual, and dexterous hand manipulation tasks, demonstrating Maniwhere's strong visual generalization and sim2real transfer abilities across 3 hardware platforms. Our experiments show that Maniwhere significantly outperforms existing state-of-the-art methods. Videos are provided at https://gemcollector.github.io/maniwhere/.
Authors: Florian Robert, Alexia Calovoulos, Laurent Facq, Fanny Decoeur, Etienne Gontier, Christophe F. Grosset, Baudouin Denis de Senneville
Abstract: Accurately segmenting and individualizing cells in SEM images is a highly promising technique for elucidating tissue architecture in oncology. While current AI-based methods are effective, errors persist, necessitating time-consuming manual corrections, particularly in areas where the quality of cell contours in the image is poor and requires gap filling. This study presents a novel AI-driven approach for refining cell boundary delineation to improve instance-based cell segmentation in SEM images, also reducing the necessity for residual manual correction. A CNN COp-Net is introduced to address gaps in cell contours, effectively filling in regions with deficient or absent information. The network takes as input cell contour probability maps with potentially inadequate or missing information and outputs corrected cell contour delineations. The lack of training data was addressed by generating low integrity probability maps using a tailored PDE. We showcase the efficacy of our approach in augmenting cell boundary precision using both private SEM images from PDX hepatoblastoma tissues and publicly accessible images datasets. The proposed cell contour closing operator exhibits a notable improvement in tested datasets, achieving respectively close to 50% (private data) and 10% (public data) increase in the accurately-delineated cell proportion compared to state-of-the-art methods. Additionally, the need for manual corrections was significantly reduced, therefore facilitating the overall digitalization process. Our results demonstrate a notable enhancement in the accuracy of cell instance segmentation, particularly in highly challenging regions where image quality compromises the integrity of cell boundaries, necessitating gap filling. Therefore, our work should ultimately facilitate the study of tumour tissue bioarchitecture in onconanotomy field.
Authors: Yakir Oz, Gilad Yehudai, Gal Vardi, Itai Antebi, Michal Irani, Niv Haim
Abstract: Current methods for reconstructing training data from trained classifiers are restricted to very small models, limited training set sizes, and low-resolution images. Such restrictions hinder their applicability to real-world scenarios. In this paper, we present a novel approach enabling data reconstruction in realistic settings for models trained on high-resolution images. Our method adapts the reconstruction scheme of arXiv:2206.07758 to real-world scenarios -- specifically, targeting models trained via transfer learning over image embeddings of large pre-trained models like DINO-ViT and CLIP. Our work employs data reconstruction in the embedding space rather than in the image space, showcasing its applicability beyond visual data. Moreover, we introduce a novel clustering-based method to identify good reconstructions from thousands of candidates. This significantly improves on previous works that relied on knowledge of the training set to identify good reconstructed images. Our findings shed light on a potential privacy risk for data leakage from models trained using transfer learning.
Authors: Zahra Sadeghi
Abstract: How do humans distinguish between general categories of objects? In a number of se-mantic-category deficits, patients are good at making broad categorization but are un-able to remember fine and specific details. It has been well accepted that general in-formation about concepts is more robust to damages related to semantic memory. Re-sults from patients with semantic memory disorders demonstrate the loss of ability in subcategory recognition. In this paper, we review the behavioural evidence for cate-gory specific disorder and show that general categories of animal and plant are visual-ly distinguishable without processing textural information. To this aim, we utilize shape descriptors with an additional phase of feature learning. The results are evaluated with both supervised and unsupervised learning mechanisms and confirm that the proposed method can effectively discriminates between animal and plant object categories in visual domain.
Authors: Zongsheng Yue, Chen Change Loy
Abstract: While deep learning-based methods for blind face restoration have achieved unprecedented success, they still suffer from two major limitations. First, most of them deteriorate when facing complex degradations out of their training data. Second, these methods require multiple constraints, e.g., fidelity, perceptual, and adversarial losses, which require laborious hyper-parameter tuning to stabilize and balance their influences. In this work, we propose a novel method named DifFace that is capable of coping with unseen and complex degradations more gracefully without complicated loss designs. The key of our method is to establish a posterior distribution from the observed low-quality (LQ) image to its high-quality (HQ) counterpart. In particular, we design a transition distribution from the LQ image to the intermediate state of a pre-trained diffusion model and then gradually transmit from this intermediate state to the HQ target by recursively applying a pre-trained diffusion model. The transition distribution only relies on a restoration backbone that is trained with $L_2$ loss on some synthetic data, which favorably avoids the cumbersome training process in existing methods. Moreover, the transition distribution can contract the error of the restoration backbone and thus makes our method more robust to unknown degradations. Comprehensive experiments show that DifFace is superior to current state-of-the-art methods, especially in cases with severe degradations. Code and model are available at https://github.com/zsyOAOA/DifFace.
Authors: Indel Pal Singh, Enjie Ghorbel, Oyebade Oyedotun, Djamila Aouada
Abstract: This paper proposes an adaptive graph-based approach for multi-label image classification. Graph-based methods have been largely exploited in the field of multi-label classification, given their ability to model label correlations. Specifically, their effectiveness has been proven not only when considering a single domain but also when taking into account multiple domains. However, the topology of the used graph is not optimal as it is pre-defined heuristically. In addition, consecutive Graph Convolutional Network (GCN) aggregations tend to destroy the feature similarity. To overcome these issues, an architecture for learning the graph connectivity in an end-to-end fashion is introduced. This is done by integrating an attention-based mechanism and a similarity-preserving strategy. The proposed framework is then extended to multiple domains using an adversarial training scheme. Numerous experiments are reported on well-known single-domain and multi-domain benchmarks. The results demonstrate that our approach achieves competitive results in terms of mean Average Precision (mAP) and model size as compared to the state-of-the-art. The code will be made publicly available.
Authors: Sukrut Rao, Moritz B\"ohle, Bernt Schiele
Abstract: Deep neural networks are very successful on many vision tasks, but hard to interpret due to their black box nature. To overcome this, various post-hoc attribution methods have been proposed to identify image regions most influential to the models' decisions. Evaluating such methods is challenging since no ground truth attributions exist. We thus propose three novel evaluation schemes to more reliably measure the faithfulness of those methods, to make comparisons between them more fair, and to make visual inspection more systematic. To address faithfulness, we propose a novel evaluation setting (DiFull) in which we carefully control which parts of the input can influence the output in order to distinguish possible from impossible attributions. To address fairness, we note that different methods are applied at different layers, which skews any comparison, and so evaluate all methods on the same layers (ML-Att) and discuss how this impacts their performance on quantitative metrics. For more systematic visualizations, we propose a scheme (AggAtt) to qualitatively evaluate the methods on complete datasets. We use these evaluation schemes to study strengths and shortcomings of some widely used attribution methods over a wide range of models. Finally, we propose a post-processing smoothing step that significantly improves the performance of some attribution methods, and discuss its applicability.
Authors: Sukrut Rao, Moritz B\"ohle, Amin Parchami-Araghi, Bernt Schiele
Abstract: Despite being highly performant, deep neural networks might base their decisions on features that spuriously correlate with the provided labels, thus hurting generalization. To mitigate this, 'model guidance' has recently gained popularity, i.e. the idea of regularizing the models' explanations to ensure that they are "right for the right reasons". While various techniques to achieve such model guidance have been proposed, experimental validation of these approaches has thus far been limited to relatively simple and / or synthetic datasets. To better understand the effectiveness of the various design choices that have been explored in the context of model guidance, in this work we conduct an in-depth evaluation across various loss functions, attribution methods, models, and 'guidance depths' on the PASCAL VOC 2007 and MS COCO 2014 datasets. As annotation costs for model guidance can limit its applicability, we also place a particular focus on efficiency. Specifically, we guide the models via bounding box annotations, which are much cheaper to obtain than the commonly used segmentation masks, and evaluate the robustness of model guidance under limited (e.g. with only 1% of annotated images) or overly coarse annotations. Further, we propose using the EPG score as an additional evaluation metric and loss function ('Energy loss'). We show that optimizing for the Energy loss leads to models that exhibit a distinct focus on object-specific features, despite only using bounding box annotations that also include background regions. Lastly, we show that such model guidance can improve generalization under distribution shifts. Code available at: https://github.com/sukrutrao/Model-Guidance.
Authors: Tahsin Reasat, Asif Sushmit, David S. Smith
Abstract: Deep learning (DL) based diagnostics systems can provide accurate and robust quantitative analysis in digital pathology. These algorithms require large amounts of annotated training data which is impractical in pathology due to the high resolution of histopathological images. Hence, self-supervised methods have been proposed to learn features using ad-hoc pretext tasks. The self-supervised training process uses a large unlabeled dataset which makes the learning process time consuming. In this work, we propose a new method for actively sampling informative members from the training set using a small proxy network, decreasing sample requirement by 93% and training time by 62% while maintaining the same performance of the traditional self-supervised learning method. The code is available on https://github.com/Reasat/data_efficient_cl
Authors: Jaemin Cho, Linjie Li, Zhengyuan Yang, Zhe Gan, Lijuan Wang, Mohit Bansal
Abstract: Spatial control is a core capability in controllable image generation. Advancements in layout-guided image generation have shown promising results on in-distribution (ID) datasets with similar spatial configurations. However, it is unclear how these models perform when facing out-of-distribution (OOD) samples with arbitrary, unseen layouts. In this paper, we propose LayoutBench, a diagnostic benchmark for layout-guided image generation that examines four categories of spatial control skills: number, position, size, and shape. We benchmark two recent representative layout-guided image generation methods and observe that the good ID layout control may not generalize well to arbitrary layouts in the wild (e.g., objects at the boundary). Next, we propose IterInpaint, a new baseline that generates foreground and background regions step-by-step via inpainting, demonstrating stronger generalizability than existing models on OOD layouts in LayoutBench. We perform quantitative and qualitative evaluation and fine-grained analysis on the four LayoutBench skills to pinpoint the weaknesses of existing models. We show comprehensive ablation studies on IterInpaint, including training task ratio, crop&paste vs. repaint, and generation order. Lastly, we evaluate the zero-shot performance of different pretrained layout-guided image generation models on LayoutBench-COCO, our new benchmark for OOD layouts with real objects, where our IterInpaint consistently outperforms SOTA baselines in all four splits. Project website: https://layoutbench.github.io
Authors: Artur Jesslen, Guofeng Zhang, Angtian Wang, Wufei Ma, Alan Yuille, Adam Kortylewski
Abstract: Discriminative models for object classification typically learn image-based representations that do not capture the compositional and 3D nature of objects. In this work, we show that explicitly integrating 3D compositional object representations into deep networks for image classification leads to a largely enhanced generalization in out-of-distribution scenarios. In particular, we introduce a novel architecture, referred to as \OURS, that consists of a feature extractor and a \textit{neural object volume} for every target object class. Each neural object volume is a composition of 3D Gaussians that emit feature vectors. This compositional object representation allows for a highly robust and fast estimation of the object class by independently matching the features of the 3D Gaussians of each category to features extracted from an input image. Additionally, the object pose can be estimated via inverse rendering of the corresponding neural object volume. To enable the classification of objects, the neural features at each 3D Gaussian are trained discriminatively to be distinct from (i) the features of 3D Gaussians in other categories, (ii) features of other 3D Gaussians of the same object, and (iii) the background features. Our experiments show that \OURS offers intriguing advantages over standard architectures due to the 3D compositional structure of the object representation, namely: (1) An exceptional robustness across a spectrum of real-world and synthetic out-of-distribution shifts and (2) an enhanced human interpretability compared to standard models, all while maintaining real-time inference and a competitive accuracy on in-distribution data.
Authors: Christos Sakaridis, David Bruggemann, Fisher Yu, Luc Van Gool
Abstract: Adaptation of semantic segmentation networks to different visual conditions is vital for robust perception in autonomous cars and robots. However, previous work has shown that most feature-level adaptation methods, which employ adversarial training and are validated on synthetic-to-real adaptation, provide marginal gains in condition-level adaptation, being outperformed by simple pixel-level adaptation via stylization. Motivated by these findings, we propose to leverage stylization in performing feature-level adaptation by aligning the internal network features extracted by the encoder of the network from the original and the stylized view of each input image with a novel feature invariance loss. In this way, we encourage the encoder to extract features that are already invariant to the style of the input, allowing the decoder to focus on parsing these features and not on further abstracting from the specific style of the input. We implement our method, named Condition-Invariant Semantic Segmentation (CISS), on the current state-of-the-art domain adaptation architecture and achieve outstanding results on condition-level adaptation. In particular, CISS sets the new state of the art in the popular daytime-to-nighttime Cityscapes$\to$Dark Zurich benchmark. Furthermore, our method achieves the second-best performance on the normal-to-adverse Cityscapes$\to$ACDC benchmark. CISS is shown to generalize well to domains unseen during training, such as BDD100K-night and ACDC-night. Code is publicly available at https://github.com/SysCV/CISS .
Authors: Giorgos Sfikas, George Retsinas
Abstract: We study matrix forms of quaternionic versions of the Fourier Transform and Convolution operations. Quaternions offer a powerful representation unit, however they are related to difficulties in their use that stem foremost from non-commutativity of quaternion multiplication, and due to that $\mu^2 = -1$ possesses infinite solutions in the quaternion domain. Handling of quaternionic matrices is consequently complicated in several aspects (definition of eigenstructure, determinant, etc.). Our research findings clarify the relation of the Quaternion Fourier Transform matrix to the standard (complex) Discrete Fourier Transform matrix, and the extend on which well-known complex-domain theorems extend to quaternions. We focus especially on the relation of Quaternion Fourier Transform matrices to Quaternion Circulant matrices (representing quaternionic convolution), and the eigenstructure of the latter. A proof-of-concept application that makes direct use of our theoretical results is presented, where we present a method to bound the Lipschitz constant of a Quaternionic Convolutional Neural Network. Code is publicly available at: \url{https://github.com/sfikas/quaternion-fourier-convolution-matrix}.
URLs: https://github.com/sfikas/quaternion-fourier-convolution-matrix
Authors: Kun Yuan, Vinkle Srivastav, Tong Yu, Joel L. Lavanchy, Pietro Mascagni, Nassir Navab, Nicolas Padoy
Abstract: Recent advancements in surgical computer vision have been driven by vision-only models, which lack language semantics, relying on manually annotated videos to predict fixed object categories. This limits their generalizability to unseen surgical procedures and tasks. We propose leveraging surgical video lectures from e-learning platforms to provide effective vision and language supervisory signals for multi-modal representation learning, bypassing manual annotations. We address surgery-specific linguistic challenges using multiple automatic speech recognition systems for text transcriptions. We introduce SurgVLP - Surgical Vision Language Pre-training - a novel method for multi-modal representation learning. SurgVLP employs a new contrastive learning objective, aligning video clip embeddings with corresponding multiple text embeddings in a joint latent space. We demonstrate the representational capability of this space through several vision-and-language surgical tasks and vision-only tasks specific to surgery. Unlike current fully supervised approaches, SurgVLP adapts to different surgical procedures and tasks without specific fine-tuning, achieving zero-shot adaptation to tasks such as surgical tool, phase, and triplet recognition without manual annotation. These results highlight the transferability and versatility of the learned multi-modal representations in surgical video analysis. The code is available at https://github.com/CAMMA-public/SurgVLP
Authors: Yuzheng Wang, Zhaoyu Chen, Jie Zhang, Dingkang Yang, Zuhao Ge, Yang Liu, Siao Liu, Yunquan Sun, Wenqiang Zhang, Lizhe Qi
Abstract: Data-Free Knowledge Distillation (DFKD) is a novel task that aims to train high-performance student models using only the pre-trained teacher network without original training data. Most of the existing DFKD methods rely heavily on additional generation modules to synthesize the substitution data resulting in high computational costs and ignoring the massive amounts of easily accessible, low-cost, unlabeled open-world data. Meanwhile, existing methods ignore the domain shift issue between the substitution data and the original data, resulting in knowledge from teachers not always trustworthy and structured knowledge from data becoming a crucial supplement. To tackle the issue, we propose a novel Open-world Data Sampling Distillation (ODSD) method for the DFKD task without the redundant generation process. First, we try to sample open-world data close to the original data's distribution by an adaptive sampling module and introduce a low-noise representation to alleviate the domain shift issue. Then, we build structured relationships of multiple data examples to exploit data knowledge through the student model itself and the teacher's structured representation. Extensive experiments on CIFAR-10, CIFAR-100, NYUv2, and ImageNet show that our ODSD method achieves state-of-the-art performance with lower FLOPs and parameters. Especially, we improve 1.50\%-9.59\% accuracy on the ImageNet dataset and avoid training the separate generator for each class.
Authors: Xiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng, Xihui Liu, Kaicheng Yu, Hengshuang Zhao
Abstract: The rapid advancement of deep learning models often attributes to their ability to leverage massive training data. In contrast, such privilege has not yet fully benefited 3D deep learning, mainly due to the limited availability of large-scale 3D datasets. Merging multiple available data sources and letting them collaboratively train a single model is a potential solution. However, due to the large domain gap between 3D point cloud datasets, such mixed supervision could adversely affect the model's performance and lead to degenerated performance (i.e., negative transfer) compared to single-dataset training. In view of this challenge, we introduce Point Prompt Training (PPT), a novel framework for multi-dataset synergistic learning in the context of 3D representation learning that supports multiple pre-training paradigms. Based on this framework, we propose Prompt-driven Normalization, which adapts the model to different datasets with domain-specific prompts and Language-guided Categorical Alignment that decently unifies the multiple-dataset label spaces by leveraging the relationship between label text. Extensive experiments verify that PPT can overcome the negative transfer associated with synergistic learning and produce generalizable representations. Notably, it achieves state-of-the-art performance on each dataset using a single weight-shared model with supervised multi-dataset training. Moreover, when served as a pre-training framework, it outperforms other pre-training approaches regarding representation quality and attains remarkable state-of-the-art performance across over ten diverse downstream tasks spanning both indoor and outdoor 3D scenarios.
Authors: Floris Erich, Naoya Chiba, Yusuke Yoshiyasu, Noriaki Ando, Ryo Hanai, Yukiyasu Domae
Abstract: We present NeuralLabeling, a labeling approach and toolset for annotating 3D scenes using either bounding boxes or meshes and generating segmentation masks, affordance maps, 2D bounding boxes, 3D bounding boxes, 6DOF object poses, depth maps, and object meshes. NeuralLabeling uses Neural Radiance Fields (NeRF) as a renderer, allowing labeling to be performed using 3D spatial tools while incorporating geometric clues such as occlusions, relying only on images captured from multiple viewpoints as input. To demonstrate the applicability of NeuralLabeling to a practical problem in robotics, we added ground truth depth maps to 30000 frames of transparent object RGB and noisy depth maps of glasses placed in a dishwasher captured using an RGBD sensor, yielding the Dishwasher30k dataset. We show that training a simple deep neural network with supervision using the annotated depth maps yields a higher reconstruction performance than training with the previously applied weakly supervised approach. We also show how instance segmentation and depth completion datasets generated using NeuralLabeling can be incorporated into a robot application for grasping transparent objects placed in a dishwasher with an accuracy of 83.3%, compared to 16.3% without depth completion.
Authors: Xiaofan Li, Bo Peng, Jie Hu, Changyou Ma, Daipeng Yang, Zhuyang Xie
Abstract: Unsupervised skin lesion segmentation offers several benefits, including conserving expert human resources, reducing discrepancies due to subjective human labeling, and adapting to novel environments. However, segmenting dermoscopic images without manual labeling guidance presents significant challenges due to dermoscopic image artifacts such as hair noise, blister noise, and subtle edge differences. To address these challenges, we introduce an innovative Uncertainty Self-Learning Network (USL-Net) designed for skin lesion segmentation. The USL-Net can effectively segment a range of lesions, eliminating the need for manual labeling guidance. Initially, features are extracted using contrastive learning, followed by the generation of Class Activation Maps (CAMs) as saliency maps using these features. The different CAM locations correspond to the importance of the lesion region based on their saliency. High-saliency regions in the map serve as pseudo-labels for lesion regions while low-saliency regions represent the background. However, intermediate regions can be hard to classify, often due to their proximity to lesion edges or interference from hair or blisters. Rather than risk potential pseudo-labeling errors or learning confusion by forcefully classifying these regions, we consider them as uncertainty regions, exempting them from pseudo-labeling and allowing the network to self-learn. Further, we employ connectivity detection and centrality detection to refine foreground pseudo-labels and reduce noise-induced errors. The application of cycle refining enhances performance further. Our method underwent thorough experimental validation on the ISIC-2017, ISIC-2018, and PH2 datasets, demonstrating that its performance is on par with weakly supervised and supervised methods, and exceeds that of other existing unsupervised methods.
Authors: Eman Ali, Muhammad Haris Khan
Abstract: Recent advances in large-scale vision-language models have achieved impressive performance in various zero-shot image classification tasks. While prior studies have demonstrated significant improvements by introducing few-shot labelled target samples, they still require labelling of target samples, which greatly degrades their scalability and generalizability while handling various visual recognition tasks. We design NtUA, a Noise-tolerant Unsupervised Adapter that allows the learning of effective target models with few unlabelled target samples. NtUA works as a key-value cache that formulates visual features and predicted pseudo-labels of the few unlabelled target samples as key-value pairs. It consists of two complementary designs. The first is adaptive cache formation that combats pseudo-label noises by weighting the key-value pairs according to their prediction confidence. The second is knowledge-guided cache refinement, which refines pair values (i.e., pseudo-labels) and cache weights by leveraging knowledge distillation from large-scale vision language models. Extensive experiments show that NtUA achieves superior performance consistently across multiple widely adopted benchmarks.
Authors: Albert Mohwald, Tomas Jenicek, Ond\v{r}ej Chum
Abstract: Image retrieval methods based on CNN descriptors rely on metric learning from a large number of diverse examples of positive and negative image pairs. Domains, such as night-time images, with limited availability and variability of training data suffer from poor retrieval performance even with methods performing well on standard benchmarks. We propose to train a GAN-based synthetic-image generator, translating available day-time image examples into night images. Such a generator is used in metric learning as a form of augmentation, supplying training data to the scarce domain. Various types of generators are evaluated and analyzed. We contribute with a novel light-weight GAN architecture that enforces the consistency between the original and translated image through edge consistency. The proposed architecture also allows a simultaneous training of an edge detector that operates on both night and day images. To further increase the variability in the training examples and to maximize the generalization of the trained model, we propose a novel method of diverse anchor mining. The proposed method improves over the state-of-the-art results on a standard Tokyo 24/7 day-night retrieval benchmark while preserving the performance on Oxford and Paris datasets. This is achieved without the need of training image pairs of matching day and night images. The source code is available at https://github.com/mohwald/gandtr .
Authors: Yangchao Wu, Tian Yu Liu, Hyoungseob Park, Stefano Soatto, Dong Lao, Alex Wong
Abstract: Unsupervised depth completion and estimation methods are trained by minimizing reconstruction error. Block artifacts from resampling, intensity saturation, and occlusions are amongst the many undesirable by-products of common data augmentation schemes that affect image reconstruction quality, and thus the training signal. Hence, typical augmentations on images viewed as essential to training pipelines in other vision tasks have seen limited use beyond small image intensity changes and flipping. The sparse depth modality in depth completion have seen even less use as intensity transformations alter the scale of the 3D scene, and geometric transformations may decimate the sparse points during resampling. We propose a method that unlocks a wide range of previously-infeasible geometric augmentations for unsupervised depth completion and estimation. This is achieved by reversing, or ``undo''-ing, geometric transformations to the coordinates of the output depth, warping the depth map back to the original reference frame. This enables computing the reconstruction losses using the original images and sparse depth maps, eliminating the pitfalls of naive loss computation on the augmented inputs and allowing us to scale up augmentations to boost performance. We demonstrate our method on indoor (VOID) and outdoor (KITTI) datasets, where we consistently improve upon recent methods across both datasets as well as generalization to four other datasets. Code available at: https://github.com/alexklwong/augundo.
