Computer Vision and Pattern Recognition (cs.CV)

Abstract: 77% of adults over 50 want to age in place today, presenting a major challenge to ensuring adequate nutritional intake. It has been reported that one in four older adults that are 65 years or older are malnourished and given the direct link between malnutrition and decreased quality of life, there have been numerous studies conducted on how to efficiently track nutritional intake of food. Recent advancements in machine learning and computer vision show promise of automated nutrition tracking methods of food, but require a large high-quality dataset in order to accurately identify the nutrients from the food on the plate. Unlike existing datasets, a collection of 3D models with nutritional information allow for view synthesis to create an infinite number of 2D images for any given viewpoint/camera angle along with the associated nutritional information. In this paper, we develop a methodology for collecting high-quality 3D models for food items with a particular focus on speed and consistency, and introduce NutritionVerse-3D, a large-scale high-quality high-resolution dataset of 105 3D food models, in conjunction with their associated weight, food name, and nutritional value. These models allow for large quantity food intake scenes, diverse and customizable scene layout, and an infinite number of camera settings and lighting conditions. NutritionVerse-3D is publicly available as a part of an open initiative to accelerate machine learning for nutrition sensing.

Abstract: With the growth in capabilities of generative models, there has been growing interest in using photo-realistic renders of common 3D food items to improve downstream tasks such as food printing, nutrition prediction, or management of food wastage. Despite 3D modelling capabilities being more accessible than ever due to the success of NeRF based view-synthesis, such rendering methods still struggle to correctly capture thin food objects, often generating meshes with significant holes. In this study, we present an optimized strategy for enabling improved rendering of thin 3D food models, and demonstrate qualitative improvements in rendering quality. Our method generates the 3D model mesh via a proposed thin-object-optimized differentiable reconstruction method and tailors the strategy at both the data collection and training stages to better handle thin objects. While simple, we find that this technique can be employed for quick and highly consistent capturing of thin 3D objects.

Abstract: Movie story analysis requires understanding characters' emotions and mental states. Towards this goal, we formulate emotion understanding as predicting a diverse and multi-label set of emotions at the level of a movie scene and for each character. We propose EmoTx, a multimodal Transformer-based architecture that ingests videos, multiple characters, and dialog utterances to make joint predictions. By leveraging annotations from the MovieGraphs dataset, we aim to predict classic emotions (e.g. happy, angry) and other mental states (e.g. honest, helpful). We conduct experiments on the most frequently occurring 10 and 25 labels, and a mapping that clusters 181 labels to 26. Ablation studies and comparison against adapted state-of-the-art emotion recognition approaches shows the effectiveness of EmoTx. Analyzing EmoTx's self-attention scores reveals that expressive emotions often look at character tokens while other mental states rely on video and dialog cues.

Abstract: Face anti-spoofing (FAS) based on domain generalization (DG) has been recently studied to improve the generalization on unseen scenarios. Previous methods typically rely on domain labels to align the distribution of each domain for learning domain-invariant representations. However, artificial domain labels are coarse-grained and subjective, which cannot reflect real domain distributions accurately. Besides, such domain-aware methods focus on domain-level alignment, which is not fine-grained enough to ensure that learned representations are insensitive to domain styles. To address these issues, we propose a novel perspective for DG FAS that aligns features on the instance level without the need for domain labels. Specifically, Instance-Aware Domain Generalization framework is proposed to learn the generalizable feature by weakening the features' sensitivity to instance-specific styles. Concretely, we propose Asymmetric Instance Adaptive Whitening to adaptively eliminate the style-sensitive feature correlation, boosting the generalization. Moreover, Dynamic Kernel Generator and Categorical Style Assembly are proposed to first extract the instance-specific features and then generate the style-diversified features with large style shifts, respectively, further facilitating the learning of style-insensitive features. Extensive experiments and analysis demonstrate the superiority of our method over state-of-the-art competitors. Code will be publicly available at https://github.com/qianyuzqy/IADG.

Abstract: We introduce the task of 3D visual grounding in large-scale dynamic scenes based on natural linguistic descriptions and online captured multi-modal visual data, including 2D images and 3D LiDAR point clouds. We present a novel method, WildRefer, for this task by fully utilizing the appearance features in images, the location and geometry features in point clouds, and the dynamic features in consecutive input frames to match the semantic features in language. In particular, we propose two novel datasets, STRefer and LifeRefer, which focus on large-scale human-centric daily-life scenarios with abundant 3D object and natural language annotations. Our datasets are significant for the research of 3D visual grounding in the wild and has huge potential to boost the development of autonomous driving and service robots. Extensive comparisons and ablation studies illustrate that our method achieves state-of-the-art performance on two proposed datasets. Code and dataset will be released when the paper is published.

Abstract: Since the differences in viewing range, resolution and relative position, the multi-modality sensing module composed of infrared and visible cameras needs to be registered so as to have more accurate scene perception. In practice, manual calibration-based registration is the most widely used process, and it is regularly calibrated to maintain accuracy, which is time-consuming and labor-intensive. To cope with these problems, we propose a scene-adaptive infrared and visible image registration. Specifically, in regard of the discrepancy between multi-modality images, an invertible translation process is developed to establish a modality-invariant domain, which comprehensively embraces the feature intensity and distribution of both infrared and visible modalities. We employ homography to simulate the deformation between different planes and develop a hierarchical framework to rectify the deformation inferred from the proposed latent representation in a coarse-to-fine manner. For that, the advanced perception ability coupled with the residual estimation conducive to the regression of sparse offsets, and the alternate correlation search facilitates a more accurate correspondence matching. Moreover, we propose the first ground truth available misaligned infrared and visible image dataset, involving three synthetic sets and one real-world set. Extensive experiments validate the effectiveness of the proposed method against the state-of-the-arts, advancing the subsequent applications.

Abstract: Contrastive Language-Image Pre-training (CLIP) is a powerful multimodal large vision model that has demonstrated significant benefits for downstream tasks, including many zero-shot learning and text-guided vision tasks. However, we notice some severe problems regarding the model's explainability, which undermines its credibility and impedes related tasks. Specifically, we find CLIP prefers the background regions than the foregrounds according to the predicted similarity map, which contradicts human understanding. Besides, there are obvious noisy activations on the visualization results at irrelevant positions. To address these two issues, we conduct in-depth analyses and reveal the reasons with new findings and evidences. Based on these insights, we propose the CLIP Surgery, a method that enables surgery-like modifications for the inference architecture and features, for better explainability and enhancement in multiple open-vocabulary tasks. The proposed method has significantly improved the explainability of CLIP for both convolutional networks and vision transformers, surpassing existing methods by large margins. Besides, our approach also demonstrates remarkable improvements in open-vocabulary segmentation and multi-label recognition tasks. For examples, the mAP improvement on NUS-Wide multi-label recognition is 4.41% without any additional training, and our CLIP Surgery surpasses the state-of-the-art method by 8.74% at mIoU on Cityscapes open-vocabulary semantic segmentation. Furthermore, our method benefits other tasks including multimodal visualization and interactive segmentation like Segment Anything Model (SAM). The code is available at https://github.com/xmed-lab/CLIP_Surgery

Abstract: This paper studies how to keep a vision backbone effective while removing token mixers in its basic building blocks. Token mixers, as self-attention for vision transformers (ViTs), are intended to perform information communication between different spatial tokens but suffer from considerable computational cost and latency. However, directly removing them will lead to an incomplete model structure prior, and thus brings a significant accuracy drop. To this end, we first develop an RepIdentityFormer base on the re-parameterizing idea, to study the token mixer free model architecture. And we then explore the improved learning paradigm to break the limitation of simple token mixer free backbone, and summarize the empirical practice into 5 guidelines. Equipped with the proposed optimization strategy, we are able to build an extremely simple vision backbone with encouraging performance, while enjoying the high efficiency during inference. Extensive experiments and ablative analysis also demonstrate that the inductive bias of network architecture, can be incorporated into simple network structure with appropriate optimization strategy. We hope this work can serve as a starting point for the exploration of optimization-driven efficient network design. Project page: https://techmonsterwang.github.io/RIFormer/.

Abstract: Most urban applications necessitate building footprints in the form of concise vector graphics with sharp boundaries rather than pixel-wise raster images. This need contrasts with the majority of existing methods, which typically generate over-smoothed footprint polygons. Editing these automatically produced polygons can be inefficient, if not more time-consuming than manual digitization. This paper introduces a semi-automatic approach for building footprint extraction through semantically-sensitive superpixels and neural graph networks. Drawing inspiration from object-based classification techniques, we first learn to generate superpixels that are not only boundary-preserving but also semantically-sensitive. The superpixels respond exclusively to building boundaries rather than other natural objects, while simultaneously producing semantic segmentation of the buildings. These intermediate superpixel representations can be naturally considered as nodes within a graph. Consequently, graph neural networks are employed to model the global interactions among all superpixels and enhance the representativeness of node features for building segmentation. Classical approaches are utilized to extract and regularize boundaries for the vectorized building footprints. Utilizing minimal clicks and straightforward strokes, we efficiently accomplish accurate segmentation outcomes, eliminating the necessity for editing polygon vertices. Our proposed approach demonstrates superior precision and efficacy, as validated by experimental assessments on various public benchmark datasets. We observe a 10\% enhancement in the metric for superpixel clustering and an 8\% increment in vector graphics evaluation, when compared with established techniques. Additionally, we have devised an optimized and sophisticated pipeline for interactive editing, poised to further augment the overall quality of the results.

Abstract: Rail detection, essential for railroad anomaly detection, aims to identify the railroad region in video frames. Although various studies on rail detection exist, neither an open benchmark nor a high-speed network is available in the community, making algorithm comparison and development difficult. Inspired by the growth of lane detection, we propose a rail database and a row-based rail detection method. In detail, we make several contributions: (i) We present a real-world railway dataset, Rail-DB, with 7432 pairs of images and annotations. The images are collected from different situations in lighting, road structures, and views. The rails are labeled with polylines, and the images are categorized into nine scenes. The Rail-DB is expected to facilitate the improvement of rail detection algorithms. (ii) We present an efficient row-based rail detection method, Rail-Net, containing a lightweight convolutional backbone and an anchor classifier. Specifically, we formulate the process of rail detection as a row-based selecting problem. This strategy reduces the computational cost compared to alternative segmentation methods. (iii) We evaluate the Rail-Net on Rail-DB with extensive experiments, including cross-scene settings and network backbones ranging from ResNet to Vision Transformers. Our method achieves promising performance in terms of both speed and accuracy. Notably, a lightweight version could achieve 92.77% accuracy and 312 frames per second. The Rail-Net outperforms the traditional method by 50.65% and the segmentation one by 5.86%. The database and code are available at: https://github.com/Sampson-Lee/Rail-Detection.

Abstract: Inverse path tracing has recently been applied to joint material and lighting estimation, given geometry and multi-view HDR observations of an indoor scene. However, it has two major limitations: path tracing is expensive to compute, and ambiguities exist between reflection and emission. We propose a novel Factorized Inverse Path Tracing (FIPT) method which utilizes a factored light transport formulation and finds emitters driven by rendering errors. Our algorithm enables accurate material and lighting optimization faster than previous work, and is more effective at resolving ambiguities. The exhaustive experiments on synthetic scenes show that our method (1) outperforms state-of-the-art indoor inverse rendering and relighting methods particularly in the presence of complex illumination effects; (2) speeds up inverse path tracing optimization to less than an hour. We further demonstrate robustness to noisy inputs through material and lighting estimates that allow plausible relighting in a real scene. The source code is available at: https://github.com/lwwu2/fipt

Abstract: We present a deep learning method for accurately localizing the center of a single corneal reflection (CR) in an eye image. Unlike previous approaches, we use a convolutional neural network (CNN) that was trained solely using simulated data. Using only simulated data has the benefit of completely sidestepping the time-consuming process of manual annotation that is required for supervised training on real eye images. To systematically evaluate the accuracy of our method, we first tested it on images with simulated CRs placed on different backgrounds and embedded in varying levels of noise. Second, we tested the method on high-quality videos captured from real eyes. Our method outperformed state-of-the-art algorithmic methods on real eye images with a 35% reduction in terms of spatial precision, and performed on par with state-of-the-art on simulated images in terms of spatial accuracy.We conclude that our method provides a precise method for CR center localization and provides a solution to the data availability problem which is one of the important common roadblocks in the development of deep learning models for gaze estimation. Due to the superior CR center localization and ease of application, our method has the potential to improve the accuracy and precision of CR-based eye trackers

Abstract: Deep neural networks (DNNs) have shown exciting performance in various tasks, yet suffer generalization failures when meeting unknown target domains. One of the most promising approaches to achieve domain generalization (DG) is generating unseen data, e.g., mixup, to cover the unknown target data. However, existing works overlook the challenges induced by the simultaneous appearance of changes in both the semantic and distribution space. Accordingly, such a challenge makes source distributions hard to fit for DNNs. To mitigate the hard-fitting issue, we propose to perform a semantic-aware mixup (SAM) for domain generalization, where whether to perform mixup depends on the semantic and domain information. The feasibility of SAM shares the same spirits with the Fourier-based mixup. Namely, the Fourier phase spectrum is expected to contain semantics information (relating to labels), while the Fourier amplitude retains other information (relating to style information). Built upon the insight, SAM applies different mixup strategies to the Fourier phase spectrum and amplitude information. For instance, SAM merely performs mixup on the amplitude spectrum when both the semantic and domain information changes. Consequently, the overwhelmingly large change can be avoided. We validate the effectiveness of SAM using image classification tasks on several DG benchmarks.

Abstract: Social group detection is a crucial aspect of various robotic applications, including robot navigation and human-robot interactions. To date, a range of model-based techniques have been employed to address this challenge, such as the F-formation and trajectory similarity frameworks. However, these approaches often fail to provide reliable results in crowded and dynamic scenarios. Recent advancements in this area have mainly focused on learning-based methods, such as deep neural networks that use visual content or human pose. Although visual content-based methods have demonstrated promising performance on large-scale datasets, their computational complexity poses a significant barrier to their practical use in real-time applications. To address these issues, we propose a simple and efficient framework for social group detection. Our approach explores the impact of motion trajectory on social grouping and utilizes a novel, reliable, and fast data-driven method. We formulate the individuals in a scene as a graph, where the nodes are represented by LSTM-encoded trajectories and the edges are defined by the distances between each pair of tracks. Our framework employs a modified graph transformer module and graph clustering losses to detect social groups. Our experiments on the popular JRDBAct dataset reveal noticeable improvements in performance, with relative improvements ranging from 2% to 11%. Furthermore, our framework is significantly faster, with up to 12x faster inference times compared to state-of-the-art methods under the same computation resources. These results demonstrate that our proposed method is suitable for real-time robotic applications.