Authors: Siyu Zhang, Yeming Chen, Yaoru Sun, Fang Wang, Jun Yang, Lizhi Bai, Shangce Gao
Abstract: The key to integrating visual language tasks is to establish a good alignment strategy. Recently, visual semantic representation has achieved fine-grained visual understanding by dividing grids or image patches. However, the coarse-grained semantic interactions in image space should not be ignored, which hinders the extraction of complex contextual semantic relations at the scene boundaries. This paper proposes superpixels as comprehensive and robust visual primitives, which mine coarse-grained semantic interactions by clustering perceptually similar pixels, speeding up the subsequent processing of primitives. To capture superpixel-level semantic features, we propose a Multiscale Difference Graph Convolutional Network (MDGCN). It allows parsing the entire image as a fine-to-coarse visual hierarchy. To reason actual semantic relations, we reduce potential noise interference by aggregating difference information between adjacent graph nodes. Finally, we propose a multi-level fusion rule in a bottom-up manner to avoid understanding deviation by mining complementary spatial information at different levels. Experiments show that the proposed method can effectively promote the learning of multiple downstream tasks. Encouragingly, our method outperforms previous methods on all metrics. Our code will be released upon publication.
Authors: Bo Yuan, Danpei Zhao
Abstract: Continual learning, also known as incremental learning or life-long learning, stands at the forefront of deep learning and AI systems. It breaks through the obstacle of one-way training on close sets and enables continuous adaptive learning on open-set conditions. In the recent decade, continual learning has been explored and applied in multiple fields especially in computer vision covering classification, detection and segmentation tasks. Continual semantic segmentation (CSS), of which the dense prediction peculiarity makes it a challenging, intricate and burgeoning task. In this paper, we present a review of CSS, committing to building a comprehensive survey on problem formulations, primary challenges, universal datasets, neoteric theories and multifarious applications. Concretely, we begin by elucidating the problem definitions and primary challenges. Based on an in-depth investigation of relevant approaches, we sort out and categorize current CSS models into two main branches including data-replay and data-free sets. In each branch, the corresponding approaches are similarity-based clustered and thoroughly analyzed, following qualitative comparison and quantitative reproductions on relevant datasets. Besides, we also introduce four CSS specialities with diverse application scenarios and development tendencies. Furthermore, we develop a benchmark for CSS encompassing representative references, evaluation results and reproductions, which is available at~\url{https://github.com/YBIO/SurveyCSS}. We hope this survey can serve as a reference-worthy and stimulating contribution to the advancement of the life-long learning field, while also providing valuable perspectives for related fields.
Authors: Yingwen Wu, Tao Li, Xinwen Cheng, Jie Yang, Xiaolin Huang
Abstract: Detecting out-of-distribution (OOD) samples is essential for ensuring the reliability of deep neural networks (DNNs) in real-world scenarios. While previous research has predominantly investigated the disparity between in-distribution (ID) and OOD data through forward information analysis, the discrepancy in parameter gradients during the backward process of DNNs has received insufficient attention. Existing studies on gradient disparities mainly focus on the utilization of gradient norms, neglecting the wealth of information embedded in gradient directions. To bridge this gap, in this paper, we conduct a comprehensive investigation into leveraging the entirety of gradient information for OOD detection. The primary challenge arises from the high dimensionality of gradients due to the large number of network parameters. To solve this problem, we propose performing linear dimension reduction on the gradient using a designated subspace that comprises principal components. This innovative technique enables us to obtain a low-dimensional representation of the gradient with minimal information loss. Subsequently, by integrating the reduced gradient with various existing detection score functions, our approach demonstrates superior performance across a wide range of detection tasks. For instance, on the ImageNet benchmark with ResNet50 model, our method achieves an average reduction of 11.15$\%$ in the false positive rate at 95$\%$ recall (FPR95) compared to the current state-of-the-art approach. The code would be released.
Authors: Zhili Chen, Maosheng Ye, Shuangjie Xu, Tongyi Cao, Qifeng Chen
Abstract: We present a new interaction mechanism of prediction and planning for end-to-end autonomous driving, called PPAD (Iterative Interaction of Prediction and Planning Autonomous Driving), which considers the timestep-wise interaction to better integrate prediction and planning. An ego vehicle performs motion planning at each timestep based on the trajectory prediction of surrounding agents (e.g., vehicles and pedestrians) and its local road conditions. Unlike existing end-to-end autonomous driving frameworks, PPAD models the interactions among ego, agents, and the dynamic environment in an autoregressive manner by interleaving the Prediction and Planning processes at every timestep, instead of a single sequential process of prediction followed by planning. Specifically, we design ego-to-agent, ego-to-map, and ego-to-BEV interaction mechanisms with hierarchical dynamic key objects attention to better model the interactions. The experiments on the nuScenes benchmark show that our approach outperforms state-of-the-art methods.
Authors: Yitao Zhu, Zhenrong Shen, Zihao Zhao, Sheng Wang, Xin Wang, Xiangyu Zhao, Dinggang Shen, Qian Wang
Abstract: The common practice in developing computer-aided diagnosis (CAD) models based on transformer architectures usually involves fine-tuning from ImageNet pre-trained weights. However, with recent advances in large-scale pre-training and the practice of scaling laws, Vision Transformers (ViT) have become much larger and less accessible to medical imaging communities. Additionally, in real-world scenarios, the deployments of multiple CAD models can be troublesome due to problems such as limited storage space and time-consuming model switching. To address these challenges, we propose a new method MeLo (Medical image Low-rank adaptation), which enables the development of a single CAD model for multiple clinical tasks in a lightweight manner. It adopts low-rank adaptation instead of resource-demanding fine-tuning. By fixing the weight of ViT models and only adding small low-rank plug-ins, we achieve competitive results on various diagnosis tasks across different imaging modalities using only a few trainable parameters. Specifically, our proposed method achieves comparable performance to fully fine-tuned ViT models on four distinct medical imaging datasets using about 0.17% trainable parameters. Moreover, MeLo adds only about 0.5MB of storage space and allows for extremely fast model switching in deployment and inference. Our source code and pre-trained weights are available on our website (https://absterzhu.github.io/melo.github.io/).
Authors: Jingnan Gao, Zhuo Chen, Yichao Yan, Bowen Pan, Zhe Wang, Jiangjing Lyu, Xiaokang Yang
Abstract: Reconstructing real-world 3D objects has numerous applications in computer vision, such as virtual reality, video games, and animations. Ideally, 3D reconstruction methods should generate high-fidelity results with 3D consistency in real-time. Traditional methods match pixels between images using photo-consistency constraints or learned features, while differentiable rendering methods like Neural Radiance Fields (NeRF) use differentiable volume rendering or surface-based representation to generate high-fidelity scenes. However, these methods require excessive runtime for rendering, making them impractical for daily applications. To address these challenges, we present $\textbf{EvaSurf}$, an $\textbf{E}$fficient $\textbf{V}$iew-$\textbf{A}$ware implicit textured $\textbf{Surf}$ace reconstruction method on mobile devices. In our method, we first employ an efficient surface-based model with a multi-view supervision module to ensure accurate mesh reconstruction. To enable high-fidelity rendering, we learn an implicit texture embedded with a set of Gaussian lobes to capture view-dependent information. Furthermore, with the explicit geometry and the implicit texture, we can employ a lightweight neural shader to reduce the expense of computation and further support real-time rendering on common mobile devices. Extensive experiments demonstrate that our method can reconstruct high-quality appearance and accurate mesh on both synthetic and real-world datasets. Moreover, our method can be trained in just 1-2 hours using a single GPU and run on mobile devices at over 40 FPS (Frames Per Second), with a final package required for rendering taking up only 40-50 MB.
Authors: Xin Zhang, Yingze Song, Tingting Song, Degang Yang, Yichen Ye, Jie Zhou, Liming Zhang
Abstract: Neural networks based on convolutional operations have achieved remarkable results in the field of deep learning, but there are two inherent flaws in standard convolutional operations. On the one hand, the convolution operation is confined to a local window, so it cannot capture information from other locations, and its sampled shapes is fixed. On the other hand, the size of the convolutional kernel are fixed to k $\times$ k, which is a fixed square shape, and the number of parameters tends to grow squarely with size. Although Deformable Convolution (Deformable Conv) address the problem of fixed sampling of standard convolutions, the number of parameters also tends to grow in a squared manner. In response to the above questions, the Linear Deformable Convolution (LDConv) is explored in this work, which gives the convolution kernel an arbitrary number of parameters and arbitrary sampled shapes to provide richer options for the trade-off between network overhead and performance. In LDConv, a novel coordinate generation algorithm is defined to generate different initial sampled positions for convolutional kernels of arbitrary size. To adapt to changing targets, offsets are introduced to adjust the shape of the samples at each position. LDConv corrects the growth trend of the number of parameters for standard convolution and Deformable Conv to a linear growth. Moreover, it completes the process of efficient feature extraction by irregular convolutional operations and brings more exploration options for convolutional sampled shapes. Object detection experiments on representative datasets COCO2017, VOC 7+12, and VisDrone-DET2021 fully demonstrate the advantages of LDConv. LDConv is a plug-and-play convolutional operation that can replace the convolutional operation to improve network performance. The code for the relevant tasks can be found at https://github.com/CV-ZhangXin/LDConv.
Authors: Haoran Li, Long Ma, Yong Liao, Lechao Cheng, Yanbin Hao, Pengyuan Zhou
Abstract: The current GAN inversion methods typically can only edit the appearance and shape of a single object and background while overlooking spatial information. In this work, we propose a 3D editing framework, 3D-GOI, to enable multifaceted editing of affine information (scale, translation, and rotation) on multiple objects. 3D-GOI realizes the complex editing function by inverting the abundance of attribute codes (object shape/appearance/scale/rotation/translation, background shape/appearance, and camera pose) controlled by GIRAFFE, a renowned 3D GAN. Accurately inverting all the codes is challenging, 3D-GOI solves this challenge following three main steps. First, we segment the objects and the background in a multi-object image. Second, we use a custom Neural Inversion Encoder to obtain coarse codes of each object. Finally, we use a round-robin optimization algorithm to get precise codes to reconstruct the image. To the best of our knowledge, 3D-GOI is the first framework to enable multifaceted editing on multiple objects. Both qualitative and quantitative experiments demonstrate that 3D-GOI holds immense potential for flexible, multifaceted editing in complex multi-object scenes.Our project and code are released at https://3d-goi.github.io .
URLs: https://3d-goi.github.io
Authors: Lingchen Meng, Shiyi Lan, Hengduo Li, Jose M. Alvarez, Zuxuan Wu, Yu-Gang Jiang
Abstract: In-context segmentation aims at segmenting novel images using a few labeled example images, termed as "in-context examples", exploring content similarities between examples and the target. The resulting models can be generalized seamlessly to novel segmentation tasks, significantly reducing the labeling and training costs compared with conventional pipelines. However, in-context segmentation is more challenging than classic ones requiring the model to learn segmentation rules conditioned on a few samples. Unlike previous work with ad-hoc or non-end-to-end designs, we propose SEGIC, an end-to-end segment-in-context framework built upon a single vision foundation model (VFM). In particular, SEGIC leverages the emergent correspondence within VFM to capture dense relationships between target images and in-context samples. As such, information from in-context samples is then extracted into three types of instructions, i.e. geometric, visual, and meta instructions, serving as explicit conditions for the final mask prediction. SEGIC is a straightforward yet effective approach that yields state-of-the-art performance on one-shot segmentation benchmarks. Notably, SEGIC can be easily generalized to diverse tasks, including video object segmentation and open-vocabulary segmentation. Code will be available at https://github.com/MengLcool/SEGIC.
Authors: Anil Batra, Davide Moltisanti, Laura Sevilla-Lara, Marcus Rohrbach, Frank Keller
Abstract: Procedural videos, exemplified by recipe demonstrations, are instrumental in conveying step-by-step instructions. However, understanding such videos is challenging as it involves the precise localization of steps and the generation of textual instructions. Manually annotating steps and writing instructions is costly, which limits the size of current datasets and hinders effective learning. Leveraging large but noisy video-transcript datasets for pre-training can boost performance but demands significant computational resources. Furthermore, transcripts contain irrelevant content and differ in style from human-written instructions. To mitigate these issues, we propose a novel technique, Sieve-&-Swap, to automatically generate high-quality training data for the recipe domain: (i) Sieve: filters irrelevant transcripts and (ii) Swap: acquires high-quality text by replacing transcripts with human-written instruction from a text-only recipe dataset. The resulting dataset is three orders of magnitude smaller than current web-scale datasets but enables efficient training of large-scale models. Alongside Sieve-&-Swap, we propose Procedure Transformer (ProcX), a model for end-to-end step localization and instruction generation for procedural videos. When pre-trained on our curated dataset, this model achieves state-of-the-art performance on YouCook2 and Tasty while using a fraction of the training data. We have released code and dataset.
Authors: Yiyang Luo, Ke Lin, Chao Gu
Abstract: Indoor scene modification has emerged as a prominent area within computer vision, particularly for its applications in Augmented Reality (AR) and Virtual Reality (VR). Traditional methods often rely on pre-existing object databases and predetermined object positions, limiting their flexibility and adaptability to new scenarios. In response to this challenge, we present a novel end-to-end multi-modal deep neural network capable of generating point cloud objects seamlessly integrated with their surroundings, driven by textual instructions. Our model revolutionizes scene modification by enabling the creation of new environments with previously unseen object layouts, eliminating the need for pre-stored CAD models. Leveraging Point-E as our generative model, we introduce innovative techniques such as quantized position prediction and Top-K estimation to address the issue of false negatives resulting from ambiguous language descriptions. Furthermore, we conduct comprehensive evaluations to showcase the diversity of generated objects, the efficacy of textual instructions, and the quantitative metrics, affirming the realism and versatility of our model in generating indoor objects. To provide a holistic assessment, we incorporate visual grounding as an additional metric, ensuring the quality and coherence of the scenes produced by our model. Through these advancements, our approach not only advances the state-of-the-art in indoor scene modification but also lays the foundation for future innovations in immersive computing and digital environment creation.
Authors: Quan Dao, Binh Ta, Tung Pham, Anh Tran
Abstract: Developing image-generative models, which are robust to outliers in the training process, has recently drawn attention from the research community. Due to the ease of integrating unbalanced optimal transport (UOT) into adversarial framework, existing works focus mainly on developing robust frameworks for generative adversarial model (GAN). Meanwhile, diffusion models have recently dominated GAN in various tasks and datasets. However, according to our knowledge, none of them are robust to corrupted datasets. Motivated by DDGAN, our work introduces the first robust-to-outlier diffusion. We suggest replacing the UOT-based generative model for GAN in DDGAN to learn the backward diffusion process. Additionally, we demonstrate that the Lipschitz property of divergence in our framework contributes to more stable training convergence. Remarkably, our method not only exhibits robustness to corrupted datasets but also achieves superior performance on clean datasets.
Authors: Ming Nie, Renyuan Peng, Chunwei Wang, Xinyue Cai, Jianhua Han, Hang Xu, Li Zhang
Abstract: Large vision-language models (VLMs) have garnered increasing interest in autonomous driving areas, due to their advanced capabilities in complex reasoning tasks essential for highly autonomous vehicle behavior. Despite their potential, research in autonomous systems is hindered by the lack of datasets with annotated reasoning chains that explain the decision-making processes in driving. To bridge this gap, we present Reason2Drive, a benchmark dataset with over 600K video-text pairs, aimed at facilitating the study of interpretable reasoning in complex driving environments. We distinctly characterize the autonomous driving process as a sequential combination of perception, prediction, and reasoning steps, and the question-answer pairs are automatically collected from a diverse range of open-source outdoor driving datasets, including nuScenes, Waymo and ONCE. Moreover, we introduce a novel aggregated evaluation metric to assess chain-based reasoning performance in autonomous systems, addressing the semantic ambiguities of existing metrics such as BLEU and CIDEr. Based on the proposed benchmark, we conduct experiments to assess various existing VLMs, revealing insights into their reasoning capabilities. Additionally, we develop an efficient approach to empower VLMs to leverage object-level perceptual elements in both feature extraction and prediction, further enhancing their reasoning accuracy. The code and dataset will be released.
Authors: Thomas Westfechtel, Dexuan Zhang, Tatsuya Harada
Abstract: Unsupervised domain adaptation (UDA) tries to overcome the tedious work of labeling data by leveraging a labeled source dataset and transferring its knowledge to a similar but different target dataset. Meanwhile, current vision-language models exhibit remarkable zero-shot prediction capabilities. In this work, we combine knowledge gained through UDA with the inherent knowledge of vision-language models. We introduce a strong-weak guidance learning scheme that employs zero-shot predictions to help align the source and target dataset. For the strong guidance, we expand the source dataset with the most confident samples of the target dataset. Additionally, we employ a knowledge distillation loss as weak guidance. The strong guidance uses hard labels but is only applied to the most confident predictions from the target dataset. Conversely, the weak guidance is employed to the whole dataset but uses soft labels. The weak guidance is implemented as a knowledge distillation loss with (shifted) zero-shot predictions. We show that our method complements and benefits from prompt adaptation techniques for vision-language models. We conduct experiments and ablation studies on three benchmarks (OfficeHome, VisDA, and DomainNet), outperforming state-of-the-art methods. Our ablation studies further demonstrate the contributions of different components of our algorithm.
Authors: Yi Wang, Jian Ma, Ruizhi Shao, Qiao Feng, Yu-Kun Lai, Yebin Liu, Kun Li
Abstract: The generation of 3D clothed humans has attracted increasing attention in recent years. However, existing work cannot generate layered high-quality 3D humans with consistent body structures. As a result, these methods are unable to arbitrarily and separately change and edit the body and clothing of the human. In this paper, we propose a text-driven layered 3D human generation framework based on a novel physically-decoupled semantic-aware diffusion model. To keep the generated clothing consistent with the target text, we propose a semantic-confidence strategy for clothing that can eliminate the non-clothing content generated by the model. To match the clothing with different body shapes, we propose a SMPL-driven implicit field deformation network that enables the free transfer and reuse of clothing. Besides, we introduce uniform shape priors based on the SMPL model for body and clothing, respectively, which generates more diverse 3D content without being constrained by specific templates. The experimental results demonstrate that the proposed method not only generates 3D humans with consistent body structures but also allows free editing in a layered manner. The source code will be made public.
Authors: Rahul Venkatesh, Honglin Chen, Kevin Feigelis, Daniel M. Bear, Khaled Jedoui, Klemen Kotar, Felix Binder, Wanhee Lee, Sherry Liu, Kevin A. Smith, Judith E. Fan, Daniel L. K. Yamins
Abstract: The ability to understand physical dynamics is critical for agents to act in the world. Here, we use Counterfactual World Modeling (CWM) to extract vision structures for dynamics understanding. CWM uses a temporally-factored masking policy for masked prediction of video data without annotations. This policy enables highly effective "counterfactual prompting" of the predictor, allowing a spectrum of visual structures to be extracted from a single pre-trained predictor without finetuning on annotated datasets. We demonstrate that these structures are useful for physical dynamics understanding, allowing CWM to achieve the state-of-the-art performance on the Physion benchmark.
Authors: Weijie Wei, Fatemeh Karimi Nejadasl, Theo Gevers, Martin R. Oswald
Abstract: The scarcity of annotated data in LiDAR point cloud understanding hinders effective representation learning. Consequently, scholars have been actively investigating efficacious self-supervised pre-training paradigms. Nevertheless, temporal information, which is inherent in the LiDAR point cloud sequence, is consistently disregarded. To better utilize this property, we propose an effective pre-training strategy, namely Temporal Masked Auto-Encoders (T-MAE), which takes as input temporally adjacent frames and learns temporal dependency. A SiamWCA backbone, containing a Siamese encoder and a windowed cross-attention (WCA) module, is established for the two-frame input. Considering that the movement of an ego-vehicle alters the view of the same instance, temporal modeling also serves as a robust and natural data augmentation, enhancing the comprehension of target objects. SiamWCA is a powerful architecture but heavily relies on annotated data. Our T-MAE pre-training strategy alleviates its demand for annotated data. Comprehensive experiments demonstrate that T-MAE achieves the best performance on both Waymo and ONCE datasets among competitive self-supervised approaches. Codes will be released at https://github.com/codename1995/T-MAE
Authors: Riccardo Marin, Enric Corona, Gerard Pons-Moll
Abstract: Aligning a template to 3D human point clouds is a long-standing problem crucial for tasks like animation, reconstruction, and enabling supervised learning pipelines. Recent data-driven methods leverage predicted surface correspondences. However, they are not robust to varied poses, identities, or noise. In contrast, industrial solutions often rely on expensive manual annotations or multi-view capturing systems. Recently, neural fields have shown promising results. Still, their purely data-driven and extrinsic nature does not incorporate any guidance toward the target surface, often resulting in a trivial misalignment of the template registration. Currently, no method can be considered the standard for 3D Human registration, limiting the scalability of downstream applications. In this work, we propose a neural scalable registration method, NSR, a pipeline that, for the first time, generalizes and scales across thousands of shapes and more than ten different data sources. Our essential contribution is NICP, an ICP-style self-supervised task tailored to neural fields. NSR takes a few seconds, is self-supervised, and works out of the box on pre-trained neural fields. NSR combines NICP with a localized neural field trained on a large MoCap dataset, achieving the state of the art over public benchmarks. The release of our code and checkpoints provides a powerful tool useful for many downstream tasks like dataset alignments, cleaning, or asset animation.
Authors: Zeqian Li, Qirui Chen, Tengda Han, Ya Zhang, Yanfeng Wang, Weidi Xie
Abstract: In this paper, we aim to establish an automatic, scalable pipeline for denoising the large-scale instructional dataset and construct a high-quality video-text dataset with multiple descriptive steps supervision, named HowToStep. We make the following contributions: (i) improving the quality of sentences in dataset by upgrading ASR systems to reduce errors from speech recognition and prompting a large language model to transform noisy ASR transcripts into descriptive steps; (ii) proposing a Transformer-based architecture with all texts as queries, iteratively attending to the visual features, to temporally align the generated steps to corresponding video segments. To measure the quality of our curated datasets, we train models for the task of multi-sentence grounding on it, i.e., given a long-form video, and associated multiple sentences, to determine their corresponding timestamps in the video simultaneously, as a result, the model shows superior performance on a series of multi-sentence grounding tasks, surpassing existing state-of-the-art methods by a significant margin on three public benchmarks, namely, 9.0% on HT-Step, 5.1% on HTM-Align and 1.9% on CrossTask. All codes, models, and the resulting dataset have been publicly released.
Authors: Yunlong Tang, Jing Bi, Siting Xu, Luchuan Song, Susan Liang, Teng Wang, Daoan Zhang, Jie An, Jingyang Lin, Rongyi Zhu, Ali Vosoughi, Chao Huang, Zeliang Zhang, Pinxin Liu, Mingqian Feng, Feng Zheng, Jianguo Zhang, Ping Luo, Jiebo Luo, Chenliang Xu
Abstract: With the burgeoning growth of online video platforms and the escalating volume of video content, the demand for proficient video understanding tools has intensified markedly. Given the remarkable capabilities of large language models (LLMs) in language and multimodal tasks, this survey provides a detailed overview of recent advancements in video understanding that harness the power of LLMs (Vid-LLMs). The emergent capabilities of Vid-LLMs are surprisingly advanced, particularly their ability for open-ended multi-granularity (general, temporal, and spatiotemporal) reasoning combined with commonsense knowledge, suggesting a promising path for future video understanding. We examine the unique characteristics and capabilities of Vid-LLMs, categorizing the approaches into three main types: Video Analyzer x LLM, Video Embedder x LLM, and (Analyzer + Embedder) x LLM. Furthermore, we identify five sub-types based on the functions of LLMs in Vid-LLMs: LLM as Summarizer, LLM as Manager, LLM as Text Decoder, LLM as Regressor, and LLM as Hidden Layer. Furthermore, this survey presents a comprehensive study of the tasks, datasets, benchmarks, and evaluation methodologies for Vid-LLMs. Additionally, it explores the expansive applications of Vid-LLMs across various domains, highlighting their remarkable scalability and versatility in real-world video understanding challenges. Finally, it summarizes the limitations of existing Vid-LLMs and outlines directions for future research. For more information, readers are recommended to visit the repository at https://github.com/yunlong10/Awesome-LLMs-for-Video-Understanding.