Abstract: We have recently seen tremendous progress in diffusion advances for generating realistic human motions. Yet, they largely disregard the rich multi-human interactions. In this paper, we present InterGen, an effective diffusion-based approach that incorporates human-to-human interactions into the motion diffusion process, which enables layman users to customize high-quality two-person interaction motions, with only text guidance. We first contribute a multimodal dataset, named InterHuman. It consists of about 107M frames for diverse two-person interactions, with accurate skeletal motions and 16,756 natural language descriptions. For the algorithm side, we carefully tailor the motion diffusion model to our two-person interaction setting. To handle the symmetry of human identities during interactions, we propose two cooperative transformer-based denoisers that explicitly share weights, with a mutual attention mechanism to further connect the two denoising processes. Then, we propose a novel representation for motion input in our interaction diffusion model, which explicitly formulates the global relations between the two performers in the world frame. We further introduce two novel regularization terms to encode spatial relations, equipped with a corresponding damping scheme during the training of our interaction diffusion model. Extensive experiments validate the effectiveness and generalizability of InterGen. Notably, it can generate more diverse and compelling two-person motions than previous methods and enables various downstream applications for human interactions.

Abstract: Recovering whole-body mesh by inferring the abstract pose and shape parameters from visual content can obtain 3D bodies with realistic structures. However, the inferring process is highly non-linear and suffers from image-mesh misalignment, resulting in inaccurate reconstruction. In contrast, 3D keypoint estimation methods utilize the volumetric representation to achieve pixel-level accuracy but may predict unrealistic body structures. To address these issues, this paper presents a novel hybrid inverse kinematics solution, HybrIK, that integrates the merits of 3D keypoint estimation and body mesh recovery in a unified framework. HybrIK directly transforms accurate 3D joints to body-part rotations via twist-and-swing decomposition. The swing rotations are analytically solved with 3D joints, while the twist rotations are derived from visual cues through neural networks. To capture comprehensive whole-body details, we further develop a holistic framework, HybrIK-X, which enhances HybrIK with articulated hands and an expressive face. HybrIK-X is fast and accurate by solving the whole-body pose with a one-stage model. Experiments demonstrate that HybrIK and HybrIK-X preserve both the accuracy of 3D joints and the realistic structure of the parametric human model, leading to pixel-aligned whole-body mesh recovery. The proposed method significantly surpasses the state-of-the-art methods on various benchmarks for body-only, hand-only, and whole-body scenarios. Code and results can be found at https://jeffli.site/HybrIK-X/

Abstract: Self-attention modules have demonstrated remarkable capabilities in capturing long-range relationships and improving the performance of point cloud tasks. However, point cloud objects are typically characterized by complex, disordered, and non-Euclidean spatial structures with multiple scales, and their behavior is often dynamic and unpredictable. The current self-attention modules mostly rely on dot product multiplication and dimension alignment among query-key-value features, which cannot adequately capture the multi-scale non-Euclidean structures of point cloud objects. To address these problems, this paper proposes a self-attention plug-in module with its variants, Multi-scale Geometry-aware Transformer (MGT). MGT processes point cloud data with multi-scale local and global geometric information in the following three aspects. At first, the MGT divides point cloud data into patches with multiple scales. Secondly, a local feature extractor based on sphere mapping is proposed to explore the geometry inner each patch and generate a fixed-length representation for each patch. Thirdly, the fixed-length representations are fed into a novel geodesic-based self-attention to capture the global non-Euclidean geometry between patches. Finally, all the modules are integrated into the framework of MGT with an end-to-end training scheme. Experimental results demonstrate that the MGT vastly increases the capability of capturing multi-scale geometry using the self-attention mechanism and achieves strong competitive performance on mainstream point cloud benchmarks.

Abstract: Pseudo depth maps are depth map predicitions which are used as ground truth during training. In this paper we leverage pseudo depth maps in order to segment objects of classes that have never been seen during training. This renders our object segmentation task an open world task. The pseudo depth maps are generated using pretrained networks, which have either been trained with the full intention to generalize to downstream tasks (LeRes and MiDaS), or which have been trained in an unsupervised fashion on video sequences (MonodepthV2). In order to tell our network which object to segment, we provide the network with a single click on the object's surface on the pseudo depth map of the image as input. We test our approach on two different scenarios: One without the RGB image and one where the RGB image is part of the input. Our results demonstrate a considerably better generalization performance from seen to unseen object types when depth is used. On the Semantic Boundaries Dataset we achieve an improvement from $61.57$ to $69.79$ IoU score on unseen classes, when only using half of the training classes during training and performing the segmentation on depth maps only.

Abstract: The retrieval of 3D objects has gained significant importance in recent years due to its broad range of applications in computer vision, computer graphics, virtual reality, and augmented reality. However, the retrieval of 3D objects presents significant challenges due to the intricate nature of 3D models, which can vary in shape, size, and texture, and have numerous polygons and vertices. To this end, we introduce a novel SHREC challenge track that focuses on retrieving relevant 3D animal models from a dataset using sketch queries and expedites accessing 3D models through available sketches. Furthermore, a new dataset named ANIMAR was constructed in this study, comprising a collection of 711 unique 3D animal models and 140 corresponding sketch queries. Our contest requires participants to retrieve 3D models based on complex and detailed sketches. We receive satisfactory results from eight teams and 204 runs. Although further improvement is necessary, the proposed task has the potential to incentivize additional research in the domain of 3D object retrieval, potentially yielding benefits for a wide range of applications. We also provide insights into potential areas of future research, such as improving techniques for feature extraction and matching, and creating more diverse datasets to evaluate retrieval performance.

Abstract: The ability to incrementally learn new classes from limited samples is crucial to the development of artificial intelligence systems for real clinical application. Although existing incremental learning techniques have attempted to address this issue, they still struggle with only few labeled data, particularly when the samples are from varied domains. In this paper, we explore the cross-domain few-shot incremental learning (CDFSCIL) problem. CDFSCIL requires models to learn new classes from very few labeled samples incrementally, and the new classes may be vastly different from the target space. To counteract this difficulty, we propose a cross-domain enhancement constraint and cross-domain data augmentation method. Experiments on MedMNIST show that the classification performance of this method is better than other similar incremental learning methods.

Abstract: The human visual system processes images with varied degrees of resolution, with the fovea, a small portion of the retina, capturing the highest acuity region, which gradually declines toward the field of view's periphery. However, the majority of existing object localization methods rely on images acquired by image sensors with space-invariant resolution, ignoring biological attention mechanisms. As a region of interest pooling, this study employs a fixation prediction model that emulates human objective-guided attention of searching for a given class in an image. The foveated pictures at each fixation point are then classified to determine whether the target is present or absent in the scene. Throughout this two-stage pipeline method, we investigate the varying results obtained by utilizing high-level or panoptic features and provide a ground-truth label function for fixation sequences that is smoother, considering in a better way the spatial structure of the problem. Finally, we present a novel dual task model capable of performing fixation prediction and detection simultaneously, allowing knowledge transfer between the two tasks. We conclude that, due to the complementary nature of both tasks, the training process benefited from the sharing of knowledge, resulting in an improvement in performance when compared to the previous approach's baseline scores.

Abstract: Recently, Meta AI Research approaches a general, promptable Segment Anything Model (SAM) pre-trained on an unprecedentedly large segmentation dataset (SA-1B). Without a double, the emergence of SAM will yield significant benefits for a wide array of practical image segmentation applications. In this study, we conduct a series of intriguing investigations into the performance of SAM across various applications, particularly in the fields of natural images, agriculture, manufacturing, remote sensing, and healthcare. We analyze and discuss the benefits and limitations of SAM and provide an outlook on future development of segmentation tasks. Note that our work does not intend to propose new algorithms or theories, but rather provide a comprehensive view of SAM in practice. This work is expected to provide insights that facilitate future research activities toward generic segmentation.

Abstract: Face anti-spoofing (FAS) is an essential mechanism for safeguarding the integrity of automated face recognition systems. Despite substantial advancements, the generalization of existing approaches to real-world applications remains challenging. This limitation can be attributed to the scarcity and lack of diversity in publicly available FAS datasets, which often leads to overfitting during training or saturation during testing. In terms of quantity, the number of spoof subjects is a critical determinant. Most datasets comprise fewer than 2,000 subjects. With regard to diversity, the majority of datasets consist of spoof samples collected in controlled environments using repetitive, mechanical processes. This data collection methodology results in homogenized samples and a dearth of scenario diversity. To address these shortcomings, we introduce the Wild Face Anti-Spoofing (WFAS) dataset, a large-scale, diverse FAS dataset collected in unconstrained settings. Our dataset encompasses 853,729 images of 321,751 spoof subjects and 529,571 images of 148,169 live subjects, representing a substantial increase in quantity. Moreover, our dataset incorporates spoof data obtained from the internet, spanning a wide array of scenarios and various commercial sensors, including 17 presentation attacks (PAs) that encompass both 2D and 3D forms. This novel data collection strategy markedly enhances FAS data diversity. Leveraging the WFAS dataset and Protocol 1 (Known-Type), we host the Wild Face Anti-Spoofing Challenge at the CVPR2023 workshop. Additionally, we meticulously evaluate representative methods using Protocol 1 and Protocol 2 (Unknown-Type). Through an in-depth examination of the challenge outcomes and benchmark baselines, we provide insightful analyses and propose potential avenues for future research. The dataset is released under Insightface.

Abstract: Representation learning aims to discover individual salient features of a domain in a compact and descriptive form that strongly identifies the unique characteristics of a given sample respective to its domain. Existing works in visual style representation literature have tried to disentangle style from content during training explicitly. A complete separation between these has yet to be fully achieved. Our paper aims to learn a representation of visual artistic style more strongly disentangled from the semantic content depicted in an image. We use Neural Style Transfer (NST) to measure and drive the learning signal and achieve state-of-the-art representation learning on explicitly disentangled metrics. We show that strongly addressing the disentanglement of style and content leads to large gains in style-specific metrics, encoding far less semantic information and achieving state-of-the-art accuracy in downstream multimodal applications.

Abstract: The task of collaborative human pose forecasting stands for predicting the future poses of multiple interacting people, given those in previous frames. Predicting two people in interaction, instead of each separately, promises better performance, due to their body-body motion correlations. But the task has remained so far primarily unexplored. In this paper, we review the progress in human pose forecasting and provide an in-depth assessment of the single-person practices that perform best for 2-body collaborative motion forecasting. Our study confirms the positive impact of frequency input representations, space-time separable and fully-learnable interaction adjacencies for the encoding GCN and FC decoding. Other single-person practices do not transfer to 2-body, so the proposed best ones do not include hierarchical body modeling or attention-based interaction encoding. We further contribute a novel initialization procedure for the 2-body spatial interaction parameters of the encoder, which benefits performance and stability. Altogether, our proposed 2-body pose forecasting best practices yield a performance improvement of 21.9% over the state-of-the-art on the most recent ExPI dataset, whereby the novel initialization accounts for 3.5%. See our project page at https://www.pinlab.org/bestpractices2body

Abstract: Year after year, the demand for ever-better smartphone photos continues to grow, in particular in the domain of portrait photography. Manufacturers thus use perceptual quality criteria throughout the development of smartphone cameras. This costly procedure can be partially replaced by automated learning-based methods for image quality assessment (IQA). Due to its subjective nature, it is necessary to estimate and guarantee the consistency of the IQA process, a characteristic lacking in the mean opinion scores (MOS) widely used for crowdsourcing IQA. In addition, existing blind IQA (BIQA) datasets pay little attention to the difficulty of cross-content assessment, which may degrade the quality of annotations. This paper introduces PIQ23, a portrait-specific IQA dataset of 5116 images of 50 predefined scenarios acquired by 100 smartphones, covering a high variety of brands, models, and use cases. The dataset includes individuals of various genders and ethnicities who have given explicit and informed consent for their photographs to be used in public research. It is annotated by pairwise comparisons (PWC) collected from over 30 image quality experts for three image attributes: face detail preservation, face target exposure, and overall image quality. An in-depth statistical analysis of these annotations allows us to evaluate their consistency over PIQ23. Finally, we show through an extensive comparison with existing baselines that semantic information (image context) can be used to improve IQA predictions. The dataset along with the proposed statistical analysis and BIQA algorithms are available: https://github.com/DXOMARK-Research/PIQ2023

Abstract: Recent works on personalized text-to-image generation usually learn to bind a special token with specific subjects or styles of a few given images by tuning its embedding through gradient descent. It is natural to question whether we can optimize the textual inversions by only accessing the process of model inference. As only requiring the forward computation to determine the textual inversion retains the benefits of less GPU memory, simple deployment, and secure access for scalable models. In this paper, we introduce a \emph{gradient-free} framework to optimize the continuous textual inversion in an iterative evolutionary strategy. Specifically, we first initialize an appropriate token embedding for textual inversion with the consideration of visual and text vocabulary information. Then, we decompose the optimization of evolutionary strategy into dimension reduction of searching space and non-convex gradient-free optimization in subspace, which significantly accelerates the optimization process with negligible performance loss. Experiments in several applications demonstrate that the performance of text-to-image model equipped with our proposed gradient-free method is comparable to that of gradient-based counterparts with variant GPU/CPU platforms, flexible employment, as well as computational efficiency.

Abstract: Power lines pose a significant safety threat to unmanned aerial vehicles (UAVs) operating at low altitudes. However, detecting power lines in aerial images is challenging due to the small size of the foreground data (i.e., power lines) and the abundance of background information. To address this challenge, we propose DUFormer, a semantic segmentation algorithm designed specifically for power line detection in aerial images. We assume that performing sufficient feature extraction with a convolutional neural network (CNN) that has a strong inductive bias is beneficial for training an efficient Transformer model. To this end, we propose a heavy token encoder responsible for overlapping feature re-mining and tokenization. The encoder comprises a pyramid CNN feature extraction module and a power line feature enhancement module. Following sufficient feature extraction for power lines, the feature fusion is carried out, and then the Transformer block is used for global modeling. The final segmentation result is obtained by fusing local and global features in the decode head. Additionally, we demonstrate the significance of the joint multi-weight loss function in power line segmentation. The experimental results demonstrate that our proposed method achieves the state-of-the-art performance in power line segmentation on the publicly available TTPLA dataset.