URLs: https://github.com/yunlong10/Awesome-LLMs-for-Video-Understanding.
Authors: Jiawei Yang, Katie Z Luo, Jiefeng Li, Congyue Deng, Leonidas Guibas, Dilip Krishnan, Kilian Q Weinberger, Yonglong Tian, Yue Wang
Abstract: We study a crucial yet often overlooked issue inherent to Vision Transformers (ViTs): feature maps of these models exhibit grid-like artifacts, which hurt the performance of ViTs in downstream dense prediction tasks such as semantic segmentation, depth prediction, and object discovery. We trace this issue down to the positional embeddings at the input stage. To mitigate this, we propose a two-stage denoising approach, termed Denoising Vision Transformers (DVT). In the first stage, we separate the clean features from those contaminated by positional artifacts by enforcing cross-view feature consistency with neural fields on a per-image basis. This per-image optimization process extracts artifact-free features from raw ViT outputs, providing clean feature estimates for offline applications. In the second stage, we train a lightweight transformer block to predict clean features from raw ViT outputs, leveraging the derived estimates of the clean features as supervision. Our method, DVT, does not require re-training the existing pre-trained ViTs, and is immediately applicable to any Vision Transformer architecture. We evaluate our method on a variety of representative ViTs (DINO, DeiT-III, EVA02, CLIP, DINOv2, DINOv2-reg) and demonstrate that DVT consistently improves existing state-of-the-art general-purpose models in semantic and geometric tasks across multiple datasets. We hope our study will encourage a re-evaluation of ViT design, especially regarding the naive use of positional embeddings. Our code and checkpoints are publicly available.
Authors: Guikun Chen, Wenguan Wang
Abstract: 3D Gaussian splatting (GS) has recently emerged as a transformative technique in the realm of explicit radiance field and computer graphics. This innovative approach, characterized by the utilization of millions of learnable 3D Gaussians, represents a significant departure from mainstream neural radiance field approaches, which predominantly use implicit, coordinate-based models to map spatial coordinates to pixel values. 3D GS, with its explicit scene representation and differentiable rendering algorithm, not only promises real-time rendering capability but also introduces unprecedented levels of editability. This positions 3D GS as a potential game-changer for the next generation of 3D reconstruction and representation. In the present paper, we provide the first systematic overview of the recent developments and critical contributions in the domain of 3D GS. We begin with a detailed exploration of the underlying principles and the driving forces behind the emergence of 3D GS, laying the groundwork for understanding its significance. A focal point of our discussion is the practical applicability of 3D GS. By enabling unprecedented rendering speed, 3D GS opens up a plethora of applications, ranging from virtual reality to interactive media and beyond. This is complemented by a comparative analysis of leading 3D GS models, evaluated across various benchmark tasks to highlight their performance and practical utility. The survey concludes by identifying current challenges and suggesting potential avenues for future research in this domain. Through this survey, we aim to provide a valuable resource for both newcomers and seasoned researchers, fostering further exploration and advancement in applicable and explicit radiance field representation.
Authors: Zeliang Zhang, Wei Yao, Xiaosen Wang
Abstract: Deep neural networks are widely known to be vulnerable to adversarial examples. However, vanilla adversarial examples generated under the white-box setting often exhibit low transferability across different models. Since adversarial transferability poses more severe threats to practical applications, various approaches have been proposed for better transferability, including gradient-based, input transformation-based, and model-related attacks, \etc. In this work, we find that several tiny changes in the existing adversarial attacks can significantly affect the attack performance, \eg, the number of iterations and step size. Based on careful studies of existing adversarial attacks, we propose a bag of tricks to enhance adversarial transferability, including momentum initialization, scheduled step size, dual example, spectral-based input transformation, and several ensemble strategies. Extensive experiments on the ImageNet dataset validate the high effectiveness of our proposed tricks and show that combining them can further boost adversarial transferability. Our work provides practical insights and techniques to enhance adversarial transferability, and offers guidance to improve the attack performance on the real-world application through simple adjustments.
Authors: Zijie Pan, Zeyu Yang, Xiatian Zhu, Li Zhang
Abstract: Generating dynamic 3D object from a single-view video is challenging due to the lack of 4D labeled data. An intuitive approach is to extend previous image-to-3D pipelines by transferring off-the-shelf image generation models such as score distillation sampling.However, this approach would be slow and expensive to scale due to the need for back-propagating the information-limited supervision signals through a large pretrained model. To address this, we propose an efficient video-to-4D object generation framework called Efficient4D. It generates high-quality spacetime-consistent images under different camera views, and then uses them as labeled data to directly reconstruct the 4D content through a 4D Gaussian splatting model. Importantly, our method can achieve real-time rendering under continuous camera trajectories. To enable robust reconstruction under sparse views, we introduce inconsistency-aware confidence-weighted loss design, along with a lightly weighted score distillation loss. Extensive experiments on both synthetic and real videos show that Efficient4D offers a remarkable 10-fold increase in speed when compared to prior art alternatives while preserving the quality of novel view synthesis. For example, Efficient4D takes only 10 minutes to model a dynamic object, vs 120 minutes by the previous art model Consistent4D.
Authors: Rameshwar Mishra, A V Subramanyam
Abstract: Advancements in generative models have sparked significant interest in generating images while adhering to specific structural guidelines. Scene graph to image generation is one such task of generating images which are consistent with the given scene graph. However, the complexity of visual scenes poses a challenge in accurately aligning objects based on specified relations within the scene graph. Existing methods approach this task by first predicting a scene layout and generating images from these layouts using adversarial training. In this work, we introduce a novel approach to generate images from scene graphs which eliminates the need of predicting intermediate layouts. We leverage pre-trained text-to-image diffusion models and CLIP guidance to translate graph knowledge into images. Towards this, we first pre-train our graph encoder to align graph features with CLIP features of corresponding images using a GAN based training. Further, we fuse the graph features with CLIP embedding of object labels present in the given scene graph to create a graph consistent CLIP guided conditioning signal. In the conditioning input, object embeddings provide coarse structure of the image and graph features provide structural alignment based on relationships among objects. Finally, we fine tune a pre-trained diffusion model with the graph consistent conditioning signal with reconstruction and CLIP alignment loss. Elaborate experiments reveal that our method outperforms existing methods on standard benchmarks of COCO-stuff and Visual Genome dataset.
Authors: Ruibo Ming, Zhewei Huang, Zhuoxuan Ju, Jianming Hu, Lihui Peng, Shuchang Zhou
Abstract: Video prediction, a fundamental task in computer vision, aims to enable models to generate sequences of future frames based on existing video content. This task has garnered widespread application across various domains. In this paper, we comprehensively survey both historical and contemporary works in this field, encompassing the most widely used datasets and algorithms. Our survey scrutinizes the challenges and evolving landscape of video prediction within the realm of computer vision. We propose a novel taxonomy centered on the stochastic nature of video prediction algorithms. This taxonomy accentuates the gradual transition from deterministic to generative prediction methodologies, underlining significant advancements and shifts in approach.
Authors: Jiezhi Yang, Khushi Desai, Charles Packer, Harshil Bhatia, Nicholas Rhinehart, Rowan McAllister, Joseph Gonzalez
Abstract: We propose CARFF, a method for predicting future 3D scenes given past observations. Our method maps 2D ego-centric images to a distribution over plausible 3D latent scene configurations and predicts the evolution of hypothesized scenes through time. Our latents condition a global Neural Radiance Field (NeRF) to represent a 3D scene model, enabling explainable predictions and straightforward downstream planning. This approach models the world as a POMDP and considers complex scenarios of uncertainty in environmental states and dynamics. Specifically, we employ a two-stage training of Pose-Conditional-VAE and NeRF to learn 3D representations, and auto-regressively predict latent scene representations utilizing a mixture density network. We demonstrate the utility of our method in scenarios using the CARLA driving simulator, where CARFF enables efficient trajectory and contingency planning in complex multi-agent autonomous driving scenarios involving occlusions.
Authors: Amin Parchami-Araghi, Moritz B\"ohle, Sukrut Rao, Bernt Schiele
Abstract: Knowledge Distillation (KD) has proven effective for compressing large teacher models into smaller student models. While it is well known that student models can achieve similar accuracies as the teachers, it has also been shown that they nonetheless often do not learn the same function. It is, however, often highly desirable that the student's and teacher's functions share similar properties such as basing the prediction on the same input features, as this ensures that students learn the 'right features' from the teachers. In this work, we explore whether this can be achieved by not only optimizing the classic KD loss but also the similarity of the explanations generated by the teacher and the student. Despite the idea being simple and intuitive, we find that our proposed 'explanation-enhanced' KD (e$^2$KD) (1) consistently provides large gains in terms of accuracy and student-teacher agreement, (2) ensures that the student learns from the teacher to be right for the right reasons and to give similar explanations, and (3) is robust with respect to the model architectures, the amount of training data, and even works with 'approximate', pre-computed explanations.
Authors: Jaeseong Lee, Junha Hyung, Sohyun Jeong, Jaegul Choo
Abstract: Face swapping has gained significant attention for its varied applications. Most previous face swapping approaches have relied on the seesaw game training scheme, also known as the target-oriented approach. However, this often leads to instability in model training and results in undesired samples with blended identities due to the target identity leakage problem. Source-oriented methods achieve more stable training with self-reconstruction objective but often fail to accurately reflect target image's skin color and illumination. This paper introduces the Shape Agnostic Masked AutoEncoder (SAMAE) training scheme, a novel self-supervised approach that combines the strengths of both target-oriented and source-oriented approaches. Our training scheme addresses the limitations of traditional training methods by circumventing the conventional seesaw game and introducing clear ground truth through its self-reconstruction training regime. Our model effectively mitigates identity leakage and reflects target albedo and illumination through learned disentangled identity and non-identity features. Additionally, we closely tackle the shape misalignment and volume discrepancy problems with new techniques, including perforation confusion and random mesh scaling. SAMAE establishes a new state-of-the-art, surpassing other baseline methods, preserving both identity and non-identity attributes without sacrificing on either aspect.
Authors: Edgar Heinert, Matthias Rottmann, Kira Maag, Karsten Kahl
Abstract: Convolutional neural networks (CNNs) for image processing tend to focus on localized texture patterns, commonly referred to as texture bias. While most of the previous works in the literature focus on the task of image classification, we go beyond this and study the texture bias of CNNs in semantic segmentation. In this work, we propose to train CNNs on pre-processed images with less texture to reduce the texture bias. Therein, the challenge is to suppress image texture while preserving shape information. To this end, we utilize edge enhancing diffusion (EED), an anisotropic image diffusion method initially introduced for image compression, to create texture reduced duplicates of existing datasets. Extensive numerical studies are performed with both CNNs and vision transformer models trained on original data and EED-processed data from the Cityscapes dataset and the CARLA driving simulator. We observe strong texture-dependence of CNNs and moderate texture-dependence of transformers. Training CNNs on EED-processed images enables the models to become completely ignorant with respect to texture, demonstrating resilience with respect to texture re-introduction to any degree. Additionally we analyze the performance reduction in depth on a level of connected components in the semantic segmentation and study the influence of EED pre-processing on domain generalization as well as adversarial robustness.
Authors: David Romero, Thamar Solorio
Abstract: We present Q-ViD, a simple approach for video question answering (video QA), that unlike prior methods, which are based on complex architectures, computationally expensive pipelines or use closed models like GPTs, Q-ViD relies on a single instruction-aware open vision-language model (InstructBLIP) to tackle videoQA using frame descriptions. Specifically, we create captioning instruction prompts that rely on the target questions about the videos and leverage InstructBLIP to obtain video frame captions that are useful to the task at hand. Subsequently, we form descriptions of the whole video using the question-dependent frame captions, and feed that information, along with a question-answering prompt, to a large language model (LLM). The LLM is our reasoning module, and performs the final step of multiple-choice QA. Our simple Q-ViD framework achieves competitive or even higher performances than current state of the art models on a diverse range of videoQA benchmarks, including NExT-QA, STAR, How2QA, TVQA and IntentQA.
Authors: Muyang He, Yexin Liu, Boya Wu, Jianhao Yuan, Yueze Wang, Tiejun Huang, Bo Zhao
Abstract: Multimodal Large Language Models (MLLMs) have demonstrated notable capabilities in general visual understanding and reasoning tasks. However, their deployment is hindered by substantial computational costs in both training and inference, limiting accessibility to the broader research and user communities. A straightforward solution is to leverage smaller pre-trained vision and language models, which inevitably cause significant performance drops. In this paper, we demonstrate the possibility of training a smaller but better MLLM with high-quality training data. Specifically, we introduce Bunny, a family of lightweight MLLMs with flexible vision and language backbones for efficient multimodal learning from selected training data. Experiments show that our Bunny-4B/8B outperforms the state-of-the-art large MLLMs on multiple benchmarks. We expect that this work can provide the community with a clean and flexible open-source tool for further research and development. The code, models, and data can be found in https://github.com/BAAI-DCAI/Bunny.
Authors: Mehedi Hasan Raju, Lee Friedman, Dillon J Lohr, Oleg V Komogortsev
Abstract: What qualities make a feature useful for biometric performance? In prior research, pre-dating the advent of deep learning (DL) approaches to biometric analysis, a strong relationship between temporal persistence, as indexed by the intraclass correlation coefficient (ICC), and biometric performance (Equal Error Rate, EER) was noted. More generally, the claim was made that good biometric performance resulted from a relatively large set of weakly intercorrelated features with high ICC. The present study aimed to determine whether the same relationships are found in a state-of-the-art DL-based eye movement biometric system (``Eye-Know-You-Too''), as applied to two publicly available eye movement datasets. To this end, we manipulate various aspects of eye-tracking signal quality, which produces variation in biometric performance, and relate that performance to the temporal persistence and intercorrelation of the resulting embeddings. Data quality indices were related to EER with either linear or logarithmic fits, and the resulting model R^2 was noted. As a general matter, we found that temporal persistence was an important predictor of DL-based biometric performance, and also that DL-learned embeddings were generally weakly intercorrelated.
Authors: Yuqi Li, Qingqing Long, Yihang Zhou, Ning Cao, Shuai Liu, Fang Zheng, Zhihong Zhu, Zhiyuan Ning, Meng Xiao, Xuezhi Wang, Pengfei Wang, Yuanchun Zhou
Abstract: Zero-shot hashing (ZSH) has shown excellent success owing to its efficiency and generalization in large-scale retrieval scenarios. While considerable success has been achieved, there still exist urgent limitations. Existing works ignore the locality relationships of representations and attributes, which have effective transferability between seeable classes and unseeable classes. Also, the continuous-value attributes are not fully harnessed. In response, we conduct a COMprehensive Attribute Exploration for ZSH, named COMAE, which depicts the relationships from seen classes to unseen ones through three meticulously designed explorations, i.e., point-wise, pair-wise and class-wise consistency constraints. By regressing attributes from the proposed attribute prototype network, COMAE learns the local features that are relevant to the visual attributes. Then COMAE utilizes contrastive learning to comprehensively depict the context of attributes, rather than instance-independent optimization. Finally, the class-wise constraint is designed to cohesively learn the hash code, image representation, and visual attributes more effectively. Experimental results on the popular ZSH datasets demonstrate that COMAE outperforms state-of-the-art hashing techniques, especially in scenarios with a larger number of unseen label classes.
Authors: Yunwei Bai, Ying Kiat Tan, Tsuhan Chen
Abstract: Few-shot-learning (FSL) commonly requires a model to identify images (queries) that belong to classes unseen during training, based on a few labelled samples of the new classes (support set) as reference. So far, plenty of algorithms involve training data augmentation to improve the generalization capability of FSL models, but outlier query or support images during inference can still pose great generalization challenges. In this work, to reduce the bias caused by unconventional test samples, we generate additional test-class samples through combining them with suitable train-class samples, whose averaged representations then become more typical for few-shot classification. We experimentally and theoretically demonstrate the effectiveness of our method, via which we can achieve around 4\% test accuracy improvement for trained FSL models. Importantly, our method is universally compatible with different off-the-shelf FSL models, whose performance can be improved without extra dataset nor further training of the models themselves.
Authors: Fahimeh Hosseini Noohdani, Parsa Hosseini, Aryan Yazdan Parast, Hamidreza Yaghoubi Araghi, Mahdieh Soleymani Baghshah
Abstract: While standard Empirical Risk Minimization (ERM) training is proven effective for image classification on in-distribution data, it fails to perform well on out-of-distribution samples. One of the main sources of distribution shift for image classification is the compositional nature of images. Specifically, in addition to the main object or component(s) determining the label, some other image components usually exist, which may lead to the shift of input distribution between train and test environments. More importantly, these components may have spurious correlations with the label. To address this issue, we propose Decompose-and-Compose (DaC), which improves robustness to correlation shift by a compositional approach based on combining elements of images. Based on our observations, models trained with ERM usually highly attend to either the causal components or the components having a high spurious correlation with the label (especially in datapoints on which models have a high confidence). In fact, according to the amount of spurious correlation and the easiness of classification based on the causal or non-causal components, the model usually attends to one of these more (on samples with high confidence). Following this, we first try to identify the causal components of images using class activation maps of models trained with ERM. Afterward, we intervene on images by combining them and retraining the model on the augmented data, including the counterfactual ones. Along with its high interpretability, this work proposes a group-balancing method by intervening on images without requiring group labels or information regarding the spurious features during training. The method has an overall better worst group accuracy compared to previous methods with the same amount of supervision on the group labels in correlation shift.
Authors: Ritwik Gupta, Shufan Li, Tyler Zhu, Jitendra Malik, Trevor Darrell, Karttikeya Mangalam
Abstract: Modern computer vision pipelines handle large images in one of two sub-optimal ways: down-sampling or cropping. These two methods incur significant losses in the amount of information and context present in an image. There are many downstream applications in which global context matters as much as high frequency details, such as in real-world satellite imagery; in such cases researchers have to make the uncomfortable choice of which information to discard. We introduce xT, a simple framework for vision transformers which effectively aggregates global context with local details and can model large images end-to-end on contemporary GPUs. We select a set of benchmark datasets across classic vision tasks which accurately reflect a vision model's ability to understand truly large images and incorporate fine details over large scales and assess our method's improvement on them. xT is a streaming, two-stage architecture that adapts existing vision backbones and long sequence language models to effectively model large images without quadratic memory growth. We are able to increase accuracy by up to 8.6% on challenging classification tasks and $F_1$ score by 11.6 on context-dependent segmentation on images as large as 29,000 x 29,000 pixels.
Authors: Zichong Meng, Jie Zhang, Changdi Yang, Zheng Zhan, Pu Zhao, Yanzhi Wang
Abstract: Class Incremental Learning (CIL) is challenging due to catastrophic forgetting. On top of that, Exemplar-free Class Incremental Learning is even more challenging due to forbidden access to previous task data. Recent exemplar-free CIL methods attempt to mitigate catastrophic forgetting by synthesizing previous task data. However, they fail to overcome the catastrophic forgetting due to the inability to deal with the significant domain gap between real and synthetic data. To overcome these issues, we propose a novel exemplar-free CIL method. Our method adopts multi-distribution matching (MDM) diffusion models to unify quality and bridge domain gaps among all domains of training data. Moreover, our approach integrates selective synthetic image augmentation (SSIA) to expand the distribution of the training data, thereby improving the model's plasticity and reinforcing the performance of our method's ultimate component, multi-domain adaptation (MDA). With the proposed integrations, our method then reformulates exemplar-free CIL into a multi-domain adaptation problem to implicitly address the domain gap problem to enhance model stability during incremental training. Extensive experiments on benchmark class incremental datasets and settings demonstrate that our method excels previous exemplar-free CIL methods and achieves state-of-the-art performance.
Authors: Zichong Meng, Changdi Yang, Jun Liu, Hao Tang, Pu Zhao, Yanzhi Wang
Abstract: Recent advances in image editing have been driven by the development of denoising diffusion models, marking a significant leap forward in this field. Despite these advances, the generalization capabilities of recent image editing approaches remain constrained. In response to this challenge, our study introduces a novel image editing framework with enhanced generalization robustness by boosting in-context learning capability and unifying language instruction. This framework incorporates a module specifically optimized for image editing tasks, leveraging the VMamba Block and an editing-shift matching strategy to augment in-context learning. Furthermore, we unveil a selective area-matching technique specifically engineered to address and rectify corrupted details in generated images, such as human facial features, to further improve the quality. Another key innovation of our approach is the integration of a language unification technique, which aligns language embeddings with editing semantics to elevate the quality of image editing. Moreover, we compile the first dataset for image editing with visual prompts and editing instructions that could be used to enhance in-context capability. Trained on this dataset, our methodology not only achieves superior synthesis quality for trained tasks, but also demonstrates robust generalization capability across unseen vision tasks through tailored prompts.
Authors: Angeliki Dimitriou, Maria Lymperaiou, Giorgos Filandrianos, Konstantinos Thomas, Giorgos Stamou
Abstract: Counterfactual explanations (CEs) based on concepts are explanations that consider alternative scenarios to understand which high-level semantic features contributed to particular model predictions. In this work, we propose CEs based on the semantic graphs accompanying input data to achieve more descriptive, accurate, and human-aligned explanations. Building upon state-of-the-art (SoTA) conceptual attempts, we adopt a model-agnostic edit-based approach and introduce leveraging GNNs for efficient Graph Edit Distance (GED) computation. With a focus on the visual domain, we represent images as scene graphs and obtain their GNN embeddings to bypass solving the NP-hard graph similarity problem for all input pairs, an integral part of the CE computation process. We apply our method to benchmark and real-world datasets with varying difficulty and availability of semantic annotations. Testing on diverse classifiers, we find that our CEs outperform previous SoTA explanation models based on semantics, including both white and black-box as well as conceptual and pixel-level approaches. Their superiority is proven quantitatively and qualitatively, as validated by human subjects, highlighting the significance of leveraging semantic edges in the presence of intricate relationships. Our model-agnostic graph-based approach is widely applicable and easily extensible, producing actionable explanations across different contexts.
Authors: Lei Shi, Paul B\"urkner, Andreas Bulling
Abstract: We present ActionDiffusion -- a novel diffusion model for procedure planning in instructional videos that is the first to take temporal inter-dependencies between actions into account in a diffusion model for procedure planning. This approach is in stark contrast to existing methods that fail to exploit the rich information content available in the particular order in which actions are performed. Our method unifies the learning of temporal dependencies between actions and denoising of the action plan in the diffusion process by projecting the action information into the noise space. This is achieved 1) by adding action embeddings in the noise masks in the noise-adding phase and 2) by introducing an attention mechanism in the noise prediction network to learn the correlations between different action steps. We report extensive experiments on three instructional video benchmark datasets (CrossTask, Coin, and NIV) and show that our method outperforms previous state-of-the-art methods on all metrics on CrossTask and NIV and all metrics except accuracy on Coin dataset. We show that by adding action embeddings into the noise mask the diffusion model can better learn action temporal dependencies and increase the performances on procedure planning.
Authors: Zhe Kong, Yong Zhang, Tianyu Yang, Tao Wang, Kaihao Zhang, Bizhu Wu, Guanying Chen, Wei Liu, Wenhan Luo
Abstract: Personalization is an important topic in text-to-image generation, especially the challenging multi-concept personalization. Current multi-concept methods are struggling with identity preservation, occlusion, and the harmony between foreground and background. In this work, we propose OMG, an occlusion-friendly personalized generation framework designed to seamlessly integrate multiple concepts within a single image. We propose a novel two-stage sampling solution. The first stage takes charge of layout generation and visual comprehension information collection for handling occlusions. The second one utilizes the acquired visual comprehension information and the designed noise blending to integrate multiple concepts while considering occlusions. We also observe that the initiation denoising timestep for noise blending is the key to identity preservation and layout. Moreover, our method can be combined with various single-concept models, such as LoRA and InstantID without additional tuning. Especially, LoRA models on civitai.com can be exploited directly. Extensive experiments demonstrate that OMG exhibits superior performance in multi-concept personalization.