Abstract: Semantic segmentation assigns category labels to each pixel in an image, enabling breakthroughs in fields such as autonomous driving and robotics. Deep Neural Networks have achieved high accuracies in semantic segmentation but require large training datasets. Some domains have difficulties building such datasets due to rarity, privacy concerns, and the need for skilled annotators. Few-Shot Learning (FSL) has emerged as a new research stream that allows models to learn new tasks from a few samples. This contribution provides an overview of FSL in semantic segmentation (FSS), proposes a new taxonomy, and describes current limitations and outlooks.

Abstract: This paper investigates the performance of diffusion models for video anomaly detection (VAD) within the most challenging but also the most operational scenario in which the data annotations are not used. As being sparse, diverse, contextual, and often ambiguous, detecting abnormal events precisely is a very ambitious task. To this end, we rely only on the information-rich spatio-temporal data, and the reconstruction power of the diffusion models such that a high reconstruction error is utilized to decide the abnormality. Experiments performed on two large-scale video anomaly detection datasets demonstrate the consistent improvement of the proposed method over the state-of-the-art generative models while in some cases our method achieves better scores than the more complex models. This is the first study using a diffusion model and examining its parameters' influence to present guidance for VAD in surveillance scenarios.

Abstract: It is desirable to predict the behavior of traffic participants conditioned on different planned trajectories of the autonomous vehicle. This allows the downstream planner to estimate the impact of its decisions. Recent approaches for conditional behavior prediction rely on a regression decoder, meaning that coordinates or polynomial coefficients are regressed. In this work we revisit set-based trajectory prediction, where the probability of each trajectory in a predefined trajectory set is determined by a classification model, and first-time employ it to the task of conditional behavior prediction. We propose RESET, which combines a new metric-driven algorithm for trajectory set generation with a graph-based encoder. For unconditional prediction, RESET achieves comparable performance to a regression-based approach. Due to the nature of set-based approaches, it has the advantageous property of being able to predict a flexible number of trajectories without influencing runtime or complexity. For conditional prediction, RESET achieves reasonable results with late fusion of the planned trajectory, which was not observed for regression-based approaches before. This means that RESET is computationally lightweight to combine with a planner that proposes multiple future plans of the autonomous vehicle, as large parts of the forward pass can be reused.

Abstract: The ability of convolutional neural networks (CNNs) to recognize objects regardless of their position in the image is due to the translation-equivariance of the convolutional operation. Group-equivariant CNNs transfer this equivariance to other transformations of the input. Dealing appropriately with objects and object parts of different scale is challenging, and scale can vary for multiple reasons such as the underlying object size or the resolution of the imaging modality. In this paper, we propose a scale-equivariant convolutional network layer for three-dimensional data that guarantees scale-equivariance in 3D CNNs. Scale-equivariance lifts the burden of having to learn each possible scale separately, allowing the neural network to focus on higher-level learning goals, which leads to better results and better data-efficiency. We provide an overview of the theoretical foundations and scientific work on scale-equivariant neural networks in the two-dimensional domain. We then transfer the concepts from 2D to the three-dimensional space and create a scale-equivariant convolutional layer for 3D data. Using the proposed scale-equivariant layer, we create a scale-equivariant U-Net for medical image segmentation and compare it with a non-scale-equivariant baseline method. Our experiments demonstrate the effectiveness of the proposed method in achieving scale-equivariance for 3D medical image analysis. We publish our code at https://github.com/wimmerth/scale-equivariant-3d-convnet for further research and application.

Abstract: StyleGANs are at the forefront of controllable image generation as they produce a latent space that is semantically disentangled, making it suitable for image editing and manipulation. However, the performance of StyleGANs severely degrades when trained via class-conditioning on large-scale long-tailed datasets. We find that one reason for degradation is the collapse of latents for each class in the $\mathcal{W}$ latent space. With NoisyTwins, we first introduce an effective and inexpensive augmentation strategy for class embeddings, which then decorrelates the latents based on self-supervision in the $\mathcal{W}$ space. This decorrelation mitigates collapse, ensuring that our method preserves intra-class diversity with class-consistency in image generation. We show the effectiveness of our approach on large-scale real-world long-tailed datasets of ImageNet-LT and iNaturalist 2019, where our method outperforms other methods by $\sim 19\%$ on FID, establishing a new state-of-the-art.

Abstract: Remarkable advances have been achieved recently in learning neural representations that characterize object geometry, while generating textured objects suitable for downstream applications and 3D rendering remains at an early stage. In particular, reconstructing textured geometry from images of real objects is a significant challenge -- reconstructed geometry is often inexact, making realistic texturing a significant challenge. We present Mesh2Tex, which learns a realistic object texture manifold from uncorrelated collections of 3D object geometry and photorealistic RGB images, by leveraging a hybrid mesh-neural-field texture representation. Our texture representation enables compact encoding of high-resolution textures as a neural field in the barycentric coordinate system of the mesh faces. The learned texture manifold enables effective navigation to generate an object texture for a given 3D object geometry that matches to an input RGB image, which maintains robustness even under challenging real-world scenarios where the mesh geometry approximates an inexact match to the underlying geometry in the RGB image. Mesh2Tex can effectively generate realistic object textures for an object mesh to match real images observations towards digitization of real environments, significantly improving over previous state of the art.

Abstract: The development of approaches for trajectory prediction requires metrics to validate and compare their performance. Currently established metrics are based on Euclidean distance, which means that errors are weighted equally in all directions. Euclidean metrics are insufficient for structured environments like roads, since they do not properly capture the agent's intent relative to the underlying lane. In order to provide a reasonable assessment of trajectory prediction approaches with regard to the downstream planning task, we propose a new metric that is lane distance-based: Lane Miss Rate (LMR). For the calculation of LMR, the ground-truth and predicted endpoints are assigned to lane segments, more precisely their centerlines. Measured by the distance along the lane segments, predictions that are within a certain threshold distance to the ground-truth count as hits, otherwise they count as misses. LMR is then defined as the ratio of sequences that yield a miss. Our results on three state-of-the-art trajectory prediction models show that LMR preserves the order of Euclidean distance-based metrics. In contrast to the Euclidean Miss Rate, qualitative results show that LMR yields misses for sequences where predictions are located on wrong lanes. Hits on the other hand result for sequences where predictions are located on the correct lane. This means that LMR implicitly weights Euclidean error relative to the lane and goes into the direction of capturing intents of traffic agents. The source code of LMR for Argoverse 1 is publicly available.

Abstract: Modern image retrieval methods typically rely on fine-tuning pre-trained encoders to extract image-level descriptors. However, the most widely used models are pre-trained on ImageNet-1K with limited classes. The pre-trained feature representation is therefore not universal enough to generalize well to the diverse open-world classes. In this paper, we first cluster the large-scale LAION400M into one million pseudo classes based on the joint textual and visual features extracted by the CLIP model. Due to the confusion of label granularity, the automatically clustered dataset inevitably contains heavy inter-class conflict. To alleviate such conflict, we randomly select partial inter-class prototypes to construct the margin-based softmax loss. To further enhance the low-dimensional feature representation, we randomly select partial feature dimensions when calculating the similarities between embeddings and class-wise prototypes. The dual random partial selections are with respect to the class dimension and the feature dimension of the prototype matrix, making the classification conflict-robust and the feature embedding compact. Our method significantly outperforms state-of-the-art unsupervised and supervised image retrieval approaches on multiple benchmarks. The code and pre-trained models are released to facilitate future research https://github.com/deepglint/unicom.