Authors: Xiaoji Zheng, Lixiu Wu, Zhijie Yan, Yuanrong Tang, Hao Zhao, Chen Zhong, Bokui Chen, Jiangtao Gong
Abstract: Motion prediction is among the most fundamental tasks in autonomous driving. Traditional methods of motion forecasting primarily encode vector information of maps and historical trajectory data of traffic participants, lacking a comprehensive understanding of overall traffic semantics, which in turn affects the performance of prediction tasks. In this paper, we utilized Large Language Models (LLMs) to enhance the global traffic context understanding for motion prediction tasks. We first conducted systematic prompt engineering, visualizing complex traffic environments and historical trajectory information of traffic participants into image prompts -- Transportation Context Map (TC-Map), accompanied by corresponding text prompts. Through this approach, we obtained rich traffic context information from the LLM. By integrating this information into the motion prediction model, we demonstrate that such context can enhance the accuracy of motion predictions. Furthermore, considering the cost associated with LLMs, we propose a cost-effective deployment strategy: enhancing the accuracy of motion prediction tasks at scale with 0.7\% LLM-augmented datasets. Our research offers valuable insights into enhancing the understanding of traffic scenes of LLMs and the motion prediction performance of autonomous driving. The source code is available at \url{https://github.com/AIR-DISCOVER/LLM-Augmented-MTR} and \url{https://aistudio.baidu.com/projectdetail/7809548}.
URLs: https://github.com/AIR-DISCOVER/LLM-Augmented-MTR, https://aistudio.baidu.com/projectdetail/7809548
Authors: Yuxin Yao, Siyu Ren, Junhui Hou, Zhi Deng, Juyong Zhang, Wenping Wang
Abstract: This paper explores the problem of reconstructing temporally consistent surfaces from a 3D point cloud sequence without correspondence. To address this challenging task, we propose DynoSurf, an unsupervised learning framework integrating a template surface representation with a learnable deformation field. Specifically, we design a coarse-to-fine strategy for learning the template surface based on the deformable tetrahedron representation. Furthermore, we propose a learnable deformation representation based on the learnable control points and blending weights, which can deform the template surface non-rigidly while maintaining the consistency of the local shape. Experimental results demonstrate the significant superiority of DynoSurf over current state-of-the-art approaches, showcasing its potential as a powerful tool for dynamic mesh reconstruction. The code is publicly available at https://github.com/yaoyx689/DynoSurf.
Authors: Fucai Ke, Zhixi Cai, Simindokht Jahangard, Weiqing Wang, Pari Delir Haghighi, Hamid Rezatofighi
Abstract: Recent advances in visual reasoning (VR), particularly with the aid of Large Vision-Language Models (VLMs), show promise but require access to large-scale datasets and face challenges such as high computational costs and limited generalization capabilities. Compositional visual reasoning approaches have emerged as effective strategies; however, they heavily rely on the commonsense knowledge encoded in Large Language Models (LLMs) to perform planning, reasoning, or both, without considering the effect of their decisions on the visual reasoning process, which can lead to errors or failed procedures. To address these challenges, we introduce HYDRA, a multi-stage dynamic compositional visual reasoning framework designed for reliable and incrementally progressive general reasoning. HYDRA integrates three essential modules: a planner, a Reinforcement Learning (RL) agent serving as a cognitive controller, and a reasoner. The planner and reasoner modules utilize an LLM to generate instruction samples and executable code from the selected instruction, respectively, while the RL agent dynamically interacts with these modules, making high-level decisions on selection of the best instruction sample given information from the historical state stored through a feedback loop. This adaptable design enables HYDRA to adjust its actions based on previous feedback received during the reasoning process, leading to more reliable reasoning outputs and ultimately enhancing its overall effectiveness. Our framework demonstrates state-of-the-art performance in various VR tasks on four different widely-used datasets.
Authors: Beichen Zhang, Pan Zhang, Xiaoyi Dong, Yuhang Zang, Jiaqi Wang
Abstract: Contrastive Language-Image Pre-training (CLIP) has been the cornerstone for zero-shot classification, text-image retrieval, and text-image generation by aligning image and text modalities. Despite its widespread adoption, a significant limitation of CLIP lies in the inadequate length of text input. The length of the text token is restricted to 77, and an empirical study shows the actual effective length is even less than 20. This prevents CLIP from handling detailed descriptions, limiting its applications for image retrieval and text-to-image generation with extensive prerequisites. To this end, we propose Long-CLIP as a plug-and-play alternative to CLIP that supports long-text input, retains or even surpasses its zero-shot generalizability, and aligns the CLIP latent space, making it readily replace CLIP without any further adaptation in downstream frameworks. Nevertheless, achieving this goal is far from straightforward, as simplistic fine-tuning can result in a significant degradation of CLIP's performance. Moreover, substituting the text encoder with a language model supporting longer contexts necessitates pretraining with vast amounts of data, incurring significant expenses. Accordingly, Long-CLIP introduces an efficient fine-tuning solution on CLIP with two novel strategies designed to maintain the original capabilities, including (1) a knowledge-preserved stretching of positional embedding and (2) a primary component matching of CLIP features. With leveraging just one million extra long text-image pairs, Long-CLIP has shown the superiority to CLIP for about 20% in long caption text-image retrieval and 6% in traditional text-image retrieval tasks, e.g., COCO and Flickr30k. Furthermore, Long-CLIP offers enhanced capabilities for generating images from detailed text descriptions by replacing CLIP in a plug-and-play manner.
Authors: Junqiao Fan, Jianfei Yang, Yuecong Xu, Lihua Xie
Abstract: Human pose estimation (HPE) from Radio Frequency vision (RF-vision) performs human sensing using RF signals that penetrate obstacles without revealing privacy (e.g., facial information). Recently, mmWave radar has emerged as a promising RF-vision sensor, providing radar point clouds by processing RF signals. However, the mmWave radar has a limited resolution with severe noise, leading to inaccurate and inconsistent human pose estimation. This work proposes mmDiff, a novel diffusion-based pose estimator tailored for noisy radar data. Our approach aims to provide reliable guidance as conditions to diffusion models. Two key challenges are addressed by mmDiff: (1) miss-detection of parts of human bodies, which is addressed by a module that isolates feature extraction from different body parts, and (2) signal inconsistency due to environmental interference, which is tackled by incorporating prior knowledge of body structure and motion. Several modules are designed to achieve these goals, whose features work as the conditions for the subsequent diffusion model, eliminating the miss-detection and instability of HPE based on RF-vision. Extensive experiments demonstrate that mmDiff outperforms existing methods significantly, achieving state-of-the-art performances on public datasets.
Authors: Souhail Hadgi, Lei Li, Maks Ovsjanikov
Abstract: Transfer learning has long been a key factor in the advancement of many fields including 2D image analysis. Unfortunately, its applicability in 3D data processing has been relatively limited. While several approaches for point cloud transfer learning have been proposed in recent literature, with contrastive learning gaining particular prominence, most existing methods in this domain have only been studied and evaluated in limited scenarios. Most importantly, there is currently a lack of principled understanding of both when and why point cloud transfer learning methods are applicable. Remarkably, even the applicability of standard supervised pre-training is poorly understood. In this work, we conduct the first in-depth quantitative and qualitative investigation of supervised and contrastive pre-training strategies and their utility in downstream 3D tasks. We demonstrate that layer-wise analysis of learned features provides significant insight into the downstream utility of trained networks. Informed by this analysis, we propose a simple geometric regularization strategy, which improves the transferability of supervised pre-training. Our work thus sheds light onto both the specific challenges of point cloud transfer learning, as well as strategies to overcome them.
Authors: Lisa Anita De Santi, J\"org Schl\"otterer, Michael Scheschenja, Joel Wessendorf, Meike Nauta, Vincenzo Positano, Christin Seifert
Abstract: Information from neuroimaging examinations is increasingly used to support diagnoses of dementia, e.g., Alzheimer's disease. While current clinical practice is mainly based on visual inspection and feature engineering, Deep Learning approaches can be used to automate the analysis and to discover new image-biomarkers. Part-prototype neural networks (PP-NN) are an alternative to standard blackbox models, and have shown promising results in general computer vision. PP-NN's base their reasoning on prototypical image regions that are learned fully unsupervised, and combined with a simple-to-understand decision layer. We present PIPNet3D, a PP-NN for volumetric images. We apply PIPNet3D to the clinical diagnosis of Alzheimer's Disease from structural Magnetic Resonance Imaging (sMRI). We assess the quality of prototypes under a systematic evaluation framework, propose new functionally grounded metrics to evaluate brain prototypes and develop an evaluation scheme to assess their coherency with domain experts. Our results show that PIPNet3D is an interpretable, compact model for Alzheimer's diagnosis with its reasoning well aligned to medical domain knowledge. Notably, PIPNet3D achieves the same accuracy as its blackbox counterpart; and removing the remaining clinically irrelevant prototypes from its decision process does not decrease predictive performance.
Authors: Qianliang Wu, Haobo Jiang, Lei Luo, Jun Li, Yaqing Ding, Jin Xie, Jian Yang
Abstract: Establishing reliable correspondences is essential for registration tasks such as 3D and 2D3D registration. Existing methods commonly leverage geometric or semantic point features to generate potential correspondences. However, these features may face challenges such as large deformation, scale inconsistency, and ambiguous matching problems (e.g., symmetry). Additionally, many previous methods, which rely on single-pass prediction, may struggle with local minima in complex scenarios. To mitigate these challenges, we introduce a diffusion matching model for robust correspondence construction. Our approach treats correspondence estimation as a denoising diffusion process within the doubly stochastic matrix space, which gradually denoises (refines) a doubly stochastic matching matrix to the ground-truth one for high-quality correspondence estimation. It involves a forward diffusion process that gradually introduces Gaussian noise into the ground truth matching matrix and a reverse denoising process that iteratively refines the noisy matching matrix. In particular, the feature extraction from the backbone occurs only once during the inference phase. Our lightweight denoising module utilizes the same feature at each reverse sampling step. Evaluation of our method on both 3D and 2D3D registration tasks confirms its effectiveness. The code is available at https://github.com/wuqianliang/Diff-Reg.
Authors: Ye Liu, Jixuan He, Wanhua Li, Junsik Kim, Donglai Wei, Hanspeter Pfister, Chang Wen Chen
Abstract: Video temporal grounding (VTG) is a fine-grained video understanding problem that aims to ground relevant clips in untrimmed videos given natural language queries. Most existing VTG models are built upon frame-wise final-layer CLIP features, aided by additional temporal backbones (e.g., SlowFast) with sophisticated temporal reasoning mechanisms. In this work, we claim that CLIP itself already shows great potential for fine-grained spatial-temporal modeling, as each layer offers distinct yet useful information under different granularity levels. Motivated by this, we propose Reversed Recurrent Tuning ($R^2$-Tuning), a parameter- and memory-efficient transfer learning framework for video temporal grounding. Our method learns a lightweight $R^2$ Block containing only 1.5% of the total parameters to perform progressive spatial-temporal modeling. Starting from the last layer of CLIP, $R^2$ Block recurrently aggregates spatial features from earlier layers, then refines temporal correlation conditioning on the given query, resulting in a coarse-to-fine scheme. $R^2$-Tuning achieves state-of-the-art performance across three VTG tasks (i.e., moment retrieval, highlight detection, and video summarization) on six public benchmarks (i.e., QVHighlights, Charades-STA, Ego4D-NLQ, TACoS, YouTube Highlights, and TVSum) even without the additional backbone, demonstrating the significance and effectiveness of the proposed scheme. Our code is available at https://github.com/yeliudev/R2-Tuning.
Authors: Tatiana Gaintseva, Martin Benning, Gregory Slabaugh
Abstract: In this paper we propose a novel modification of Contrastive Language-Image Pre-Training (CLIP) guidance for the task of unsupervised backlit image enhancement. Our work builds on the state-of-the-art CLIP-LIT approach, which learns a prompt pair by constraining the text-image similarity between a prompt (negative/positive sample) and a corresponding image (backlit image/well-lit image) in the CLIP embedding space. Learned prompts then guide an image enhancement network. Based on the CLIP-LIT framework, we propose two novel methods for CLIP guidance. First, we show that instead of tuning prompts in the space of text embeddings, it is possible to directly tune their embeddings in the latent space without any loss in quality. This accelerates training and potentially enables the use of additional encoders that do not have a text encoder. Second, we propose a novel approach that does not require any prompt tuning. Instead, based on CLIP embeddings of backlit and well-lit images from training data, we compute the residual vector in the embedding space as a simple difference between the mean embeddings of the well-lit and backlit images. This vector then guides the enhancement network during training, pushing a backlit image towards the space of well-lit images. This approach further dramatically reduces training time, stabilizes training and produces high quality enhanced images without artifacts, both in supervised and unsupervised training regimes. Additionally, we show that residual vectors can be interpreted, revealing biases in training data, and thereby enabling potential bias correction.
Authors: Yuetian Weng, Mingfei Han, Haoyu He, Xiaojun Chang, Bohan Zhuang
Abstract: Empowered by Large Language Models (LLMs), recent advancements in Video-based LLMs (VideoLLMs) have driven progress in various video understanding tasks. These models encode video representations through pooling or query aggregation over a vast number of visual tokens, making computational and memory costs affordable. Despite successfully providing an overall comprehension of video content, existing VideoLLMs still face challenges in achieving detailed understanding due to overlooking local information in long-term videos. To tackle this challenge, we introduce LongVLM, a simple yet powerful VideoLLM for long video understanding, building upon the observation that long videos often consist of sequential key events, complex actions, and camera movements. Our approach proposes to decompose long videos into multiple short-term segments and encode local features for each segment via a hierarchical token merging module. These features are concatenated in temporal order to maintain the storyline across sequential short-term segments. Additionally, we propose to integrate global semantics into each local feature to enhance context understanding. In this way, we encode video representations that incorporate both local and global information, enabling the LLM to generate comprehensive responses for long-term videos. Experimental results on the VideoChatGPT benchmark and zero-shot video question-answering datasets demonstrate the superior capabilities of our model over the previous state-of-the-art methods. Qualitative examples show that our model produces more precise responses for long video understanding. Code is available at https://github.com/ziplab/LongVLM.
Authors: Yu Cai, Weiwen Zhang, Hao Chen, Kwang-Ting Cheng
Abstract: Anomaly detection (AD) aims at detecting abnormal samples that deviate from the expected normal patterns. Generally, it can be trained merely on normal data, without a requirement for abnormal samples, and thereby plays an important role in the recognition of rare diseases and health screening in the medical domain. Despite the emergence of numerous methods for medical AD, we observe a lack of a fair and comprehensive evaluation, which causes ambiguous conclusions and hinders the development of this field. To address this problem, this paper builds a benchmark with unified comparison. Seven medical datasets with five image modalities, including chest X-rays, brain MRIs, retinal fundus images, dermatoscopic images, and histopathology whole slide images, are curated for extensive evaluation. Thirty typical AD methods, including reconstruction and self-supervised learning-based methods, are involved in comparison of image-level anomaly classification and pixel-level anomaly segmentation. Furthermore, for the first time, we formally explore the effect of key components in existing methods, clearly revealing unresolved challenges and potential future directions. The datasets and code are available at \url{https://github.com/caiyu6666/MedIAnomaly}.
Authors: Shiming Chen, Wenjin Hou, Salman Khan, Fahad Shahbaz Khan
Abstract: Zero-shot learning (ZSL) recognizes the unseen classes by conducting visual-semantic interactions to transfer semantic knowledge from seen classes to unseen ones, supported by semantic information (e.g., attributes). However, existing ZSL methods simply extract visual features using a pre-trained network backbone (i.e., CNN or ViT), which fail to learn matched visual-semantic correspondences for representing semantic-related visual features as lacking of the guidance of semantic information, resulting in undesirable visual-semantic interactions. To tackle this issue, we propose a progressive semantic-guided vision transformer for zero-shot learning (dubbed ZSLViT). ZSLViT mainly considers two properties in the whole network: i) discover the semantic-related visual representations explicitly, and ii) discard the semantic-unrelated visual information. Specifically, we first introduce semantic-embedded token learning to improve the visual-semantic correspondences via semantic enhancement and discover the semantic-related visual tokens explicitly with semantic-guided token attention. Then, we fuse low semantic-visual correspondence visual tokens to discard the semantic-unrelated visual information for visual enhancement. These two operations are integrated into various encoders to progressively learn semantic-related visual representations for accurate visual-semantic interactions in ZSL. The extensive experiments show that our ZSLViT achieves significant performance gains on three popular benchmark datasets, i.e., CUB, SUN, and AWA2. Codes are available at: https://github.com/shiming-chen/ZSLViT .
Authors: Ming Li, Taojiannan Yang, Huafeng Kuang, Jie Wu, Zhaoning Wang, Xuefeng Xiao, Chen Chen
Abstract: To enhance the controllability of text-to-image diffusion models, existing efforts like ControlNet incorporated image-based conditional controls. In this paper, we reveal that existing methods still face significant challenges in generating images that align with the image conditional controls. To this end, we propose ControlNet++, a novel approach that improves controllable generation by explicitly optimizing pixel-level cycle consistency between generated images and conditional controls. Specifically, for an input conditional control, we use a pre-trained discriminative reward model to extract the corresponding condition of the generated images, and then optimize the consistency loss between the input conditional control and extracted condition. A straightforward implementation would be generating images from random noises and then calculating the consistency loss, but such an approach requires storing gradients for multiple sampling timesteps, leading to considerable time and memory costs. To address this, we introduce an efficient reward strategy that deliberately disturbs the input images by adding noise, and then uses the single-step denoised images for reward fine-tuning. This avoids the extensive costs associated with image sampling, allowing for more efficient reward fine-tuning. Extensive experiments show that ControlNet++ significantly improves controllability under various conditional controls. For example, it achieves improvements over ControlNet by 11.1% mIoU, 13.4% SSIM, and 7.6% RMSE, respectively, for segmentation mask, line-art edge, and depth conditions. All the code, models, demo and organized data have been open sourced on our Github Repo.
Authors: Muhammad Saad Saeed, Shah Nawaz, Muhammad Salman Tahir, Rohan Kumar Das, Muhammad Zaigham Zaheer, Marta Moscati, Markus Schedl, Muhammad Haris Khan, Karthik Nandakumar, Muhammad Haroon Yousaf
Abstract: The advancements of technology have led to the use of multimodal systems in various real-world applications. Among them, the audio-visual systems are one of the widely used multimodal systems. In the recent years, associating face and voice of a person has gained attention due to presence of unique correlation between them. The Face-voice Association in Multilingual Environments (FAME) Challenge 2024 focuses on exploring face-voice association under a unique condition of multilingual scenario. This condition is inspired from the fact that half of the world's population is bilingual and most often people communicate under multilingual scenario. The challenge uses a dataset namely, Multilingual Audio-Visual (MAV-Celeb) for exploring face-voice association in multilingual environments. This report provides the details of the challenge, dataset, baselines and task details for the FAME Challenge.
Authors: Fangwei Zhong, Kui Wu, Hai Ci, Churan Wang, Hao Chen
Abstract: Embodied visual tracking is to follow a target object in dynamic 3D environments using an agent's egocentric vision. This is a vital and challenging skill for embodied agents. However, existing methods suffer from inefficient training and poor generalization. In this paper, we propose a novel framework that combines visual foundation models(VFM) and offline reinforcement learning(offline RL) to empower embodied visual tracking. We use a pre-trained VFM, such as "Tracking Anything", to extract semantic segmentation masks with text prompts. We then train a recurrent policy network with offline RL, e.g., Conservative Q-Learning, to learn from the collected demonstrations without online interactions. To further improve the robustness and generalization of the policy network, we also introduce a mask re-targeting mechanism and a multi-level data collection strategy. In this way, we can train a robust policy within an hour on a consumer-level GPU, e.g., Nvidia RTX 3090. We evaluate our agent on several high-fidelity environments with challenging situations, such as distraction and occlusion. The results show that our agent outperforms state-of-the-art methods in terms of sample efficiency, robustness to distractors, and generalization to unseen scenarios and targets. We also demonstrate the transferability of the learned agent from virtual environments to a real-world robot.
Authors: Giannis Daras, Alexandros G. Dimakis, Constantinos Daskalakis
Abstract: Ambient diffusion is a recently proposed framework for training diffusion models using corrupted data. Both Ambient Diffusion and alternative SURE-based approaches for learning diffusion models from corrupted data resort to approximations which deteriorate performance. We present the first framework for training diffusion models that provably sample from the uncorrupted distribution given only noisy training data, solving an open problem in this space. Our key technical contribution is a method that uses a double application of Tweedie's formula and a consistency loss function that allows us to extend sampling at noise levels below the observed data noise. We also provide further evidence that diffusion models memorize from their training sets by identifying extremely corrupted images that are almost perfectly reconstructed, raising copyright and privacy concerns. Our method for training using corrupted samples can be used to mitigate this problem. We demonstrate this by fine-tuning Stable Diffusion XL to generate samples from a distribution using only noisy samples. Our framework reduces the amount of memorization of the fine-tuning dataset, while maintaining competitive performance.
Authors: Niklas Gard, Anna Hilsmann, Peter Eisert
Abstract: In this paper, we present SPVLoc, a global indoor localization method that accurately determines the six-dimensional (6D) camera pose of a query image and requires minimal scene-specific prior knowledge and no scene-specific training. Our approach employs a novel matching procedure to localize the perspective camera's viewport, given as an RGB image, within a set of panoramic semantic layout representations of the indoor environment. The panoramas are rendered from an untextured 3D reference model, which only comprises approximate structural information about room shapes, along with door and window annotations. We demonstrate that a straightforward convolutional network structure can successfully achieve image-to-panorama and ultimately image-to-model matching. Through a viewport classification score, we rank reference panoramas and select the best match for the query image. Then, a 6D relative pose is estimated between the chosen panorama and query image. Our experiments demonstrate that this approach not only efficiently bridges the domain gap but also generalizes well to previously unseen scenes that are not part of the training data. Moreover, it achieves superior localization accuracy compared to the state of the art methods and also estimates more degrees of freedom of the camera pose. Our source code is publicly available at https://fraunhoferhhi.github.io/spvloc .
Authors: Chen Xu, Tianhui Song, Weixin Feng, Xubin Li, Tiezheng Ge, Bo Zheng, Limin Wang
Abstract: Diffusion models have significantly advanced the state of the art in image, audio, and video generation tasks. However, their applications in practical scenarios are hindered by slow inference speed. Drawing inspiration from the approximation strategies utilized in consistency models, we propose the Sub-path Linear Approximation Model (SLAM), which accelerates diffusion models while maintaining high-quality image generation. SLAM treats the PF-ODE trajectory as a series of PF-ODE sub-paths divided by sampled points, and harnesses sub-path linear (SL) ODEs to form a progressive and continuous error estimation along each individual PF-ODE sub-path. The optimization on such SL-ODEs allows SLAM to construct denoising mappings with smaller cumulative approximated errors. An efficient distillation method is also developed to facilitate the incorporation of more advanced diffusion models, such as latent diffusion models. Our extensive experimental results demonstrate that SLAM achieves an efficient training regimen, requiring only 6 A100 GPU days to produce a high-quality generative model capable of 2 to 4-step generation with high performance. Comprehensive evaluations on LAION, MS COCO 2014, and MS COCO 2017 datasets also illustrate that SLAM surpasses existing acceleration methods in few-step generation tasks, achieving state-of-the-art performance both on FID and the quality of the generated images.