Abstract: Visual Place Recognition is a task that aims to predict the coordinates of an image (called query) based solely on visual clues. Most commonly, a retrieval approach is adopted, where the query is matched to the most similar images from a large database of geotagged photos, using learned global descriptors. Despite recent advances, recognizing the same place when the query comes from a significantly different distribution is still a major hurdle for state of the art retrieval methods. Examples are heavy illumination changes (e.g. night-time images) or substantial occlusions (e.g. transient objects). In this work we explore whether re-ranking methods based on spatial verification can tackle these challenges, following the intuition that local descriptors are inherently more robust than global features to domain shifts. To this end, we provide a new, comprehensive benchmark on current state of the art models. We also introduce two new demanding datasets with night and occluded queries, to be matched against a city-wide database. Code and datasets are available at https://github.com/gbarbarani/re-ranking-for-VPR.

Abstract: The prevalence of breast cancer continues to grow, affecting about 300,000 females in the United States in 2023. However, there are different levels of severity of breast cancer requiring different treatment strategies, and hence, grading breast cancer has become a vital component of breast cancer diagnosis and treatment planning. Specifically, the gold-standard Scarff-Bloom-Richardson (SBR) grade has been shown to consistently indicate a patient's response to chemotherapy. Unfortunately, the current method to determine the SBR grade requires removal of some cancer cells from the patient which can lead to stress and discomfort along with costly expenses. In this paper, we study the efficacy of deep learning for breast cancer grading based on synthetic correlated diffusion (CDI$^s$) imaging, a new magnetic resonance imaging (MRI) modality and found that it achieves better performance on SBR grade prediction compared to those learnt using gold-standard imaging modalities. Hence, we introduce Cancer-Net BCa-S, a volumetric deep radiomics approach for predicting SBR grade based on volumetric CDI$^s$ data. Given the promising results, this proposed method to identify the severity of the cancer would allow for better treatment decisions without the need for a biopsy. Cancer-Net BCa-S has been made publicly available as part of a global open-source initiative for advancing machine learning for cancer care.

Abstract: Masked image modeling (MIM) has attracted much research attention due to its promising potential for learning scalable visual representations. In typical approaches, models usually focus on predicting specific contents of masked patches, and their performances are highly related to pre-defined mask strategies. Intuitively, this procedure can be considered as training a student (the model) on solving given problems (predict masked patches). However, we argue that the model should not only focus on solving given problems, but also stand in the shoes of a teacher to produce a more challenging problem by itself. To this end, we propose Hard Patches Mining (HPM), a brand-new framework for MIM pre-training. We observe that the reconstruction loss can naturally be the metric of the difficulty of the pre-training task. Therefore, we introduce an auxiliary loss predictor, predicting patch-wise losses first and deciding where to mask next. It adopts a relative relationship learning strategy to prevent overfitting to exact reconstruction loss values. Experiments under various settings demonstrate the effectiveness of HPM in constructing masked images. Furthermore, we empirically find that solely introducing the loss prediction objective leads to powerful representations, verifying the efficacy of the ability to be aware of where is hard to reconstruct.

Abstract: Multiscale video transformers have been explored in a wide variety of vision tasks. To date, however, the multiscale processing has been confined to the encoder or decoder alone. We present a unified multiscale encoder-decoder transformer that is focused on dense prediction tasks in videos. Multiscale representation at both encoder and decoder yields key benefits of implicit extraction of spatiotemporal features (i.e. without reliance on input optical flow) as well as temporal consistency at encoding and coarseto-fine detection for high-level (e.g. object) semantics to guide precise localization at decoding. Moreover, we propose a transductive learning scheme through many-to-many label propagation to provide temporally consistent predictions. We showcase our Multiscale Encoder-Decoder Video Transformer (MED-VT) on Automatic Video Object Segmentation (AVOS) and actor/action segmentation, where we outperform state-of-the-art approaches on multiple benchmarks using only raw images, without using optical flow.

Abstract: Sign languages are used as a primary language by approximately 70 million D/deaf people world-wide. However, most communication technologies operate in spoken and written languages, creating inequities in access. To help tackle this problem, we release ASL Citizen, the largest Isolated Sign Language Recognition (ISLR) dataset to date, collected with consent and containing 83,912 videos for 2,731 distinct signs filmed by 52 signers in a variety of environments. We propose that this dataset be used for sign language dictionary retrieval for American Sign Language (ASL), where a user demonstrates a sign to their own webcam with the aim of retrieving matching signs from a dictionary. We show that training supervised machine learning classifiers with our dataset greatly advances the state-of-the-art on metrics relevant for dictionary retrieval, achieving, for instance, 62% accuracy and a recall-at-10 of 90%, evaluated entirely on videos of users who are not present in the training or validation sets. An accessible PDF of this article is available at https://aashakadesai.github.io/research/ASL_Dataset__arxiv_.pdf

Abstract: Variational autoencoders (VAEs) are powerful generative modelling methods, however they suffer from blurry generated samples and reconstructions compared to the images they have been trained on. Significant research effort has been spent to increase the generative capabilities by creating more flexible models but often flexibility comes at the cost of higher complexity and computational cost. Several works have focused on altering the reconstruction term of the evidence lower bound (ELBO), however, often at the expense of losing the mathematical link to maximizing the likelihood of the samples under the modeled distribution. Here we propose a new formulation of the reconstruction term for the VAE that specifically penalizes the generation of blurry images while at the same time still maximizing the ELBO under the modeled distribution. We show the potential of the proposed loss on three different data sets, where it outperforms several recently proposed reconstruction losses for VAEs.

Abstract: Visual localization is a core component in many applications, including augmented reality (AR). Localization algorithms compute the camera pose of a query image w.r.t. a scene representation, which is typically built from images. This often requires capturing and storing large amounts of data, followed by running Structure-from-Motion (SfM) algorithms. An interesting, and underexplored, source of data for building scene representations are 3D models that are readily available on the Internet, e.g., hand-drawn CAD models, 3D models generated from building footprints, or from aerial images. These models allow to perform visual localization right away without the time-consuming scene capturing and model building steps. Yet, it also comes with challenges as the available 3D models are often imperfect reflections of reality. E.g., the models might only have generic or no textures at all, might only provide a simple approximation of the scene geometry, or might be stretched. This paper studies how the imperfections of these models affect localization accuracy. We create a new benchmark for this task and provide a detailed experimental evaluation based on multiple 3D models per scene. We show that 3D models from the Internet show promise as an easy-to-obtain scene representation. At the same time, there is significant room for improvement for visual localization pipelines. To foster research on this interesting and challenging task, we release our benchmark at v-pnk.github.io/cadloc.