Authors: Zirui Wang, Wenjing Bian, Omkar Parkhi, Yuheng Ren, Victor Adrian Prisacariu
Abstract: We introduce a novel cross-reference image quality assessment method that effectively fills the gap in the image assessment landscape, complementing the array of established evaluation schemes -- ranging from full-reference metrics like SSIM, no-reference metrics such as NIQE, to general-reference metrics including FID, and Multi-modal-reference metrics, e.g., CLIPScore. Utilising a neural network with the cross-attention mechanism and a unique data collection pipeline from NVS optimisation, our method enables accurate image quality assessment without requiring ground truth references. By comparing a query image against multiple views of the same scene, our method addresses the limitations of existing metrics in novel view synthesis (NVS) and similar tasks where direct reference images are unavailable. Experimental results show that our method is closely correlated to the full-reference metric SSIM, while not requiring ground truth references.
Authors: Hang Hua, Jing Shi, Kushal Kafle, Simon Jenni, Daoan Zhang, John Collomosse, Scott Cohen, Jiebo Luo
Abstract: Recent progress in large-scale pre-training has led to the development of advanced vision-language models (VLMs) with remarkable proficiency in comprehending and generating multimodal content. Despite the impressive ability to perform complex reasoning for VLMs, current models often struggle to effectively and precisely capture the compositional information on both the image and text sides. To address this, we propose FineMatch, a new aspect-based fine-grained text and image matching benchmark, focusing on text and image mismatch detection and correction. This benchmark introduces a novel task for boosting and evaluating the VLMs' compositionality for aspect-based fine-grained text and image matching. In this task, models are required to identify mismatched aspect phrases within a caption, determine the aspect's class, and propose corrections for an image-text pair that may contain between 0 and 3 mismatches. To evaluate the models' performance on this new task, we propose a new evaluation metric named ITM-IoU for which our experiments show a high correlation to human evaluation. In addition, we also provide a comprehensive experimental analysis of existing mainstream VLMs, including fully supervised learning and in-context learning settings. We have found that models trained on FineMatch demonstrate enhanced proficiency in detecting fine-grained text and image mismatches. Moreover, models (e.g., GPT-4V, Gemini Pro Vision) with strong abilities to perform multimodal in-context learning are not as skilled at fine-grained compositional image and text matching analysis. With FineMatch, we are able to build a system for text-to-image generation hallucination detection and correction.
Authors: Henry Peng Zou, Vinay Samuel, Yue Zhou, Weizhi Zhang, Liancheng Fang, Zihe Song, Philip S. Yu, Cornelia Caragea
Abstract: Existing datasets for attribute value extraction (AVE) predominantly focus on explicit attribute values while neglecting the implicit ones, lack product images, are often not publicly available, and lack an in-depth human inspection across diverse domains. To address these limitations, we present ImplicitAVE, the first, publicly available multimodal dataset for implicit attribute value extraction. ImplicitAVE, sourced from the MAVE dataset, is carefully curated and expanded to include implicit AVE and multimodality, resulting in a refined dataset of 68k training and 1.6k testing data across five domains. We also explore the application of multimodal large language models (MLLMs) to implicit AVE, establishing a comprehensive benchmark for MLLMs on the ImplicitAVE dataset. Six recent MLLMs with eleven variants are evaluated across diverse settings, revealing that implicit value extraction remains a challenging task for MLLMs. The contributions of this work include the development and release of ImplicitAVE, and the exploration and benchmarking of various MLLMs for implicit AVE, providing valuable insights and potential future research directions. Dataset and code are available at https://github.com/HenryPengZou/ImplicitAVE
Authors: Zhenrong Zhang, Jianan Liu, Xi Zhou, Tao Huang, Qing-Long Han, Jingxin Liu, Hongbin Liu
Abstract: Cooperative perception is essential to enhance the efficiency and safety of future transportation systems, requiring extensive data sharing among vehicles on the road, which raises significant privacy concerns. Federated learning offers a promising solution by enabling data privacy-preserving collaborative enhancements in perception, decision-making, and planning among connected and autonomous vehicles (CAVs). However, federated learning is impeded by significant challenges arising from data heterogeneity across diverse clients, potentially diminishing model accuracy and prolonging convergence periods. This study introduces a specialized federated learning framework for CP, termed the federated dynamic weighted aggregation (FedDWA) algorithm, facilitated by dynamic adjusting loss (DALoss) function. This framework employs dynamic client weighting to direct model convergence and integrates a novel loss function that utilizes Kullback-Leibler divergence (KLD) to counteract the detrimental effects of non-independently and identically distributed (Non-IID) and unbalanced data. Utilizing the BEV transformer as the primary model, our rigorous testing on the OpenV2V dataset, augmented with FedBEVT data, demonstrates significant improvements in the average intersection over union (IoU). These results highlight the substantial potential of our federated learning framework to address data heterogeneity challenges in CP, thereby enhancing the accuracy of environmental perception models and facilitating more robust and efficient collaborative learning solutions in the transportation sector.
Authors: Minghao Chen, Iro Laina, Andrea Vedaldi
Abstract: We consider the problem of editing 3D objects and scenes based on open-ended language instructions. A common approach to this problem is to use a 2D image generator or editor to guide the 3D editing process, obviating the need for 3D data. However, this process is often inefficient due to the need for iterative updates of costly 3D representations, such as neural radiance fields, either through individual view edits or score distillation sampling. A major disadvantage of this approach is the slow convergence caused by aggregating inconsistent information across views, as the guidance from 2D models is not multi-view consistent. We thus introduce the Direct Gaussian Editor (DGE), a method that addresses these issues in two stages. First, we modify a given high-quality image editor like InstructPix2Pix to be multi-view consistent. To do so, we propose a training-free approach that integrates cues from the 3D geometry of the underlying scene. Second, given a multi-view consistent edited sequence of images, we directly and efficiently optimize the 3D representation, which is based on 3D Gaussian Splatting. Because it avoids incremental and iterative edits, DGE is significantly more accurate and efficient than existing approaches and offers additional benefits, such as enabling selective editing of parts of the scene.
Authors: Mohamad Al Mdfaa, Raghad Salameh, Sergey Zagoruyko, Gonzalo Ferrer
Abstract: In the field of robotics and computer vision, efficient and accurate semantic mapping remains a significant challenge due to the growing demand for intelligent machines that can comprehend and interact with complex environments. Conventional panoptic mapping methods, however, are limited by predefined semantic classes, thus making them ineffective for handling novel or unforeseen objects. In response to this limitation, we introduce the Unified Promptable Panoptic Mapping (UPPM) method. UPPM utilizes recent advances in foundation models to enable real-time, on-demand label generation using natural language prompts. By incorporating a dynamic labeling strategy into traditional panoptic mapping techniques, UPPM provides significant improvements in adaptability and versatility while maintaining high performance levels in map reconstruction. We demonstrate our approach on real-world and simulated datasets. Results show that UPPM can accurately reconstruct scenes and segment objects while generating rich semantic labels through natural language interactions. A series of ablation experiments validated the advantages of foundation model-based labeling over fixed label sets.
Authors: Yunfeng Li, Bo Wang, Ye Li, Zhiwen Yu, Liang Wang
Abstract: How to better fuse cross-modal features is the core issue of RGB-T tracking. Some previous methods either insufficiently fuse RGB and TIR features, or depend on intermediaries containing information from both modalities to achieve cross-modal information interaction. The former does not fully exploit the potential of using only RGB and TIR information of the template or search region for channel and spatial feature fusion, and the latter lacks direct interaction between the template and search area, which limits the model's ability to fully exploit the original semantic information of both modalities. To alleviate these limitations, we explore how to improve the performance of a visual Transformer by using direct fusion of cross-modal channels and spatial features, and propose CSTNet. CSTNet uses ViT as a backbone and inserts cross-modal channel feature fusion modules (CFM) and cross-modal spatial feature fusion modules (SFM) for direct interaction between RGB and TIR features. The CFM performs parallel joint channel enhancement and joint multilevel spatial feature modeling of RGB and TIR features and sums the features, and then globally integrates the sum feature with the original features. The SFM uses cross-attention to model the spatial relationship of cross-modal features and then introduces a convolutional feedforward network for joint spatial and channel integration of multimodal features. We retrain the model with CSNet as the pre-training weights in the model with CFM and SFM removed, and propose CSTNet-small, which achieves 36% reduction in parameters and 24% reduction in Flops, and 50% speedup with a 1-2% performance decrease. Comprehensive experiments show that CSTNet achieves state-of-the-art performance on three public RGB-T tracking benchmarks. Code is available at https://github.com/LiYunfengLYF/CSTNet.
Authors: Huaiyuan Xu, Junliang Chen, Shiyu Meng, Yi Wang, Lap-Pui Chau
Abstract: 3D occupancy perception technology aims to observe and understand dense 3D environments for autonomous vehicles. Owing to its comprehensive perception capability, this technology is emerging as a trend in autonomous driving perception systems, and is attracting significant attention from both industry and academia. Similar to traditional bird's-eye view (BEV) perception, 3D occupancy perception has the nature of multi-source input and the necessity for information fusion. However, the difference is that it captures vertical structures that are ignored by 2D BEV. In this survey, we review the most recent works on 3D occupancy perception, and provide in-depth analyses of methodologies with various input modalities. Specifically, we summarize general network pipelines, highlight information fusion techniques, and discuss effective network training. We evaluate and analyze the occupancy perception performance of the state-of-the-art on the most popular datasets. Furthermore, challenges and future research directions are discussed. We hope this paper will inspire the community and encourage more research work on 3D occupancy perception. A comprehensive list of studies in this survey is publicly available in an active repository that continuously collects the latest work: https://github.com/HuaiyuanXu/3D-Occupancy-Perception.
URLs: https://github.com/HuaiyuanXu/3D-Occupancy-Perception.
Authors: Fotios Logothetis, Ignas Budvytis, Roberto Cipolla
Abstract: In this work we present a novel multi-view photometric stereo (MVPS) method. Like many works in 3D reconstruction we are leveraging neural shape representations and learnt renderers. However, our work differs from the state-of-the-art multi-view PS methods such as PS-NeRF or Supernormal in that we explicitly leverage per-pixel intensity renderings rather than relying mainly on estimated normals. We model point light attenuation and explicitly raytrace cast shadows in order to best approximate the incoming radiance for each point. The estimated incoming radiance is used as input to a fully neural material renderer that uses minimal prior assumptions and it is jointly optimised with the surface. Estimated normals and segmentation maps are also incorporated in order to maximise the surface accuracy. Our method is among the first (along with Supernormal) to outperform the classical MVPS approach proposed by the DiLiGenT-MV benchmark and achieves average 0.2mm Chamfer distance for objects imaged at approx 1.5m distance away with approximate 400x400 resolution. Moreover, our method shows high robustness to the sparse MVPS setup (6 views, 6 lights) greatly outperforming the SOTA competitor (0.38mm vs 0.61mm), illustrating the importance of neural rendering in multi-view photometric stereo.
Authors: Peng Gao, Yujian Lee, Hui Zhang, Xubo Liu, Yiyang Hu, Guquan Jing
Abstract: Visible-infrared person re-identification (VI-ReID) aims to match people with the same identity between visible and infrared modalities. VI-ReID is a challenging task due to the large differences in individual appearance under different modalities. Existing methods generally try to bridge the cross-modal differences at image or feature level, which lacks exploring the discriminative embeddings. Effectively minimizing these cross-modal discrepancies relies on obtaining representations that are guided by identity and consistent across modalities, while also filtering out representations that are irrelevant to identity. To address these challenges, we introduce a dynamic identity-guided attention network (DIAN) to mine identity-guided and modality-consistent embeddings, facilitating effective bridging the gap between different modalities. Specifically, in DIAN, to pursue a semantically richer representation, we first use orthogonal projection to fuse the features from two connected coarse and fine layers. Furthermore, we first use dynamic convolution kernels to mine identity-guided and modality-consistent representations. More notably, a cross embedding balancing loss is introduced to effectively bridge cross-modal discrepancies by above embeddings. Experimental results on SYSU-MM01 and RegDB datasets show that DIAN achieves state-of-the-art performance. Specifically, for indoor search on SYSU-MM01, our method achieves 86.28% rank-1 accuracy and 87.41% mAP, respectively. Our code will be available soon.
Authors: Raymond Wang, Nicholas R. Record, D. Whitney King, Tahiya Chowdhury
Abstract: Marine debris poses a significant ecological threat to birds, fish, and other animal life. Traditional methods for assessing debris accumulation involve labor-intensive and costly manual surveys. This study introduces a framework that utilizes aerial imagery captured by drones to conduct remote trash surveys. Leveraging computer vision techniques, our approach detects, classifies, and maps marine debris distributions. The framework uses Grounding DINO, a transformer-based zero-shot object detector, and CLIP, a vision-language model for zero-shot object classification, enabling the detection and classification of debris objects based on material type without the need for training labels. To mitigate over-counting due to different views of the same object, Scale-Invariant Feature Transform (SIFT) is employed for duplicate matching using local object features. Additionally, we have developed a user-friendly web application that facilitates end-to-end analysis of drone images, including object detection, classification, and visualization on a map to support cleanup efforts. Our method achieves competitive performance in detection (0.69 mean IoU) and classification (0.74 F1 score) across seven debris object classes without labeled data, comparable to state-of-the-art supervised methods. This framework has the potential to streamline automated trash sampling surveys, fostering efficient and sustainable community-led cleanup initiatives.
Authors: Li Zhang, Yuankun Yang, Ziyang Xie, Zhiyuan Yuan, Jianfeng Feng, Xiatian Zhu, Yu-Gang Jiang
Abstract: Understanding the hidden mechanisms behind human's visual perception is a fundamental question in neuroscience. To that end, investigating into the neural responses of human mind activities, such as functional Magnetic Resonance Imaging (fMRI), has been a significant research vehicle. However, analyzing fMRI signals is challenging, costly, daunting, and demanding for professional training. Despite remarkable progress in artificial intelligence (AI) based fMRI analysis, existing solutions are limited and far away from being biologically meaningful and practically useful. In this context, we leap forward to demonstrate how AI can go beyond the current state of the art by advancing from 2D visuals to visually plausible and functionally more comprehensive 3D visuals decoded from brain signals, enabling automatic more sophisticated modeling of fMRI data. Innovationally, we reformulate the task of analyzing fMRI data as a conditional 3D object generation problem. We design a novel 3D object representation learning method, Brain3D, that takes as input the fMRI data of a subject who was presented with a 2D image, and yields as output the corresponding 3D object visuals. Importantly, we show that our AI agent captures the distinct functionalities of each region of human vision system as well as their intricate interplay relationships, aligning remarkably with the established discoveries of neuroscience. Non-expert diagnosis indicate that \ourmodel{} can successfully identify the disordered brain regions in simulated scenarios, such as V1, V2, V3, V4, and the medial temporal lobe (MTL) within the human visual system. We also present results in cross-modal 3D visual generation setting, showcasing the perception quality of our 3D generation.
Authors: \'Alvaro Becerra, Javier Irigoyen, Roberto Daza, Ruth Cobos, Aythami Morales, Julian Fierrez, Mutlu Cukurova
Abstract: In this article, we explore computer vision approaches to detect abnormal head pose during e-learning sessions and we introduce a study on the effects of mobile phone usage during these sessions. We utilize behavioral data collected from 120 learners monitored while participating in a MOOC learning sessions. Our study focuses on the influence of phone-usage events on behavior and physiological responses, specifically attention, heart rate, and meditation, before, during, and after phone usage. Additionally, we propose an approach for estimating head pose events using images taken by the webcam during the MOOC learning sessions to detect phone-usage events. Our hypothesis suggests that head posture undergoes significant changes when learners interact with a mobile phone, contrasting with the typical behavior seen when learners face a computer during e-learning sessions. We propose an approach designed to detect deviations in head posture from the average observed during a learner's session, operating as a semi-supervised method. This system flags events indicating alterations in head posture for subsequent human review and selection of mobile phone usage occurrences with a sensitivity over 90%.
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: Hanjun Luo, Ziye Deng, Ruizhe Chen, Zuozhu Liu
Abstract: The rapid development and reduced barriers to entry for Text-to-Image (T2I) models have raised concerns about the biases in their outputs, but existing research lacks a holistic definition and evaluation framework of biases, limiting the enhancement of debiasing techniques. To address this issue, we introduce FAIntbench, a holistic and precise benchmark for biases in T2I models. In contrast to existing benchmarks that evaluate bias in limited aspects, FAIntbench evaluate biases from four dimensions: manifestation of bias, visibility of bias, acquired attributes, and protected attributes. We applied FAIntbench to evaluate seven recent large-scale T2I models and conducted human evaluation, whose results demonstrated the effectiveness of FAIntbench in identifying various biases. Our study also revealed new research questions about biases, including the side-effect of distillation. The findings presented here are preliminary, highlighting the potential of FAIntbench to advance future research aimed at mitigating the biases in T2I models. Our benchmark is publicly available to ensure the reproducibility.
Authors: Qilong Zhangli, Jindong Jiang, Di Liu, Licheng Yu, Xiaoliang Dai, Ankit Ramchandani, Guan Pang, Dimitris N. Metaxas, Praveen Krishnan
Abstract: While diffusion models have significantly advanced the quality of image generation their capability to accurately and coherently render text within these images remains a substantial challenge. Conventional diffusion-based methods for scene text generation are typically limited by their reliance on an intermediate layout output. This dependency often results in a constrained diversity of text styles and fonts an inherent limitation stemming from the deterministic nature of the layout generation phase. To address these challenges this paper introduces SceneTextGen a novel diffusion-based model specifically designed to circumvent the need for a predefined layout stage. By doing so SceneTextGen facilitates a more natural and varied representation of text. The novelty of SceneTextGen lies in its integration of three key components: a character-level encoder for capturing detailed typographic properties coupled with a character-level instance segmentation model and a word-level spotting model to address the issues of unwanted text generation and minor character inaccuracies. We validate the performance of our method by demonstrating improved character recognition rates on generated images across different public visual text datasets in comparison to both standard diffusion based methods and text specific methods.
Authors: Daizong Liu, Yang Liu, Wencan Huang, Wei Hu
Abstract: Text-guided 3D visual grounding (T-3DVG), which aims to locate a specific object that semantically corresponds to a language query from a complicated 3D scene, has drawn increasing attention in the 3D research community over the past few years. Compared to 2D visual grounding, this task presents great potential and challenges due to its closer proximity to the real world and the complexity of data collection and 3D point cloud source processing. In this survey, we attempt to provide a comprehensive overview of the T-3DVG progress, including its fundamental elements, recent research advances, and future research directions. To the best of our knowledge, this is the first systematic survey on the T-3DVG task. Specifically, we first provide a general structure of the T-3DVG pipeline with detailed components in a tutorial style, presenting a complete background overview. Then, we summarize the existing T-3DVG approaches into different categories and analyze their strengths and weaknesses. We also present the benchmark datasets and evaluation metrics to assess their performances. Finally, we discuss the potential limitations of existing T-3DVG and share some insights on several promising research directions. The latest papers are continually collected at https://github.com/liudaizong/Awesome-3D-Visual-Grounding.
URLs: https://github.com/liudaizong/Awesome-3D-Visual-Grounding.
Authors: Yiren Song, Shijie Huang, Chen Yao, Xiaojun Ye, Hai Ci, Jiaming Liu, Yuxuan Zhang, Mike Zheng Shou
Abstract: The painting process of artists is inherently stepwise and varies significantly among different painters and styles. Generating detailed, step-by-step painting processes is essential for art education and research, yet remains largely underexplored. Traditional stroke-based rendering methods break down images into sequences of brushstrokes, yet they fall short of replicating the authentic processes of artists, with limitations confined to basic brushstroke modifications. Text-to-image models utilizing diffusion processes generate images through iterative denoising, also diverge substantially from artists' painting process. To address these challenges, we introduce ProcessPainter, a text-to-video model that is initially pre-trained on synthetic data and subsequently fine-tuned with a select set of artists' painting sequences using the LoRA model. This approach successfully generates painting processes from text prompts for the first time. Furthermore, we introduce an Artwork Replication Network capable of accepting arbitrary-frame input, which facilitates the controlled generation of painting processes, decomposing images into painting sequences, and completing semi-finished artworks. This paper offers new perspectives and tools for advancing art education and image generation technology.
Authors: Yuqi Cheng, Yunkang Cao, Rui Chen, Weiming Shen
Abstract: Robustness against noisy imaging is crucial for practical image anomaly detection systems. This study introduces a Robust Anomaly Detection (RAD) dataset with free views, uneven illuminations, and blurry collections to systematically evaluate the robustness of current anomaly detection methods. Specifically, RAD aims to identify foreign objects on working platforms as anomalies. The collection process incorporates various sources of imaging noise, such as viewpoint changes, uneven illuminations, and blurry collections, to replicate real-world inspection scenarios. Subsequently, we assess and analyze 11 state-of-the-art unsupervised and zero-shot methods on RAD. Our findings indicate that: 1) Variations in viewpoint, illumination, and blurring affect anomaly detection methods to varying degrees; 2) Methods relying on memory banks and assisted by synthetic anomalies demonstrate stronger robustness; 3) Effectively leveraging the general knowledge of foundational models is a promising avenue for enhancing the robustness of anomaly detection methods. The dataset is available at https://github.com/hustCYQ/RAD-dataset.
Authors: Zhiyuan Ma, Liangliang Zhao, Biqing Qi, Bowen Zhou
Abstract: The most advanced diffusion models have recently adopted increasingly deep stacked networks (e.g., U-Net or Transformer) to promote the generative emergence capabilities of vision generation models similar to large language models (LLMs). However, progressively deeper stacked networks will intuitively cause numerical propagation errors and reduce noisy prediction capabilities on generative data, which hinders massively deep scalable training of vision generation models. In this paper, we first uncover the nature that neural networks being able to effectively perform generative denoising lies in the fact that the intrinsic residual unit has consistent dynamic property with the input signal's reverse diffusion process, thus supporting excellent generative abilities. Afterwards, we stand on the shoulders of two common types of deep stacked networks to propose a unified and massively scalable Neural Residual Diffusion Models framework (Neural-RDM for short), which is a simple yet meaningful change to the common architecture of deep generative networks by introducing a series of learnable gated residual parameters that conform to the generative dynamics. Experimental results on various generative tasks show that the proposed neural residual models obtain state-of-the-art scores on image's and video's generative benchmarks. Rigorous theoretical proofs and extensive experiments also demonstrate the advantages of this simple gated residual mechanism consistent with dynamic modeling in improving the fidelity and consistency of generated content and supporting large-scale scalable training. Code is available at https://github.com/Anonymous/Neural-RDM.
Authors: Mingshu Zhao, Yi Luo, Yong Ouyang
Abstract: In the realm of resource-constrained mobile vision tasks, the pursuit of efficiency and performance consistently drives innovation in lightweight Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). While ViTs excel at capturing global context through self-attention mechanisms, their deployment in resource-limited environments is hindered by computational complexity and latency. Conversely, lightweight CNNs are favored for their parameter efficiency and low latency. This study investigates the complementary advantages of CNNs and ViTs to develop a versatile vision backbone tailored for resource-constrained applications. We introduce RepNeXt, a novel model series integrates multi-scale feature representations and incorporates both serial and parallel structural reparameterization (SRP) to enhance network depth and width without compromising inference speed. Extensive experiments demonstrate RepNeXt's superiority over current leading lightweight CNNs and ViTs, providing advantageous latency across various vision benchmarks. RepNeXt-M4 matches RepViT-M1.5's 82.3\% accuracy on ImageNet within 1.5ms on an iPhone 12, outperforms its AP$^{box}$ by 1.3 on MS-COCO, and reduces parameters by 0.7M. Codes and models are available at https://github.com/suous/RepNeXt.
Authors: Ahmad Mohammadshirazi, Ali Nosrati Firoozsalari, Mengxi Zhou, Dheeraj Kulshrestha, Rajiv Ramnath
Abstract: Automating the annotation of scanned documents is challenging, requiring a balance between computational efficiency and accuracy. DocParseNet addresses this by combining deep learning and multi-modal learning to process both text and visual data. This model goes beyond traditional OCR and semantic segmentation, capturing the interplay between text and images to preserve contextual nuances in complex document structures. Our evaluations show that DocParseNet significantly outperforms conventional models, achieving mIoU scores of 49.12 on validation and 49.78 on the test set. This reflects a 58% accuracy improvement over state-of-the-art baseline models and an 18% gain compared to the UNext baseline. Remarkably, DocParseNet achieves these results with only 2.8 million parameters, reducing the model size by approximately 25 times and speeding up training by 5 times compared to other models. These metrics, coupled with a computational efficiency of 0.039 TFLOPs (BS=1), highlight DocParseNet's high performance in document annotation. The model's adaptability and scalability make it well-suited for real-world corporate document processing applications. The code is available at https://github.com/ahmad-shirazi/DocParseNet
Authors: Jiangtong Tan, Feng Zhao
Abstract: Image restoration aims to enhance low quality images, producing high quality images that exhibit natural visual characteristics and fine semantic attributes. Recently, the diffusion model has emerged as a powerful technique for image generation, and it has been explicitly employed as a backbone in image restoration tasks, yielding excellent results. However, it suffers from the drawbacks of slow inference speed and large model parameters due to its intrinsic characteristics. In this paper, we introduce a new perspective that implicitly leverages the diffusion model to assist the training of image restoration network, called DiffLoss, which drives the restoration results to be optimized for naturalness and semantic-aware visual effect. To achieve this, we utilize the mode coverage capability of the diffusion model to approximate the distribution of natural images and explore its ability to capture image semantic attributes. On the one hand, we extract intermediate noise to leverage its modeling capability of the distribution of natural images, which serves as a naturalness-oriented optimization space. On the other hand, we utilize the bottleneck features of diffusion model to harness its semantic attributes serving as a constraint on semantic level. By combining these two designs, the overall loss function is able to improve the perceptual quality of image restoration, resulting in visually pleasing and semantically enhanced outcomes. To validate the effectiveness of our method, we conduct experiments on various common image restoration tasks and benchmarks. Extensive experimental results demonstrate that our approach enhances the visual quality and semantic perception of the restoration network.
Authors: KuanChao Chu, Satoshi Yamazaki, Hideki Nakayama
Abstract: This work focuses on training dataset enhancement of informative relational triplets for Scene Graph Generation (SGG). Due to the lack of effective supervision, the current SGG model predictions perform poorly for informative relational triplets with inadequate training samples. Therefore, we propose two novel training dataset enhancement modules: Feature Space Triplet Augmentation (FSTA) and Soft Transfer. FSTA leverages a feature generator trained to generate representations of an object in relational triplets. The biased prediction based sampling in FSTA efficiently augments artificial triplets focusing on the challenging ones. In addition, we introduce Soft Transfer, which assigns soft predicate labels to general relational triplets to make more supervisions for informative predicate classes effectively. Experimental results show that integrating FSTA and Soft Transfer achieve high levels of both Recall and mean Recall in Visual Genome dataset. The mean of Recall and mean Recall is the highest among all the existing model-agnostic methods.
Authors: Ranjan Sapkota, Rizwan Qureshi, Marco Flores Calero, Chetan Badjugar, Upesh Nepal, Alwin Poulose, Peter Zeno, Uday Bhanu Prakash Vaddevolu, Sheheryar Khan, Maged Shoman, Hong Yan, Manoj Karkee
Abstract: This review systematically examines the progression of the You Only Look Once (YOLO) object detection algorithms from YOLOv1 to the recently unveiled YOLOv10. Employing a reverse chronological analysis, this study examines the advancements introduced by YOLO algorithms, beginning with YOLOv10 and progressing through YOLOv9, YOLOv8, and subsequent versions to explore each version's contributions to enhancing speed, accuracy, and computational efficiency in real-time object detection. The study highlights the transformative impact of YOLO across five critical application areas: automotive safety, healthcare, industrial manufacturing, surveillance, and agriculture. By detailing the incremental technological advancements in subsequent YOLO versions, this review chronicles the evolution of YOLO, and discusses the challenges and limitations in each earlier versions. The evolution signifies a path towards integrating YOLO with multimodal, context-aware, and General Artificial Intelligence (AGI) systems for the next YOLO decade, promising significant implications for future developments in AI-driven applications.
Authors: Wentong Li, Yuqian Yuan, Jian Liu, Dongqi Tang, Song Wang, Jianke Zhu, Lei Zhang
Abstract: The visual projector serves as an essential bridge between the visual encoder and the Large Language Model (LLM) in a Multimodal LLM (MLLM). Typically, MLLMs adopt a simple MLP to preserve all visual contexts via one-to-one transformation. However, the visual tokens are redundant and can be considerably increased when dealing with high-resolution images, impairing the efficiency of MLLMs significantly. Some recent works have introduced resampler or abstractor to reduce the number of resulting visual tokens. Unfortunately, they fail to capture finer details and undermine the visual reasoning capabilities of MLLMs. In this work, we propose a novel visual projector, which adopts a coarse-to-fine scheme to inject the enriched characteristics to generate the condensed visual tokens. In specific, we first interpolate the visual features as a low-resolution point query, providing the overall visual representation as the foundation. Then, we introduce a region-to-point injection module that utilizes high-resolution, multi-level region-based cues as fine-grained reference keys and values, allowing them to be fully absorbed within the corresponding local context region. This step effectively updates the coarse point query, transforming it into an enriched one for the subsequent LLM reasoning. Extensive experiments demonstrate that our approach compresses the visual tokens by 75%~89%, while achieves comparable or even better performance across diverse benchmarks with significantly higher efficiency. The source codes can be found at https://github.com/CircleRadon/TokenPacker.
Authors: Thong Nguyen, Yi Bin, Xiaobao Wu, Xinshuai Dong, Zhiyuan Hu, Khoi Le, Cong-Duy Nguyen, See-Kiong Ng, Luu Anh Tuan
Abstract: Data quality stands at the forefront of deciding the effectiveness of video-language representation learning. However, video-text pairs in previous data typically do not align perfectly with each other, which might lead to video-language representations that do not accurately reflect cross-modal semantics. Moreover, previous data also possess an uneven distribution of concepts, thereby hampering the downstream performance across unpopular subjects. To address these problems, we propose a contrastive objective with a subtractive angular margin to regularize cross-modal representations in their effort to reach perfect similarity. Furthermore, to adapt to the non-uniform concept distribution, we propose a multi-layer perceptron (MLP)-parameterized weighting function that maps loss values to sample weights which enable dynamic adjustment of the model's focus throughout the training. With the training guided by a small amount of unbiased meta-data and augmented by video-text data generated by large vision-language model, we improve video-language representations and achieve superior performances on commonly used video question answering and text-video retrieval datasets.
Authors: Elad Hirsch, Gefen Dawidowicz, Ayellet Tal
Abstract: Medical report generation from X-ray images is a challenging task, particularly in an unpaired setting where paired image-report data is unavailable for training. To address this challenge, we propose a novel model that leverages the available information in two distinct datasets, one comprising reports and the other consisting of images. The core idea of our model revolves around the notion that combining auto-encoding report generation with multi-modal (report-image) alignment can offer a solution. However, the challenge persists regarding how to achieve this alignment when pair correspondence is absent. Our proposed solution involves the use of auxiliary tasks, particularly contrastive learning and classification, to position related images and reports in close proximity to each other. This approach differs from previous methods that rely on pre-processing steps, such as using external information stored in a knowledge graph. Our model, named MedRAT, surpasses previous state-of-the-art methods, demonstrating the feasibility of generating comprehensive medical reports without the need for paired data or external tools.
Authors: Ananthu Aniraj, Cassio F. Dantas, Dino Ienco, Diego Marcos
Abstract: Computer vision methods that explicitly detect object parts and reason on them are a step towards inherently interpretable models. Existing approaches that perform part discovery driven by a fine-grained classification task make very restrictive assumptions on the geometric properties of the discovered parts; they should be small and compact. Although this prior is useful in some cases, in this paper we show that pre-trained transformer-based vision models, such as self-supervised DINOv2 ViT, enable the relaxation of these constraints. In particular, we find that a total variation (TV) prior, which allows for multiple connected components of any size, substantially outperforms previous work. We test our approach on three fine-grained classification benchmarks: CUB, PartImageNet and Oxford Flowers, and compare our results to previously published methods as well as a re-implementation of the state-of-the-art method PDiscoNet with a transformer-based backbone. We consistently obtain substantial improvements across the board, both on part discovery metrics and the downstream classification task, showing that the strong inductive biases in self-supervised ViT models require to rethink the geometric priors that can be used for unsupervised part discovery.
Authors: Hannah Lee, Changyeon Lee, Kevin Farhat, Lin Qiu, Steve Geluso, Aerin Kim, Oren Etzioni
Abstract: Multimodal generative models are rapidly evolving, leading to a surge in the generation of realistic video and audio that offers exciting possibilities but also serious risks. Deepfake videos, which can convincingly impersonate individuals, have particularly garnered attention due to their potential misuse in spreading misinformation and creating fraudulent content. This survey paper examines the dual landscape of deepfake video generation and detection, emphasizing the need for effective countermeasures against potential abuses. We provide a comprehensive overview of current deepfake generation techniques, including face swapping, reenactment, and audio-driven animation, which leverage cutting-edge technologies like GANs and diffusion models to produce highly realistic fake videos. Additionally, we analyze various detection approaches designed to differentiate authentic from altered videos, from detecting visual artifacts to deploying advanced algorithms that pinpoint inconsistencies across video and audio signals. The effectiveness of these detection methods heavily relies on the diversity and quality of datasets used for training and evaluation. We discuss the evolution of deepfake datasets, highlighting the importance of robust, diverse, and frequently updated collections to enhance the detection accuracy and generalizability. As deepfakes become increasingly indistinguishable from authentic content, developing advanced detection techniques that can keep pace with generation technologies is crucial. We advocate for a proactive approach in the "tug-of-war" between deepfake creators and detectors, emphasizing the need for continuous research collaboration, standardization of evaluation metrics, and the creation of comprehensive benchmarks.
Authors: Yiran Yang, Jinchao Zhang, Ying Deng, Jie Zhou
Abstract: Inspired by the success of the text-to-image (T2I) generation task, many researchers are devoting themselves to the text-to-video (T2V) generation task. Most of the T2V frameworks usually inherit from the T2I model and add extra-temporal layers of training to generate dynamic videos, which can be viewed as a fine-tuning task. However, the traditional 3D-Unet is a serial mode and the temporal layers follow the spatial layers, which will result in high GPU memory and training time consumption according to its serial feature flow. We believe that this serial mode will bring more training costs with the large diffusion model and massive datasets, which are not environmentally friendly and not suitable for the development of the T2V. Therefore, we propose a highly efficient spatial-temporal parallel training paradigm for T2V tasks, named Mobius. In our 3D-Unet, the temporal layers and spatial layers are parallel, which optimizes the feature flow and backpropagation. The Mobius will save 24% GPU memory and 12% training time, which can greatly improve the T2V fine-tuning task and provide a novel insight for the AIGC community. We will release our codes in the future.
Authors: Yang Liu, Weixing Chen, Yongjie Bai, Guanbin Li, Wen Gao, Liang Lin
Abstract: Embodied Artificial Intelligence (Embodied AI) is crucial for achieving Artificial General Intelligence (AGI) and serves as a foundation for various applications that bridge cyberspace and the physical world. Recently, the emergence of Multi-modal Large Models (MLMs) and World Models (WMs) have attracted significant attention due to their remarkable perception, interaction, and reasoning capabilities, making them a promising architecture for the brain of embodied agents. However, there is no comprehensive survey for Embodied AI in the era of MLMs. In this survey, we give a comprehensive exploration of the latest advancements in Embodied AI. Our analysis firstly navigates through the forefront of representative works of embodied robots and simulators, to fully understand the research focuses and their limitations. Then, we analyze four main research targets: 1) embodied perception, 2) embodied interaction, 3) embodied agent, and 4) sim-to-real adaptation, covering the state-of-the-art methods, essential paradigms, and comprehensive datasets. Additionally, we explore the complexities of MLMs in virtual and real embodied agents, highlighting their significance in facilitating interactions in dynamic digital and physical environments. Finally, we summarize the challenges and limitations of embodied AI and discuss their potential future directions. We hope this survey will serve as a foundational reference for the research community and inspire continued innovation. The associated project can be found at https://github.com/HCPLab-SYSU/Embodied_AI_Paper_List.
URLs: https://github.com/HCPLab-SYSU/Embodied_AI_Paper_List.
Authors: Hao Wang, Pengzhen Ren, Zequn Jie, Xiao Dong, Chengjian Feng, Yinlong Qian, Lin Ma, Dongmei Jiang, Yaowei Wang, Xiangyuan Lan, Xiaodan Liang
Abstract: Open-vocabulary detection is a challenging task due to the requirement of detecting objects based on class names, including those not encountered during training. Existing methods have shown strong zero-shot detection capabilities through pre-training and pseudo-labeling on diverse large-scale datasets. However, these approaches encounter two main challenges: (i) how to effectively eliminate data noise from pseudo-labeling, and (ii) how to efficiently leverage the language-aware capability for region-level cross-modality fusion and alignment. To address these challenges, we propose a novel unified open-vocabulary detection method called OV-DINO, which is pre-trained on diverse large-scale datasets with language-aware selective fusion in a unified framework. Specifically, we introduce a Unified Data Integration (UniDI) pipeline to enable end-to-end training and eliminate noise from pseudo-label generation by unifying different data sources into detection-centric data format. In addition, we propose a Language-Aware Selective Fusion (LASF) module to enhance the cross-modality alignment through a language-aware query selection and fusion process. We evaluate the performance of the proposed OV-DINO on popular open-vocabulary detection benchmarks, achieving state-of-the-art results with an AP of 50.6% on the COCO benchmark and 40.1% on the LVIS benchmark in a zero-shot manner, demonstrating its strong generalization ability. Furthermore, the fine-tuned OV-DINO on COCO achieves 58.4% AP, outperforming many existing methods with the same backbone. The code for OV-DINO is available at https://github.com/wanghao9610/OV-DINO.
Authors: Liang Wu, Wenjing Lu, Liming Tang, Zhuang Fang
Abstract: The Retinex theory models the image as a product of illumination and reflection components, which has received extensive attention and is widely used in image enhancement, segmentation and color restoration. However, it has been rarely used in additive noise removal due to the inclusion of both multiplication and addition operations in the Retinex noisy image modeling. In this paper, we propose an exponential Retinex decomposition model based on hybrid non-convex regularization and weak space oscillation-modeling for image denoising. The proposed model utilizes non-convex first-order total variation (TV) and non-convex second-order TV to regularize the reflection component and the illumination component, respectively, and employs weak $H^{-1}$ norm to measure the residual component. By utilizing different regularizers, the proposed model effectively decomposes the image into reflection, illumination, and noise components. An alternating direction multipliers method (ADMM) combined with the Majorize-Minimization (MM) algorithm is developed to solve the proposed model. Furthermore, we provide a detailed proof of the convergence property of the algorithm. Numerical experiments validate both the proposed model and algorithm. Compared with several state-of-the-art denoising models, the proposed model exhibits superior performance in terms of peak signal-to-noise ratio (PSNR) and mean structural similarity (MSSIM).
Authors: Wenjing Lu, Zhuang Fang, Liang Wu, Liming Tang, Hanxin Liu, Chuanjiang He
Abstract: The low-rank matrix completion (LRMC) technology has achieved remarkable results in low-level visual tasks. There is an underlying assumption that the real-world matrix data is low-rank in LRMC. However, the real matrix data does not satisfy the strict low-rank property, which undoubtedly present serious challenges for the above-mentioned matrix recovery methods. Fortunately, there are feasible schemes that devise appropriate and effective priori representations for describing the intrinsic information of real data. In this paper, we firstly model the matrix data ${\bf{Y}}$ as the sum of a low-rank approximation component $\bf{X}$ and an approximation error component $\cal{E}$. This finer-grained data decomposition architecture enables each component of information to be portrayed more precisely. Further, we design an overlapping group error representation (OGER) function to characterize the above error structure and propose a generalized low-rank matrix completion model based on OGER. Specifically, the low-rank component describes the global structure information of matrix data, while the OGER component not only compensates for the approximation error between the low-rank component and the real data but also better captures the local block sparsity information of matrix data. Finally, we develop an alternating direction method of multipliers (ADMM) that integrates the majorization-minimization (MM) algorithm, which enables the efficient solution of the proposed model. And we analyze the convergence of the algorithm in detail both theoretically and experimentally. In addition, the results of numerical experiments demonstrate that the proposed model outperforms existing competing models in performance.
Authors: Zujin Guo, Wei Li, Chen Change Loy
Abstract: Motion modeling is critical in flow-based Video Frame Interpolation (VFI). Existing paradigms either consider linear combinations of bidirectional flows or directly predict bilateral flows for given timestamps without exploring favorable motion priors, thus lacking the capability of effectively modeling spatiotemporal dynamics in real-world videos. To address this limitation, in this study, we introduce Generalizable Implicit Motion Modeling (GIMM), a novel and effective approach to motion modeling for VFI. Specifically, to enable GIMM as an effective motion modeling paradigm, we design a motion encoding pipeline to model spatiotemporal motion latent from bidirectional flows extracted from pre-trained flow estimators, effectively representing input-specific motion priors. Then, we implicitly predict arbitrary-timestep optical flows within two adjacent input frames via an adaptive coordinate-based neural network, with spatiotemporal coordinates and motion latent as inputs. Our GIMM can be smoothly integrated with existing flow-based VFI works without further modifications. We show that GIMM performs better than the current state of the art on the VFI benchmarks.
Authors: Shuang Wang, Qingchuan Tao, Zhenming Tang
Abstract: This study presents ResVMUNetX, a novel image enhancement network for low-light conditions, addressing the limitations of existing deep learning methods in capturing long-range image information. Leveraging error regression and an efficient VMamba architecture, ResVMUNetX enhances brightness, recovers structural details, and removes noise through a two-step process involving direct pixel addition and a specialized Denoise CNN module. Demonstrating superior performance on the LOL dataset, ResVMUNetX significantly improves image clarity and quality with reduced computational demands, achieving real-time processing speeds of up to 70 frames per second. This confirms its effectiveness in enhancing low-light images and its potential for practical, real-time applications.
Authors: Hao Fang, Jiawei Kong, Bin Chen, Tao Dai, Hao Wu, Shu-Tao Xia
Abstract: Transferable targeted adversarial attacks aim to mislead models into outputting adversary-specified predictions in black-box scenarios. Recent studies have introduced \textit{single-target} generative attacks that train a generator for each target class to generate highly transferable perturbations, resulting in substantial computational overhead when handling multiple classes. \textit{Multi-target} attacks address this by training only one class-conditional generator for multiple classes. However, the generator simply uses class labels as conditions, failing to leverage the rich semantic information of the target class. To this end, we design a \textbf{C}LIP-guided \textbf{G}enerative \textbf{N}etwork with \textbf{C}ross-attention modules (CGNC) to enhance multi-target attacks by incorporating textual knowledge of CLIP into the generator. Extensive experiments demonstrate that CGNC yields significant improvements over previous multi-target generative attacks, e.g., a 21.46\% improvement in success rate from ResNet-152 to DenseNet-121. Moreover, we propose a masked fine-tuning mechanism to further strengthen our method in attacking a single class, which surpasses existing single-target methods.
Authors: Yuchen Yang, Kwonjoon Lee, Behzad Dariush, Yinzhi Cao, Shao-Yuan Lo
Abstract: Video Anomaly Detection (VAD) is crucial for applications such as security surveillance and autonomous driving. However, existing VAD methods provide little rationale behind detection, hindering public trust in real-world deployments. In this paper, we approach VAD with a reasoning framework. Although Large Language Models (LLMs) have shown revolutionary reasoning ability, we find that their direct use falls short of VAD. Specifically, the implicit knowledge pre-trained in LLMs focuses on general context and thus may not apply to every specific real-world VAD scenario, leading to inflexibility and inaccuracy. To address this, we propose AnomalyRuler, a novel rule-based reasoning framework for VAD with LLMs. AnomalyRuler comprises two main stages: induction and deduction. In the induction stage, the LLM is fed with few-shot normal reference samples and then summarizes these normal patterns to induce a set of rules for detecting anomalies. The deduction stage follows the induced rules to spot anomalous frames in test videos. Additionally, we design rule aggregation, perception smoothing, and robust reasoning strategies to further enhance AnomalyRuler's robustness. AnomalyRuler is the first reasoning approach for the one-class VAD task, which requires only few-normal-shot prompting without the need for full-shot training, thereby enabling fast adaption to various VAD scenarios. Comprehensive experiments across four VAD benchmarks demonstrate AnomalyRuler's state-of-the-art detection performance and reasoning ability. AnomalyRuler is open-source and available at: https://github.com/Yuchen413/AnomalyRuler
Authors: Ziheng Chen, Yue Song, Xiao-Jun Wu, Gaowen Liu, Nicu Sebe
Abstract: Global Covariance Pooling (GCP) has been demonstrated to improve the performance of Deep Neural Networks (DNNs) by exploiting second-order statistics of high-level representations. GCP typically performs classification of the covariance matrices by applying matrix function normalization, such as matrix logarithm or power, followed by a Euclidean classifier. However, covariance matrices inherently lie in a Riemannian manifold, known as the Symmetric Positive Definite (SPD) manifold. The current literature does not provide a satisfactory explanation of why Euclidean classifiers can be applied directly to Riemannian features after the normalization of the matrix power. To mitigate this gap, this paper provides a comprehensive and unified understanding of the matrix logarithm and power from a Riemannian geometry perspective. The underlying mechanism of matrix functions in GCP is interpreted from two perspectives: one based on tangent classifiers (Euclidean classifiers on the tangent space) and the other based on Riemannian classifiers. Via theoretical analysis and empirical validation through extensive experiments on fine-grained and large-scale visual classification datasets, we conclude that the working mechanism of the matrix functions should be attributed to the Riemannian classifiers they implicitly respect.
Authors: Tianxiang Hao, Xiaohan Ding, Juexiao Feng, Yuhong Yang, Hui Chen, Guiguang Ding
Abstract: In the past few years, large-scale pre-trained vision-language models like CLIP have achieved tremendous success in various fields. Naturally, how to transfer the rich knowledge in such huge pre-trained models to downstream tasks and datasets becomes a hot topic. During downstream adaptation, the most challenging problems are overfitting and catastrophic forgetting, which can cause the model to overly focus on the current data and lose more crucial domain-general knowledge. Existing works use classic regularization techniques to solve the problems. As solutions become increasingly complex, the ever-growing storage and inference costs are also a significant problem that urgently needs to be addressed. While in this paper, we start from an observation that proper random noise can suppress overfitting and catastrophic forgetting. Then we regard quantization error as a kind of noise, and explore quantization for regularizing vision-language model, which is quite efficiency and effective. Furthermore, to improve the model's generalization capability while maintaining its specialization capacity at minimal cost, we deeply analyze the characteristics of the weight distribution in prompts, conclude several principles for quantization module design and follow such principles to create several competitive baselines. The proposed method is significantly efficient due to its inherent lightweight nature, making it possible to adapt on extremely resource-limited devices. Our method can be fruitfully integrated into many existing approaches like MaPLe, enhancing accuracy while reducing storage overhead, making it more powerful yet versatile. Extensive experiments on 11 datasets shows great superiority of our method sufficiently. Code is available at https://github.com/beyondhtx/QPrompt.