Abstract: Continuous mid-air hand gesture recognition based on captured hand pose streams is fundamental for human-computer interaction, particularly in AR / VR. However, many of the methods proposed to recognize heterogeneous hand gestures are tested only on the classification task, and the real-time low-latency gesture segmentation in a continuous stream is not well addressed in the literature. For this task, we propose the On-Off deep Multi-View Multi-Task paradigm (OO-dMVMT). The idea is to exploit multiple time-local views related to hand pose and movement to generate rich gesture descriptions, along with using heterogeneous tasks to achieve high accuracy. OO-dMVMT extends the classical MVMT paradigm, where all of the multiple tasks have to be active at each time, by allowing specific tasks to switch on/off depending on whether they can apply to the input. We show that OO-dMVMT defines the new SotA on continuous/online 3D skeleton-based gesture recognition in terms of gesture classification accuracy, segmentation accuracy, false positives, and decision latency while maintaining real-time operation.

Abstract: Hyperspectral image (HSI) classification is an important topic in the field of remote sensing, and has a wide range of applications in Earth science. HSIs contain hundreds of continuous bands, which are characterized by high dimension and high correlation between adjacent bands. The high dimension and redundancy of HSI data bring great difficulties to HSI classification. In recent years, a large number of HSI feature extraction and classification methods based on deep learning have been proposed. However, their ability to model the global relationships among samples in both spatial and spectral domains is still limited. In order to solve this problem, an HSI classification method with spectral-spatial diffusion models is proposed. The proposed method realizes the reconstruction of spectral-spatial distribution of the training samples with the forward and reverse spectral-spatial diffusion process, thus modeling the global spatial-spectral relationship between samples. Then, we use the spectral-spatial denoising network of the reverse process to extract the unsupervised diffusion features. Features extracted by the spectral-spatial diffusion models can achieve cross-sample perception from the reconstructed distribution of the training samples, thus obtaining better classification performance. Experiments on three public HSI datasets show that the proposed method can achieve better performance than the state-of-the-art methods. The source code and the pre-trained spectral-spatial diffusion model will be publicly available at https://github.com/chenning0115/SpectralDiff.

Abstract: We present ImageReward -- the first general-purpose text-to-image human preference reward model -- to address various prevalent issues in generative models and align them with human values and preferences. Its training is based on our systematic annotation pipeline that covers both the rating and ranking components, collecting a dataset of 137k expert comparisons to date. In human evaluation, ImageReward outperforms existing scoring methods (e.g., CLIP by 38.6\%), making it a promising automatic metric for evaluating and improving text-to-image synthesis. The reward model is publicly available via the \texttt{image-reward} package at \url{https://github.com/THUDM/ImageReward}.

Abstract: In recent years, the success of large-scale vision-language models (VLMs) such as CLIP has led to their increased usage in various computer vision tasks. These models enable zero-shot inference through carefully crafted instructional text prompts without task-specific supervision. However, the potential of VLMs for generalization tasks in remote sensing (RS) has not been fully realized. To address this research gap, we propose a novel image-conditioned prompt learning strategy called the Visual Attention Parameterized Prompts Learning Network (APPLeNet). APPLeNet emphasizes the importance of multi-scale feature learning in RS scene classification and disentangles visual style and content primitives for domain generalization tasks. To achieve this, APPLeNet combines visual content features obtained from different layers of the vision encoder and style properties obtained from feature statistics of domain-specific batches. An attention-driven injection module is further introduced to generate visual tokens from this information. We also introduce an anti-correlation regularizer to ensure discrimination among the token embeddings, as this visual information is combined with the textual tokens. To validate APPLeNet, we curated four available RS benchmarks and introduced experimental protocols and datasets for three domain generalization tasks. Our results consistently outperform the relevant literature and code is available at https://github.com/mainaksingha01/APPLeNet

Abstract: The swift and precise detection of vehicles holds significant research significance in intelligent transportation systems (ITS). However, current vehicle detection algorithms encounter challenges such as high computational complexity, low detection rate, and limited feasibility on mobile devices. To address these issues, this paper proposes a lightweight vehicle detection algorithm for YOLOv7-tiny called Ghost-YOLOv7. The model first scales the width multiple to 0.5 and replaces the standard convolution of the backbone network with Ghost convolution to achieve a lighter network and improve the detection speed; secondly, a Ghost bi-directional feature pyramid network (Ghost-BiFPN) neck network is designed to enhance feature extraction capability of the algorithm and enrich semantic information; thirdly, a Ghost Decouoled Head (GDH) is employed for accurate prediction of vehicle location and class, enhancing model accuracy; finally, a coordinate attention mechanism is introduced in the output layer to suppress environmental interference, and the WIoU loss function is employed to enhance the detection accuracy further. Experimental results on the PASCAL VOC dataset demonstrate that Ghost-YOLOv7 outperforms the original YOLOv7-tiny model, achieving a 29.8% reduction in computation, 37.3% reduction in the number of parameters, 35.1% reduction in model weights, and 1.1% higher mean average precision (mAP), while achieving a detection speed of 428 FPS. These results validate the effectiveness of the proposed method.

Abstract: In the realm of 3D-computer vision applications, point cloud few-shot learning plays a critical role. However, it poses an arduous challenge due to the sparsity, irregularity, and unordered nature of the data. Current methods rely on complex local geometric extraction techniques such as convolution, graph, and attention mechanisms, along with extensive data-driven pre-training tasks. These approaches contradict the fundamental goal of few-shot learning, which is to facilitate efficient learning. To address this issue, we propose GPr-Net (Geometric Prototypical Network), a lightweight and computationally efficient geometric prototypical network that captures the intrinsic topology of point clouds and achieves superior performance. Our proposed method, IGI++ (Intrinsic Geometry Interpreter++) employs vector-based hand-crafted intrinsic geometry interpreters and Laplace vectors to extract and evaluate point cloud morphology, resulting in improved representations for FSL (Few-Shot Learning). Additionally, Laplace vectors enable the extraction of valuable features from point clouds with fewer points. To tackle the distribution drift challenge in few-shot metric learning, we leverage hyperbolic space and demonstrate that our approach handles intra and inter-class variance better than existing point cloud few-shot learning methods. Experimental results on the ModelNet40 dataset show that GPr-Net outperforms state-of-the-art methods in few-shot learning on point clouds, achieving utmost computational efficiency that is $170\times$ better than all existing works. The code is publicly available at https://github.com/TejasAnvekar/GPr-Net.

Abstract: Computer vision applications in transportation logistics and warehousing have a huge potential for process automation. We present a structured literature review on research in the field to help leverage this potential. All literature is categorized w.r.t. the application, i.e. the task it tackles and w.r.t. the computer vision techniques that are used. Regarding applications, we subdivide the literature in two areas: Monitoring, i.e. observing and retrieving relevant information from the environment, and manipulation, where approaches are used to analyze and interact with the environment. In addition to that, we point out directions for future research and link to recent developments in computer vision that are suitable for application in logistics. Finally, we present an overview of existing datasets and industrial solutions. We conclude that while already many research areas have been investigated, there is still huge potential for future research. The results of our analysis are also available online at https://a-nau.github.io/cv-in-logistics.

Abstract: The most recent efforts in video matting have focused on eliminating trimap dependency since trimap annotations are expensive and trimap-based methods are less adaptable for real-time applications. Despite the latest tripmap-free methods showing promising results, their performance often degrades when dealing with highly diverse and unstructured videos. We address this limitation by introducing Adaptive Matting for Dynamic Videos, termed AdaM, which is a framework designed for simultaneously differentiating foregrounds from backgrounds and capturing alpha matte details of human subjects in the foreground. Two interconnected network designs are employed to achieve this goal: (1) an encoder-decoder network that produces alpha mattes and intermediate masks which are used to guide the transformer in adaptively decoding foregrounds and backgrounds, and (2) a transformer network in which long- and short-term attention combine to retain spatial and temporal contexts, facilitating the decoding of foreground details. We benchmark and study our methods on recently introduced datasets, showing that our model notably improves matting realism and temporal coherence in complex real-world videos and achieves new best-in-class generalizability. Further details and examples are available at https://github.com/microsoft/AdaM.