Authors: Chunliang Li, Wencheng Han, Junbo Yin, Sanyuan Zhao, Jianbing Shen
Abstract: Concurrent processing of multiple autonomous driving 3D perception tasks within the same spatiotemporal scene poses a significant challenge, in particular due to the computational inefficiencies and feature competition between tasks when using traditional multi-task learning approaches. This paper addresses these issues by proposing a novel unified representation, RepVF, which harmonizes the representation of various perception tasks such as 3D object detection and 3D lane detection within a single framework. RepVF characterizes the structure of different targets in the scene through a vector field, enabling a single-head, multi-task learning model that significantly reduces computational redundancy and feature competition. Building upon RepVF, we introduce RFTR, a network designed to exploit the inherent connections between different tasks by utilizing a hierarchical structure of queries that implicitly model the relationships both between and within tasks. This approach eliminates the need for task-specific heads and parameters, fundamentally reducing the conflicts inherent in traditional multi-task learning paradigms. We validate our approach by combining labels from the OpenLane dataset with the Waymo Open dataset. Our work presents a significant advancement in the efficiency and effectiveness of multi-task perception in autonomous driving, offering a new perspective on handling multiple 3D perception tasks synchronously and in parallel. The code will be available at: https://github.com/jbji/RepVF
Authors: Yucheng Lu, Dovile Juodelyte, Jonathan D. Victor, Veronika Cheplygina
Abstract: In this paper, we use spectral analysis to investigate transfer learning and study model sensitivity to frequency shortcuts in medical imaging. By analyzing the power spectrum density of both pre-trained and fine-tuned model gradients, as well as artificially generated frequency shortcuts, we observe notable differences in learning priorities between models pre-trained on natural vs medical images, which generally persist during fine-tuning. We find that when a model's learning priority aligns with the power spectrum density of an artifact, it results in overfitting to that artifact. Based on these observations, we show that source data editing can alter the model's resistance to shortcut learning.
Authors: Boqiang Zhang, Zuan Gao, Yadong Qu, Hongtao Xie
Abstract: Visual text generation has significantly advanced through diffusion models aimed at producing images with readable and realistic text. Recent works primarily use a ControlNet-based framework, employing standard font text images to control diffusion models. Recognizing the critical role of control information in generating high-quality text, we investigate its influence from three perspectives: input encoding, role at different stages, and output features. Our findings reveal that: 1) Input control information has unique characteristics compared to conventional inputs like Canny edges and depth maps. 2) Control information plays distinct roles at different stages of the denoising process. 3) Output control features significantly differ from the base and skip features of the U-Net decoder in the frequency domain. Based on these insights, we propose TextGen, a novel framework designed to enhance generation quality by optimizing control information. We improve input and output features using Fourier analysis to emphasize relevant information and reduce noise. Additionally, we employ a two-stage generation framework to align the different roles of control information at different stages. Furthermore, we introduce an effective and lightweight dataset for training. Our method achieves state-of-the-art performance in both Chinese and English text generation. The code and dataset available at https://github.com/CyrilSterling/TextGen.
Authors: Pengyu Zhang, Hao Yin, Zeren Wang, Wenyue Chen, Shengming Li, Dong Wang, Huchuan Lu, Xu Jia
Abstract: Sign language is one of the most effective communication tools for people with hearing difficulties. Most existing works focus on improving the performance of sign language tasks on RGB videos, which may suffer from degraded recording conditions, such as fast movement of hands with motion blur and textured signer's appearance. The bio-inspired event camera, which asynchronously captures brightness change with high speed, could naturally perceive dynamic hand movements, providing rich manual clues for sign language tasks. In this work, we aim at exploring the potential of event camera in continuous sign language recognition (CSLR) and sign language translation (SLT). To promote the research, we first collect an event-based benchmark EvSign for those tasks with both gloss and spoken language annotations. EvSign dataset offers a substantial amount of high-quality event streams and an extensive vocabulary of glosses and words, thereby facilitating the development of sign language tasks. In addition, we propose an efficient transformer-based framework for event-based SLR and SLT tasks, which fully leverages the advantages of streaming events. The sparse backbone is employed to extract visual features from sparse events. Then, the temporal coherence is effectively utilized through the proposed local token fusion and gloss-aware temporal aggregation modules. Extensive experimental results are reported on both simulated (PHOENIX14T) and EvSign datasets. Our method performs favorably against existing state-of-the-art approaches with only 0.34% computational cost (0.84G FLOPS per video) and 44.2% network parameters. The project is available at https://zhang-pengyu.github.io/EVSign.
Authors: Mohammad Erfan Sadeghi, Arash Fayyazi, Suhas Somashekar, Massoud Pedram
Abstract: Vision Transformers (ViTs) represent a groundbreaking shift in machine learning approaches to computer vision. Unlike traditional approaches, ViTs employ the self-attention mechanism, which has been widely used in natural language processing, to analyze image patches. Despite their advantages in modeling visual tasks, deploying ViTs on hardware platforms, notably Field-Programmable Gate Arrays (FPGAs), introduces considerable challenges. These challenges stem primarily from the non-linear calculations and high computational and memory demands of ViTs. This paper introduces CHOSEN, a software-hardware co-design framework to address these challenges and offer an automated framework for ViT deployment on the FPGAs in order to maximize performance. Our framework is built upon three fundamental contributions: multi-kernel design to maximize the bandwidth, mainly targeting benefits of multi DDR memory banks, approximate non-linear functions that exhibit minimal accuracy degradation, and efficient use of available logic blocks on the FPGA, and efficient compiler to maximize the performance and memory-efficiency of the computing kernels by presenting a novel algorithm for design space exploration to find optimal hardware configuration that achieves optimal throughput and latency. Compared to the state-of-the-art ViT accelerators, CHOSEN achieves a 1.5x and 1.42x improvement in the throughput on the DeiT-S and DeiT-B models.
Authors: Sherwin Bahmani, Ivan Skorokhodov, Aliaksandr Siarohin, Willi Menapace, Guocheng Qian, Michael Vasilkovsky, Hsin-Ying Lee, Chaoyang Wang, Jiaxu Zou, Andrea Tagliasacchi, David B. Lindell, Sergey Tulyakov
Abstract: Modern text-to-video synthesis models demonstrate coherent, photorealistic generation of complex videos from a text description. However, most existing models lack fine-grained control over camera movement, which is critical for downstream applications related to content creation, visual effects, and 3D vision. Recently, new methods demonstrate the ability to generate videos with controllable camera poses these techniques leverage pre-trained U-Net-based diffusion models that explicitly disentangle spatial and temporal generation. Still, no existing approach enables camera control for new, transformer-based video diffusion models that process spatial and temporal information jointly. Here, we propose to tame video transformers for 3D camera control using a ControlNet-like conditioning mechanism that incorporates spatiotemporal camera embeddings based on Plucker coordinates. The approach demonstrates state-of-the-art performance for controllable video generation after fine-tuning on the RealEstate10K dataset. To the best of our knowledge, our work is the first to enable camera control for transformer-based video diffusion models.
Authors: Shuchang Zhang, Hui Zhang, Hongxia Wang
Abstract: Preconditioned Proximal Point (PPP) algorithms provide a unified framework for splitting methods in image restoration. Recent advancements with RED (Regularization by Denoising) and PnP (Plug-and-Play) priors have achieved state-of-the-art performance in this domain, emphasizing the need for a meaningful particular solution. However, degenerate PPP algorithms typically exhibit weak convergence in infinite-dimensional Hilbert space, leading to uncertain solutions. To address this issue, we propose the Halpern-type Preconditioned Proximal Point (HPPP) algorithm, which leverages the strong convergence properties of Halpern iteration to achieve a particular solution. Based on the implicit regularization defined by gradient RED, we further introduce the Gradient REgularization by Denoising via HPPP called GraRED-HP3 algorithm. The HPPP algorithm is shown to have the regularity converging to a particular solution by a toy example. Additionally, experiments in image deblurring and inpainting validate the effectiveness of GraRED-HP3, showing it surpasses classical methods such as Chambolle-Pock (CP), PPP, RED, and RED-PRO.
Authors: Yulin He, Wei Chen, Tianci Xun, Yusong Tan
Abstract: Occupancy prediction plays a pivotal role in autonomous driving (AD) due to the fine-grained geometric perception and general object recognition capabilities. However, existing methods often incur high computational costs, which contradicts the real-time demands of AD. To this end, we first evaluate the speed and memory usage of most public available methods, aiming to redirect the focus from solely prioritizing accuracy to also considering efficiency. We then identify a core challenge in achieving both fast and accurate performance: \textbf{the strong coupling between geometry and semantic}. To address this issue, 1) we propose a Geometric-Semantic Dual-Branch Network (GSDBN) with a hybrid BEV-Voxel representation. In the BEV branch, a BEV-level temporal fusion module and a U-Net encoder is introduced to extract dense semantic features. In the voxel branch, a large-kernel re-parameterized 3D convolution is proposed to refine sparse 3D geometry and reduce computation. Moreover, we propose a novel BEV-Voxel lifting module that projects BEV features into voxel space for feature fusion of the two branches. In addition to the network design, 2) we also propose a Geometric-Semantic Decoupled Learning (GSDL) strategy. This strategy initially learns semantics with accurate geometry using ground-truth depth, and then gradually mixes predicted depth to adapt the model to the predicted geometry. Extensive experiments on the widely-used Occ3D-nuScenes benchmark demonstrate the superiority of our method, which achieves a 39.4 mIoU with 20.0 FPS. This result is $\sim 3 \times$ faster and +1.9 mIoU higher compared to FB-OCC, the winner of CVPR2023 3D Occupancy Prediction Challenge. Our code will be made open-source.
Authors: Pei-Kai Huang, Tzu-Hsien Chen, Ya-Ting Chan, Kuan-Wen Chen, Chiou-Ting Hsu
Abstract: Many remote photoplethysmography (rPPG) estimation models have achieved promising performance on the training domain but often fail to measure the physiological signals or heart rates (HR) on test domains. Domain generalization (DG) or domain adaptation (DA) techniques are therefore adopted in the offline training stage to adapt the model to the unobserved or observed test domain by referring to all the available source domain data. However, in rPPG estimation problems, the adapted model usually confronts challenges of estimating target data with various domain information, such as different video capturing settings, individuals of different age ranges, or of different HR distributions. In contrast, Test-Time Adaptation (TTA), by online adapting to unlabeled target data without referring to any source data, enables the model to adaptively estimate rPPG signals of various unseen domains. In this paper, we first propose a novel TTA-rPPG benchmark, which encompasses various domain information and HR distributions, to simulate the challenges encountered in rPPG estimation. Next, we propose a novel synthetic signal-guided rPPG estimation framework with a two-fold purpose. First, we design an effective spectral-based entropy minimization to enforce the rPPG model to learn new target domain information. Second, we develop a synthetic signal-guided feature learning, by synthesizing pseudo rPPG signals as pseudo ground-truths to guide a conditional generator to generate latent rPPG features. The synthesized rPPG signals and the generated rPPG features are used to guide the rPPG model to broadly cover various HR distributions. Our extensive experiments on the TTA-rPPG benchmark show that the proposed method achieves superior performance and outperforms previous DG and DA methods across most protocols of the proposed TTA-rPPG benchmark.
Authors: Zongrui Li, Minghui Hu, Qian Zheng, Xudong Jiang
Abstract: Although recent advancements in text-to-3D generation have significantly improved generation quality, issues like limited level of detail and low fidelity still persist, which requires further improvement. To understand the essence of those issues, we thoroughly analyze current score distillation methods by connecting theories of consistency distillation to score distillation. Based on the insights acquired through analysis, we propose an optimization framework, Guided Consistency Sampling (GCS), integrated with 3D Gaussian Splatting (3DGS) to alleviate those issues. Additionally, we have observed the persistent oversaturation in the rendered views of generated 3D assets. From experiments, we find that it is caused by unwanted accumulated brightness in 3DGS during optimization. To mitigate this issue, we introduce a Brightness-Equalized Generation (BEG) scheme in 3DGS rendering. Experimental results demonstrate that our approach generates 3D assets with more details and higher fidelity than state-of-the-art methods. The codes are released at https://github.com/LMozart/ECCV2024-GCS-BEG.
Authors: Wei Zhang, Miaoxin Cai, Tong Zhang, Jun Li, Yin Zhuang, Xuerui Mao
Abstract: Recent advances in visual prompting in the natural image area have allowed users to interact with artificial intelligence (AI) tools through various visual marks such as box, point, and free-form shapes. However, due to the significant difference between the natural and remote sensing (RS) images, existing visual prompting models face challenges in RS scenarios. Moreover, RS MLLMs mainly focus on interpreting image-level RS data and only support interaction with language instruction, restricting flexibility applications in the real world. To address those limitations, the first visual prompting model named EarthMarker is proposed, which excels in image-level, region-level, and point-level RS imagery interpretation. Specifically, the visual prompts alongside images and text instruction input into the large language model (LLM), adapt models toward specific predictions and tasks. Subsequently, a sharing visual encoding method is introduced to refine multi-scale image features and visual prompt information uniformly. Furthermore, to endow the EarthMarker with versatile multi-granularity visual perception abilities, the cross-domain phased learning strategy is developed, and the disjoint parameters are optimized in a lightweight manner by leveraging both the natural and RS domain-specific knowledge. In addition, to tackle the lack of RS visual prompting data, a dataset named RSVP featuring multi-modal fine-grained visual prompting instruction is constructed. Extensive experiments are conducted to demonstrate the proposed EarthMarker's competitive performance, representing a significant advance in multi-granularity RS imagery interpretation under the visual prompting learning framework.
Authors: Yixiang Qiu, Hao Fang, Hongyao Yu, Bin Chen, MeiKang Qiu, Shu-Tao Xia
Abstract: Model Inversion (MI) attacks aim to reconstruct privacy-sensitive training data from released models by utilizing output information, raising extensive concerns about the security of Deep Neural Networks (DNNs). Recent advances in generative adversarial networks (GANs) have contributed significantly to the improved performance of MI attacks due to their powerful ability to generate realistic images with high fidelity and appropriate semantics. However, previous MI attacks have solely disclosed private information in the latent space of GAN priors, limiting their semantic extraction and transferability across multiple target models and datasets. To address this challenge, we propose a novel method, Intermediate Features enhanced Generative Model Inversion (IF-GMI), which disassembles the GAN structure and exploits features between intermediate blocks. This allows us to extend the optimization space from latent code to intermediate features with enhanced expressive capabilities. To prevent GAN priors from generating unrealistic images, we apply a L1 ball constraint to the optimization process. Experiments on multiple benchmarks demonstrate that our method significantly outperforms previous approaches and achieves state-of-the-art results under various settings, especially in the out-of-distribution (OOD) scenario. Our code is available at: https://github.com/final-solution/IF-GMI
Authors: Wenxuan Lu, Mengshun Hu, Yansheng Qiu, Liang Liao, Zheng Wang
Abstract: With the development of VR-related techniques, viewers can enjoy a realistic and immersive experience through a head-mounted display, while omnidirectional video with a low frame rate can lead to user dizziness. However, the prevailing plane frame interpolation methodologies are unsuitable for Omnidirectional Video Interpolation, chiefly due to the lack of models tailored to such videos with strong distortion, compounded by the scarcity of valuable datasets for Omnidirectional Video Frame Interpolation. In this paper, we introduce the benchmark dataset, 360VFI, for Omnidirectional Video Frame Interpolation. We present a practical implementation that introduces a distortion prior from omnidirectional video into the network to modulate distortions. We especially propose a pyramid distortion-sensitive feature extractor that uses the unique characteristics of equirectangular projection (ERP) format as prior information. Moreover, we devise a decoder that uses an affine transformation to facilitate the synthesis of intermediate frames further. 360VFI is the first dataset and benchmark that explores the challenge of Omnidirectional Video Frame Interpolation. Through our benchmark analysis, we presented four different distortion conditions scenes in the proposed 360VFI dataset to evaluate the challenge triggered by distortion during interpolation. Besides, experimental results demonstrate that Omnidirectional Video Interpolation can be effectively improved by modeling for omnidirectional distortion.
Authors: Yu Linghu, Tiago de Freitas Pereira, Christophe Ecabert, S\'ebastien Marcel, Manuel G\"unther
Abstract: Fair biometric algorithms have similar verification performance across different demographic groups given a single decision threshold. Unfortunately, for state-of-the-art face recognition networks, score distributions differ between demographics. Contrary to work that tries to align those distributions by extra training or fine-tuning, we solely focus on score post-processing methods. As proved, well-known sample-centered score normalization techniques, Z-norm and T-norm, do not improve fairness for high-security operating points. Thus, we extend the standard Z/T-norm to integrate demographic information in normalization. Additionally, we investigate several possibilities to incorporate cohort similarities for both genuine and impostor pairs per demographic to improve fairness across different operating points. We run experiments on two datasets with different demographics (gender and ethnicity) and show that our techniques generally improve the overall fairness of five state-of-the-art pre-trained face recognition networks, without downgrading verification performance. We also indicate that an equal contribution of False Match Rate (FMR) and False Non-Match Rate (FNMR) in fairness evaluation is required for the highest gains. Code and protocols are available.
Authors: Lisa Anita De Santi, J\"org Schl\"otterer, Meike Nauta, Vincenzo Positano, Christin Seifert
Abstract: Volumetric neuroimaging examinations like structural Magnetic Resonance Imaging (sMRI) are routinely applied to support the clinical diagnosis of dementia like Alzheimer's Disease (AD). Neuroradiologists examine 3D sMRI to detect and monitor abnormalities in brain morphology due to AD, like global and/or local brain atrophy and shape alteration of characteristic structures. There is a strong research interest in developing diagnostic systems based on Deep Learning (DL) models to analyse sMRI for AD. However, anatomical information extracted from an sMRI examination needs to be interpreted together with patient's age to distinguish AD patterns from the regular alteration due to a normal ageing process. In this context, part-prototype neural networks integrate the computational advantages of DL in an interpretable-by-design architecture and showed promising results in medical imaging applications. We present PIMPNet, the first interpretable multimodal model for 3D images and demographics applied to the binary classification of AD from 3D sMRI and patient's age. Despite age prototypes do not improve predictive performance compared to the single modality model, this lays the foundation for future work in the direction of the model's design and multimodal prototype training process
Authors: Jakub Gwizda{\l}a, Doruk Oner, Soumava Kumar Roy, Mian Akbar Shah, Ad Eberhard, Ivan Egorov, Philipp Kr\"usi, Grigory Yakushev, Pascal Fua
Abstract: Power lines are dangerous for low-flying aircrafts, especially in low-visibility conditions. Thus, a vision-based system able to analyze the aircraft's surroundings and to provide the pilots with a "second pair of eyes" can contribute to enhancing their safety. To this end, we have developed a deep learning approach to jointly detect power line cables and pylons from images captured at distances of several hundred meters by aircraft-mounted cameras. In doing so, we have combined a modern convolutional architecture with transfer learning and a loss function adapted to curvilinear structure delineation. We use a single network for both detection tasks and demonstrated its performance on two benchmarking datasets. We have integrated it within an onboard system and run it in flight, and have demonstrated with our experiments that it outperforms the prior distant cable detection method on both datasets, while also successfully detecting pylons, given their annotations are available for the data.
Authors: Minlong Peng, Zidi Xiong, Quang H. Nguyen, Mingming Sun, Khoa D. Doan, Ping Li
Abstract: By injecting a small number of poisoned samples into the training set, backdoor attacks aim to make the victim model produce designed outputs on any input injected with pre-designed backdoors. In order to achieve a high attack success rate using as few poisoned training samples as possible, most existing attack methods change the labels of the poisoned samples to the target class. This practice often results in severe over-fitting of the victim model over the backdoors, making the attack quite effective in output control but easier to be identified by human inspection or automatic defense algorithms. In this work, we proposed a label-smoothing strategy to overcome the over-fitting problem of these attack methods, obtaining a \textit{Label-Smoothed Backdoor Attack} (LSBA). In the LSBA, the label of the poisoned sample $\bm{x}$ will be changed to the target class with a probability of $p_n(\bm{x})$ instead of 100\%, and the value of $p_n(\bm{x})$ is specifically designed to make the prediction probability the target class be only slightly greater than those of the other classes. Empirical studies on several existing backdoor attacks show that our strategy can considerably improve the stealthiness of these attacks and, at the same time, achieve a high attack success rate. In addition, our strategy makes it able to manually control the prediction probability of the design output through manipulating the applied and activated number of LSBAs\footnote{Source code will be published at \url{https://github.com/v-mipeng/LabelSmoothedAttack.git}}.
Authors: Florinel-Alin Croitoru, Nicolae-Catalin Ristea, Radu Tudor Ionescu, Nicu Sebe
Abstract: Most curriculum learning methods require an approach to sort the data samples by difficulty, which is often cumbersome to perform. In this work, we propose a novel curriculum learning approach termed Learning Rate Curriculum (LeRaC), which leverages the use of a different learning rate for each layer of a neural network to create a data-agnostic curriculum during the initial training epochs. More specifically, LeRaC assigns higher learning rates to neural layers closer to the input, gradually decreasing the learning rates as the layers are placed farther away from the input. The learning rates increase at various paces during the first training iterations, until they all reach the same value. From this point on, the neural model is trained as usual. This creates a model-level curriculum learning strategy that does not require sorting the examples by difficulty and is compatible with any neural network, generating higher performance levels regardless of the architecture. We conduct comprehensive experiments on 12 data sets from the computer vision (CIFAR-10, CIFAR-100, Tiny ImageNet, ImageNet-200, Food-101, UTKFace, PASCAL VOC), language (BoolQ, QNLI, RTE) and audio (ESC-50, CREMA-D) domains, considering various convolutional (ResNet-18, Wide-ResNet-50, DenseNet-121, YOLOv5), recurrent (LSTM) and transformer (CvT, BERT, SepTr) architectures. We compare our approach with the conventional training regime, as well as with Curriculum by Smoothing (CBS), a state-of-the-art data-agnostic curriculum learning approach. Unlike CBS, our performance improvements over the standard training regime are consistent across all data sets and models. Furthermore, we significantly surpass CBS in terms of training time (there is no additional cost over the standard training regime for LeRaC). Our code is freely available at: https://github.com/CroitoruAlin/LeRaC.
Authors: Dapeng Feng, Yuhua Qi, Shipeng Zhong, Zhiqiang Chen, Qiming Chen, Hongbo Chen, Jin Wu, Jun Ma
Abstract: The burgeoning demand for collaborative robotic systems to execute complex tasks collectively has intensified the research community's focus on advancing simultaneous localization and mapping (SLAM) in a cooperative context. Despite this interest, the scalability and diversity of existing datasets for collaborative trajectories remain limited, especially in scenarios with constrained perspectives where the generalization capabilities of Collaborative SLAM (C-SLAM) are critical for the feasibility of multi-agent missions. Addressing this gap, we introduce S3E, an expansive multimodal dataset. Captured by a fleet of unmanned ground vehicles traversing four distinct collaborative trajectory paradigms, S3E encompasses 13 outdoor and 5 indoor sequences. These sequences feature meticulously synchronized and spatially calibrated data streams, including 360-degree LiDAR point cloud, high-resolution stereo imagery, high-frequency inertial measurement units (IMU), and Ultra-wideband (UWB) relative observations. Our dataset not only surpasses previous efforts in scale, scene diversity, and data intricacy but also provides a thorough analysis and benchmarks for both collaborative and individual SLAM methodologies. For access to the dataset and the latest information, please visit our repository at https://pengyu-team.github.io/S3E.
Authors: Rui Zhu, Di Tang, Siyuan Tang, XiaoFeng Wang, Haixu Tang
Abstract: In this paper, we present a simple yet surprisingly effective technique to induce "selective amnesia" on a backdoored model. Our approach, called SEAM, has been inspired by the problem of catastrophic forgetting (CF), a long standing issue in continual learning. Our idea is to retrain a given DNN model on randomly labeled clean data, to induce a CF on the model, leading to a sudden forget on both primary and backdoor tasks; then we recover the primary task by retraining the randomized model on correctly labeled clean data. We analyzed SEAM by modeling the unlearning process as continual learning and further approximating a DNN using Neural Tangent Kernel for measuring CF. Our analysis shows that our random-labeling approach actually maximizes the CF on an unknown backdoor in the absence of triggered inputs, and also preserves some feature extraction in the network to enable a fast revival of the primary task. We further evaluated SEAM on both image processing and Natural Language Processing tasks, under both data contamination and training manipulation attacks, over thousands of models either trained on popular image datasets or provided by the TrojAI competition. Our experiments show that SEAM vastly outperforms the state-of-the-art unlearning techniques, achieving a high Fidelity (measuring the gap between the accuracy of the primary task and that of the backdoor) within a few minutes (about 30 times faster than training a model from scratch using the MNIST dataset), with only a small amount of clean data (0.1% of training data for TrojAI models).
Authors: Yatong Bai, Brendon G. Anderson, Aerin Kim, Somayeh Sojoudi
Abstract: While prior research has proposed a plethora of methods that build neural classifiers robust against adversarial robustness, practitioners are still reluctant to adopt them due to their unacceptably severe clean accuracy penalties. This paper significantly alleviates this accuracy-robustness trade-off by mixing the output probabilities of a standard classifier and a robust classifier, where the standard network is optimized for clean accuracy and is not robust in general. We show that the robust base classifier's confidence difference for correct and incorrect examples is the key to this improvement. In addition to providing intuitions and empirical evidence, we theoretically certify the robustness of the mixed classifier under realistic assumptions. Furthermore, we adapt an adversarial input detector into a mixing network that adaptively adjusts the mixture of the two base models, further reducing the accuracy penalty of achieving robustness. The proposed flexible method, termed "adaptive smoothing", can work in conjunction with existing or even future methods that improve clean accuracy, robustness, or adversary detection. Our empirical evaluation considers strong attack methods, including AutoAttack and adaptive attack. On the CIFAR-100 dataset, our method achieves an 85.21% clean accuracy while maintaining a 38.72% $\ell_\infty$-AutoAttacked ($\epsilon = 8/255$) accuracy, becoming the second most robust method on the RobustBench CIFAR-100 benchmark as of submission, while improving the clean accuracy by ten percentage points compared with all listed models. The code that implements our method is available at https://github.com/Bai-YT/AdaptiveSmoothing.
Authors: Harini S I, Somesh Singh, Yaman K Singla, Aanisha Bhattacharyya, Veeky Baths, Changyou Chen, Rajiv Ratn Shah, Balaji Krishnamurthy
Abstract: Marketers spend billions of dollars on advertisements, but to what end? At purchase time, if customers cannot recognize the brand for which they saw an ad, the money spent on the ad is essentially wasted. Despite its importance in marketing, until now, there has been no large-scale study on the memorability of ads. All previous memorability studies have been conducted on short-term recall on specific content types like action videos. On the other hand, the advertising industry only cares about long-term memorability, and ads are almost always highly multimodal. Therefore, we release the first memorability dataset, LAMBDA, consisting of 1749 participants and 2205 ads covering 276 brands. Running statistical tests over different participant subpopulations and ad types, we find many interesting insights into what makes an ad memorable, e.g., fast-moving ads are more memorable than those with slower scenes; people who use ad-blockers remember a lower number of ads than those who don't. Next, we present a model, Henry, to predict the memorability of a content. Henry achieves state-of-the-art performance across all prominent literature memorability datasets. It shows strong generalization performance with better results in 0-shot on unseen datasets. Finally, with the intent of memorable ad generation, we present a scalable method to build a high-quality memorable ad generation model by leveraging automatically annotated data. Our approach, SEED (Self rEwarding mEmorability Modeling), starts with a language model trained on LAMBDA as seed data and progressively trains an LLM to generate more memorable ads. We show that the generated advertisements have 44% higher memorability scores than the original ads. We release this large-scale ad dataset, UltraLAMBDA, consisting of 5 million ads. Our code and datasets are available at https://behavior-in-the-wild.github.io/memorability.
Authors: Yufan Deng, Ruida Wang, Yuhao Zhang, Yu-Wing Tai, Chi-Keung Tang
Abstract: Video generation models have shown their superior ability to generate photo-realistic video. However, how to accurately control (or edit) the video remains a formidable challenge. The main issues are: 1) how to perform direct and accurate user control in editing; 2) how to execute editings like changing shape, expression, and layout without unsightly distortion and artifacts to the edited content; and 3) how to maintain spatio-temporal consistency of video after editing. To address the above issues, we propose DragVideo, a general drag-style video editing framework. Inspired by DragGAN, DragVideo addresses issues 1) and 2) by proposing the drag-style video latent optimization method which gives desired control by updating noisy video latent according to drag instructions through video-level drag objective function. We amend issue 3) by integrating the video diffusion model with sample-specific LoRA and Mutual Self-Attention in DragVideo to ensure the edited result is spatio-temporally consistent. We also present a series of testing examples for drag-style video editing and conduct extensive experiments across a wide array of challenging editing tasks, such as motion, skeleton editing, etc, underscoring DragVideo can edit video in an intuitive, faithful to the user's intention manner, with nearly unnoticeable distortion and artifacts, while maintaining spatio-temporal consistency. While traditional prompt-based video editing fails to do the former two and directly applying image drag editing fails in the last, DragVideo's versatility and generality are emphasized. Github link: https://github.com/RickySkywalker/DragVideo-Official.
Authors: Gaurav Verma, Minje Choi, Kartik Sharma, Jamelle Watson-Daniels, Sejoon Oh, Srijan Kumar
Abstract: Multimodal large language models (MLLMs) like LLaVA and GPT-4(V) enable general-purpose conversations about images with the language modality. As off-the-shelf MLLMs may have limited capabilities on images from domains like dermatology and agriculture, they must be fine-tuned to unlock domain-specific applications. The prevalent architecture of current open-source MLLMs comprises two major modules: an image-language (cross-modal) projection network and a large language model. It is desirable to understand the roles of these two modules in modeling domain-specific visual attributes to inform the design of future models and streamline the interpretability efforts on the current models. To this end, via experiments on 4 datasets and under 2 fine-tuning settings, we find that as the MLLM is fine-tuned, it indeed gains domain-specific visual capabilities, but the updates do not lead to the projection extracting relevant domain-specific visual attributes. Our results indicate that the domain-specific visual attributes are modeled by the LLM, even when only the projection is fine-tuned. Through this study, we offer a potential reinterpretation of the role of cross-modal projections in MLLM architectures. Project webpage: https://claws-lab.github.io/projection-in-MLLMs/
Authors: Raghav Kapoor, Yash Parag Butala, Melisa Russak, Jing Yu Koh, Kiran Kamble, Waseem Alshikh, Ruslan Salakhutdinov
Abstract: For decades, human-computer interaction has fundamentally been manual. Even today, almost all productive work done on the computer necessitates human input at every step. Autonomous virtual agents represent an exciting step in automating many of these menial tasks. Virtual agents would empower users with limited technical proficiency to harness the full possibilities of computer systems. They could also enable the efficient streamlining of numerous computer tasks, ranging from calendar management to complex travel bookings, with minimal human intervention. In this paper, we introduce OmniACT, the first-of-a-kind dataset and benchmark for assessing an agent's capability to generate executable programs to accomplish computer tasks. Our scope extends beyond traditional web automation, covering a diverse range of desktop applications. The dataset consists of fundamental tasks such as "Play the next song", as well as longer horizon tasks such as "Send an email to John Doe mentioning the time and place to meet". Specifically, given a pair of screen image and a visually-grounded natural language task, the goal is to generate a script capable of fully executing the task. We run several strong baseline language model agents on our benchmark. The strongest baseline, GPT-4, performs the best on our benchmark However, its performance level still reaches only 15% of the human proficiency in generating executable scripts capable of completing the task, demonstrating the challenge of our task for conventional web agents. Our benchmark provides a platform to measure and evaluate the progress of language model agents in automating computer tasks and motivates future work towards building multimodal models that bridge large language models and the visual grounding of computer screens.
Authors: Eda Yilmaz, Hacer Yalim Keles
Abstract: We introduce Adversarial Sparse Teacher (AST), a robust defense method against distillation-based model stealing attacks. Our approach trains a teacher model using adversarial examples to produce sparse logit responses and increase the entropy of the output distribution. Typically, a model generates a peak in its output corresponding to its prediction. By leveraging adversarial examples, AST modifies the teacher model's original response, embedding a few altered logits into the output while keeping the primary response slightly higher. Concurrently, all remaining logits are elevated to further increase the output distribution's entropy. All these complex manipulations are performed using an optimization function with our proposed Exponential Predictive Divergence (EPD) loss function. EPD allows us to maintain higher entropy levels compared to traditional KL divergence, effectively confusing attackers. Experiments on CIFAR-10 and CIFAR-100 datasets demonstrate that AST outperforms state-of-the-art methods, providing effective defense against model stealing while preserving high accuracy. The source codes will be made publicly available here soon.
Authors: Dongyu Yan, Guanyu Huang, Fengyu Quan, Haoyao Chen
Abstract: Panoramic observation using fisheye cameras is significant in virtual reality (VR) and robot perception. However, panoramic images synthesized by traditional methods lack depth information and can only provide three degrees-of-freedom (3DoF) rotation rendering in VR applications. To fully preserve and exploit the parallax information within the original fisheye cameras, we introduce MSI-NeRF, which combines deep learning omnidirectional depth estimation and novel view synthesis. We construct a multi-sphere image as a cost volume through feature extraction and warping of the input images. We further build an implicit radiance field using spatial points and interpolated 3D feature vectors as input, which can simultaneously realize omnidirectional depth estimation and 6DoF view synthesis. Leveraging the knowledge from depth estimation task, our method can learn scene appearance by source view supervision only. It does not require novel target views and can be trained conveniently on existing panorama depth estimation datasets. Our network has the generalization ability to reconstruct unknown scenes efficiently using only four images. Experimental results show that our method outperforms existing methods in both depth estimation and novel view synthesis tasks.
Authors: Billel Essaid, Hamza Kheddar, Noureddine Batel, Muhammad E. H. Chowdhury, Abderrahmane Lakas
Abstract: Automatic speech recognition (ASR) plays a pivotal role in our daily lives, offering utility not only for interacting with machines but also for facilitating communication for individuals with partial or profound hearing impairments. The process involves receiving the speech signal in analog form, followed by various signal processing algorithms to make it compatible with devices of limited capacities, such as cochlear implants (CIs). Unfortunately, these implants, equipped with a finite number of electrodes, often result in speech distortion during synthesis. Despite efforts by researchers to enhance received speech quality using various state-of-the-art (SOTA) signal processing techniques, challenges persist, especially in scenarios involving multiple sources of speech, environmental noise, and other adverse conditions. The advent of new artificial intelligence (AI) methods has ushered in cutting-edge strategies to address the limitations and difficulties associated with traditional signal processing techniques dedicated to CIs. This review aims to comprehensively cover advancements in CI-based ASR and speech enhancement, among other related aspects. The primary objective is to provide a thorough overview of metrics and datasets, exploring the capabilities of AI algorithms in this biomedical field, and summarizing and commenting on the best results obtained. Additionally, the review will delve into potential applications and suggest future directions to bridge existing research gaps in this domain.
Authors: Junchen Fu, Xuri Ge, Xin Xin, Alexandros Karatzoglou, Ioannis Arapakis, Jie Wang, Joemon M. Jose
Abstract: Multimodal foundation models are transformative in sequential recommender systems, leveraging powerful representation learning capabilities. While Parameter-efficient Fine-tuning (PEFT) is commonly used to adapt foundation models for recommendation tasks, most research prioritizes parameter efficiency, often overlooking critical factors like GPU memory efficiency and training speed. Addressing this gap, our paper introduces IISAN (Intra- and Inter-modal Side Adapted Network for Multimodal Representation), a simple plug-and-play architecture using a Decoupled PEFT structure and exploiting both intra- and inter-modal adaptation. IISAN matches the performance of full fine-tuning (FFT) and state-of-the-art PEFT. More importantly, it significantly reduces GPU memory usage - from 47GB to just 3GB for multimodal sequential recommendation tasks. Additionally, it accelerates training time per epoch from 443s to 22s compared to FFT. This is also a notable improvement over the Adapter and LoRA, which require 37-39 GB GPU memory and 350-380 seconds per epoch for training. Furthermore, we propose a new composite efficiency metric, TPME (Training-time, Parameter, and GPU Memory Efficiency) to alleviate the prevalent misconception that "parameter efficiency represents overall efficiency". TPME provides more comprehensive insights into practical efficiency comparisons between different methods. Besides, we give an accessible efficiency analysis of all PEFT and FFT approaches, which demonstrate the superiority of IISAN. We release our codes and other materials at https://github.com/GAIR-Lab/IISAN.
Authors: Xueyi Liu, Kangbo Lyu, Jieqiong Zhang, Tao Du, Li Yi
Abstract: We explore the dexterous manipulation transfer problem by designing simulators. The task wishes to transfer human manipulations to dexterous robot hand simulations and is inherently difficult due to its intricate, highly-constrained, and discontinuous dynamics and the need to control a dexterous hand with a DoF to accurately replicate human manipulations. Previous approaches that optimize in high-fidelity black-box simulators or a modified one with relaxed constraints only demonstrate limited capabilities or are restricted by insufficient simulation fidelity. We introduce parameterized quasi-physical simulators and a physics curriculum to overcome these limitations. The key ideas are 1) balancing between fidelity and optimizability of the simulation via a curriculum of parameterized simulators, and 2) solving the problem in each of the simulators from the curriculum, with properties ranging from high task optimizability to high fidelity. We successfully enable a dexterous hand to track complex and diverse manipulations in high-fidelity simulated environments, boosting the success rate by 11\%+ from the best-performed baseline. The project website is available at https://meowuu7.github.io/QuasiSim/.
Authors: Minyoung Huh, Brian Cheung, Tongzhou Wang, Phillip Isola
Abstract: We argue that representations in AI models, particularly deep networks, are converging. First, we survey many examples of convergence in the literature: over time and across multiple domains, the ways by which different neural networks represent data are becoming more aligned. Next, we demonstrate convergence across data modalities: as vision models and language models get larger, they measure distance between datapoints in a more and more alike way. We hypothesize that this convergence is driving toward a shared statistical model of reality, akin to Plato's concept of an ideal reality. We term such a representation the platonic representation and discuss several possible selective pressures toward it. Finally, we discuss the implications of these trends, their limitations, and counterexamples to our analysis.
Authors: Abdur Rahman, Rajat Chawla, Muskaan Kumar, Arkajit Datta, Adarsh Jha, Mukunda NS, Ishaan Bhola
Abstract: In the rapidly evolving landscape of AI research and application, Multimodal Large Language Models (MLLMs) have emerged as a transformative force, adept at interpreting and integrating information from diverse modalities such as text, images, and Graphical User Interfaces (GUIs). Despite these advancements, the nuanced interaction and understanding of GUIs pose a significant challenge, limiting the potential of existing models to enhance automation levels. To bridge this gap, this paper presents V-Zen, an innovative Multimodal Large Language Model (MLLM) meticulously crafted to revolutionise the domain of GUI understanding and grounding. Equipped with dual-resolution image encoders, V-Zen establishes new benchmarks in efficient grounding and next-action prediction, thereby laying the groundwork for self-operating computer systems. Complementing V-Zen is the GUIDE dataset, an extensive collection of real-world GUI elements and task-based sequences, serving as a catalyst for specialised fine-tuning. The successful integration of V-Zen and GUIDE marks the dawn of a new era in multimodal AI research, opening the door to intelligent, autonomous computing experiences. This paper extends an invitation to the research community to join this exciting journey, shaping the future of GUI automation. In the spirit of open science, our code, data, and model will be made publicly available, paving the way for multimodal dialogue scenarios with intricate and precise interactions.
Authors: Samar Agnihotri, Renu Rameshan, Ritwik Ghosal
Abstract: Lossless image compression is required in various applications to reduce storage or transmission costs of images, while requiring the reconstructed images to have zero information loss compared to the original. Existing lossless image compression methods either have simple design but poor compression performance, or complex design, better performance, but with no performance guarantees. In our endeavor to develop a lossless image compression method with low complexity and guaranteed performance, we argue that compressibility of a color image is essentially derived from the patterns in its spatial structure, intensity variations, and color variations. Thus, we divide the overall design of a lossless image compression scheme into three parts that exploit corresponding redundancies. We further argue that the binarized version of an image captures its fundamental spatial structure. In this first part of our work, we propose a scheme for lossless compression of binary images. The proposed scheme first learns dictionaries of $16\times16$, $8\times8$, $4\times4$, and $2\times 2$ square pixel patterns from various datasets of binary images. It then uses these dictionaries to encode binary images. These dictionaries have various interesting properties that are further exploited to construct an efficient and scalable scheme. Our preliminary results show that the proposed scheme consistently outperforms existing conventional and learning based lossless compression approaches, and provides, on average, as much as $1.5\times$ better performance than a common general purpose lossless compression scheme (WebP), more than $3\times$ better performance than a state of the art learning based scheme, and better performance than a specialized scheme for binary image compression (JBIG2).
Authors: Felix Wagner, Wentian Xu, Pramit Saha, Ziyun Liang, Daniel Whitehouse, David Menon, Virginia Newcombe, Natalie Voets, J. Alison Noble, Konstantinos Kamnitsas
Abstract: Segmentation models for brain lesions in MRI are commonly developed for a specific disease and trained on data with a predefined set of MRI modalities. Each such model cannot segment the disease using data with a different set of MRI modalities, nor can it segment any other type of disease. Moreover, this training paradigm does not allow a model to benefit from learning from heterogeneous databases that may contain scans and segmentation labels for different types of brain pathologies and diverse sets of MRI modalities. Additionally, the sensitivity of patient data often prevents centrally aggregating data, necessitating a decentralized approach. Is it feasible to use Federated Learning (FL) to train a single model on client databases that contain scans and labels of different brain pathologies and diverse sets of MRI modalities? We demonstrate promising results by combining appropriate, simple, and practical modifications to the model and training strategy: Designing a model with input channels that cover the whole set of modalities available across clients, training with random modality drop, and exploring the effects of feature normalization methods. Evaluation on 7 brain MRI databases with 5 different diseases shows that such FL framework can train a single model that is shown to be very promising in segmenting all disease types seen during training. Importantly, it is able to segment these diseases in new databases that contain sets of modalities different from those in training clients. These results demonstrate, for the first time, the feasibility and effectiveness of using Federated Learning to train a single 3D segmentation model on decentralised data with diverse brain diseases and MRI modalities, a necessary step towards leveraging heterogeneous real-world databases. Code will be made available at: https://github.com/FelixWag/FL-MultiDisease-MRI
Authors: Jonas Ngnaw\'e, Sabyasachi Sahoo, Yann Pequignot, Fr\'ed\'eric Precioso, Christian Gagn\'e
Abstract: Despite extensive research on adversarial training strategies to improve robustness, the decisions of even the most robust deep learning models can still be quite sensitive to imperceptible perturbations, creating serious risks when deploying them for high-stakes real-world applications. While detecting such cases may be critical, evaluating a model's vulnerability at a per-instance level using adversarial attacks is computationally too intensive and unsuitable for real-time deployment scenarios. The input space margin is the exact score to detect non-robust samples and is intractable for deep neural networks. This paper introduces the concept of margin consistency -- a property that links the input space margins and the logit margins in robust models -- for efficient detection of vulnerable samples. First, we establish that margin consistency is a necessary and sufficient condition to use a model's logit margin as a score for identifying non-robust samples. Next, through comprehensive empirical analysis of various robustly trained models on CIFAR10 and CIFAR100 datasets, we show that they indicate strong margin consistency with a strong correlation between their input space margins and the logit margins. Then, we show that we can effectively use the logit margin to confidently detect brittle decisions with such models and accurately estimate robust accuracy on an arbitrarily large test set by estimating the input margins only on a small subset. Finally, we address cases where the model is not sufficiently margin-consistent by learning a pseudo-margin from the feature representation. Our findings highlight the potential of leveraging deep representations to efficiently assess adversarial vulnerability in deployment scenarios.
Authors: Nick John Eliopoulos, Purvish Jajal, James Davis, Gaowen Liu, George K. Thiravathukal, Yung-Hsiang Lu
Abstract: This paper investigates how to efficiently deploy vision transformers on edge devices. Recent methods reduce the latency of transformer neural networks by removing or merging tokens, with small accuracy degradation. However, these methods are not designed with edge device deployment in mind: jthey do not leverage information about the latency vs. workload trends to improve efficiency. First, we show the latency-workload size relationship is nonlinear for certain workload sizes. We consider this relationship to create a token pruning schedule. Second, we demonstrate a training-free, token pruning method utilizing this schedule. We show that for single batch inference, other methods increase latency by 18.6-30.3% with respect to baseline, while we can reduce it by 9%. For similar latency (within 5.2%) across devices we achieve 78.6%-84.5% ImageNet1K accuracy, while the state-of-the-art, Token Merging, achieves 45.8%-85.4%.
Authors: Jian-Qing Zheng, Yuanhan Mo, Yang Sun, Jiahua Li, Fuping Wu, Ziyang Wang, Tonia Vincent, Bart{\l}omiej W. Papie\.z
Abstract: In medical imaging, the diffusion models have shown great potential in synthetic image generation tasks. However, these models often struggle with the interpretable connections between the generated and existing images and could create illusions. To address these challenges, our research proposes a novel diffusion-based generative model based on deformation diffusion and recovery. This model, named Deformation-Recovery Diffusion Model (DRDM), diverges from traditional score/intensity and latent feature-based approaches, emphasizing morphological changes through deformation fields rather than direct image synthesis. This is achieved by introducing a topological-preserving deformation field generation method, which randomly samples and integrates a set of multi-scale Deformation Vector Fields (DVF). DRDM is trained to learn to recover unreasonable deformation components, thereby restoring each randomly deformed image to a realistic distribution. These innovations facilitate the generation of diverse and anatomically plausible deformations, enhancing data augmentation and synthesis for further analysis in downstream tasks, such as few-shot learning and image registration. Experimental results in cardiac MRI and pulmonary CT show DRDM is capable of creating diverse, large (over 10\% image size deformation scale), and high-quality (negative rate of the Jacobian matrix's determinant is lower than 1\%) deformation fields. The further experimental results in downstream tasks, 2D image segmentation and 3D image registration, indicate significant improvements resulting from DRDM, showcasing the potential of our model to advance image manipulation and synthesis in medical imaging and beyond. Project page: https://jianqingzheng.github.io/def_diff_rec/
Authors: Zhuoyuan Li, Yao Li, Chuanbo Tang, Li Li, Dong Liu, Feng Wu
Abstract: Inter prediction is a key technology to reduce the temporal redundancy in video coding. In natural videos, there are usually multiple moving objects with variable velocity, resulting in complex motion fields that are difficult to represent compactly. In Versatile Video Coding (VVC), existing inter prediction methods usually assume uniform speed motion between consecutive frames and use the linear models for motion estimation (ME) and motion compensation (MC), which may not well handle the complex motion fields in the real world. To address these issues, we introduce a uniformly accelerated motion model (UAMM) to exploit motion-related elements (velocity, acceleration) of moving objects between the video frames, and further combine them to assist the inter prediction methods to handle the variable motion in the temporal domain. Specifically, first, the theory of UAMM is mentioned. Second, based on that, we propose the UAMM-based parameter derivation and extrapolation schemes in the coding process. Third, we integrate the UAMM into existing inter prediction modes (Merge, MMVD, CIIP) to achieve higher prediction accuracy. The proposed method is implemented into the VVC reference software, VTM version 12.0. Experimental results show that the proposed method achieves up to 0.38% and on average 0.13% BD-rate reduction compared to the VTM anchor, under the Low-delay P configuration, with a slight increase of time complexity on the encoding/decoding side.
Authors: Chao Huang, Dejan Markovic, Chenliang Xu, Alexander Richard
Abstract: While rendering and animation of photorealistic 3D human body models have matured and reached an impressive quality over the past years, modeling the spatial audio associated with such full body models has been largely ignored so far. In this work, we present a framework that allows for high-quality spatial audio generation, capable of rendering the full 3D soundfield generated by a human body, including speech, footsteps, hand-body interactions, and others. Given a basic audio-visual representation of the body in form of 3D body pose and audio from a head-mounted microphone, we demonstrate that we can render the full acoustic scene at any point in 3D space efficiently and accurately. To enable near-field and realtime rendering of sound, we borrow the idea of volumetric primitives from graphical neural rendering and transfer them into the acoustic domain. Our acoustic primitives result in an order of magnitude smaller soundfield representations and overcome deficiencies in near-field rendering compared to previous approaches.