Abstract: Due to the difficulty in collecting large-scale and perfectly aligned paired training data for Under-Display Camera (UDC) image restoration, previous methods resort to monitor-based image systems or simulation-based methods, sacrificing the realness of the data and introducing domain gaps. In this work, we revisit the classic stereo setup for training data collection -- capturing two images of the same scene with one UDC and one standard camera. The key idea is to "copy" details from a high-quality reference image and "paste" them on the UDC image. While being able to generate real training pairs, this setting is susceptible to spatial misalignment due to perspective and depth of field changes. The problem is further compounded by the large domain discrepancy between the UDC and normal images, which is unique to UDC restoration. In this paper, we mitigate the non-trivial domain discrepancy and spatial misalignment through a novel Transformer-based framework that generates well-aligned yet high-quality target data for the corresponding UDC input. This is made possible through two carefully designed components, namely, the Domain Alignment Module (DAM) and Geometric Alignment Module (GAM), which encourage robust and accurate discovery of correspondence between the UDC and normal views. Extensive experiments show that high-quality and well-aligned pseudo UDC training pairs are beneficial for training a robust restoration network. Code and the dataset are available at https://github.com/jnjaby/AlignFormer.

Abstract: We propose $\textbf{VidStyleODE}$, a spatiotemporally continuous disentangled $\textbf{Vid}$eo representation based upon $\textbf{Style}$GAN and Neural-$\textbf{ODE}$s. Effective traversal of the latent space learned by Generative Adversarial Networks (GANs) has been the basis for recent breakthroughs in image editing. However, the applicability of such advancements to the video domain has been hindered by the difficulty of representing and controlling videos in the latent space of GANs. In particular, videos are composed of content (i.e., appearance) and complex motion components that require a special mechanism to disentangle and control. To achieve this, VidStyleODE encodes the video content in a pre-trained StyleGAN $\mathcal{W}_+$ space and benefits from a latent ODE component to summarize the spatiotemporal dynamics of the input video. Our novel continuous video generation process then combines the two to generate high-quality and temporally consistent videos with varying frame rates. We show that our proposed method enables a variety of applications on real videos: text-guided appearance manipulation, motion manipulation, image animation, and video interpolation and extrapolation. Project website: https://cyberiada.github.io/VidStyleODE

Abstract: This paper presents a new annotation method called Sparse Annotation (SA) for crowd counting, which reduces human labeling efforts by sparsely labeling individuals in an image. We argue that sparse labeling can reduce the redundancy of full annotation and capture more diverse information from distant individuals that is not fully captured by Partial Annotation methods. Besides, we propose a point-based Progressive Point Matching network (PPM) to better explore the crowd from the whole image with sparse annotation, which includes a Proposal Matching Network (PMN) and a Performance Restoration Network (PRN). The PMN generates pseudo-point samples using a basic point classifier, while the PRN refines the point classifier with the pseudo points to maximize performance. Our experimental results show that PPM outperforms previous semi-supervised crowd counting methods with the same amount of annotation by a large margin and achieves competitive performance with state-of-the-art fully-supervised methods.

Abstract: Segmenting anything is a ground-breaking step toward artificial general intelligence, and the Segment Anything Model (SAM) greatly fosters the foundation models for computer vision. We could not be more excited to probe the performance traits of SAM. In particular, exploring situations in which SAM does not perform well is interesting. In this report, we choose three concealed scenes, i.e., camouflaged animals, industrial defects, and medical lesions, to evaluate SAM under unprompted settings. Our main observation is that SAM looks unskilled in concealed scenes.

Abstract: Automatic perception of human behaviors during social interactions is crucial for AR/VR applications, and an essential component is estimation of plausible 3D human pose and shape of our social partners from the egocentric view. One of the biggest challenges of this task is severe body truncation due to close social distances in egocentric scenarios, which brings large pose ambiguities for unseen body parts. To tackle this challenge, we propose a novel scene-conditioned diffusion method to model the body pose distribution. Conditioned on the 3D scene geometry, the diffusion model generates bodies in plausible human-scene interactions, with the sampling guided by a physics-based collision score to further resolve human-scene inter-penetrations. The classifier-free training enables flexible sampling with different conditions and enhanced diversity. A visibility-aware graph convolution model guided by per-joint visibility serves as the diffusion denoiser to incorporate inter-joint dependencies and per-body-part control. Extensive evaluations show that our method generates bodies in plausible interactions with 3D scenes, achieving both superior accuracy for visible joints and diversity for invisible body parts. The code will be available at https://sanweiliti.github.io/egohmr/egohmr.html.

Abstract: We present DreamPose, a diffusion-based method for generating animated fashion videos from still images. Given an image and a sequence of human body poses, our method synthesizes a video containing both human and fabric motion. To achieve this, we transform a pretrained text-to-image model (Stable Diffusion) into a pose-and-image guided video synthesis model, using a novel finetuning strategy, a set of architectural changes to support the added conditioning signals, and techniques to encourage temporal consistency. We fine-tune on a collection of fashion videos from the UBC Fashion dataset. We evaluate our method on a variety of clothing styles and poses, and demonstrate that our method produces state-of-the-art results on fashion video animation. Video results are available on our project page.

Abstract: Recent works demonstrate a remarkable ability to customize text-to-image diffusion models while only providing a few example images. What happens if you try to customize such models using multiple, fine-grained concepts in a sequential (i.e., continual) manner? In our work, we show that recent state-of-the-art customization of text-to-image models suffer from catastrophic forgetting when new concepts arrive sequentially. Specifically, when adding a new concept, the ability to generate high quality images of past, similar concepts degrade. To circumvent this forgetting, we propose a new method, C-LoRA, composed of a continually self-regularized low-rank adaptation in cross attention layers of the popular Stable Diffusion model. Furthermore, we use customization prompts which do not include the word of the customized object (i.e., "person" for a human face dataset) and are initialized as completely random embeddings. Importantly, our method induces only marginal additional parameter costs and requires no storage of user data for replay. We show that C-LoRA not only outperforms several baselines for our proposed setting of text-to-image continual customization, which we refer to as Continual Diffusion, but that we achieve a new state-of-the-art in the well-established rehearsal-free continual learning setting for image classification. The high achieving performance of C-LoRA in two separate domains positions it as a compelling solution for a wide range of applications, and we believe it has significant potential for practical impact.

Abstract: We present RECLIP (Resource-efficient CLIP), a simple method that minimizes computational resource footprint for CLIP (Contrastive Language Image Pretraining). Inspired by the notion of coarse-to-fine in computer vision, we leverage small images to learn from large-scale language supervision efficiently, and finetune the model with high-resolution data in the end. Since the complexity of the vision transformer heavily depends on input image size, our approach significantly reduces the training resource requirements both in theory and in practice. Using the same batch size and training epoch, RECLIP achieves highly competitive zero-shot classification and image text retrieval accuracy with 6 to 8$\times$ less computational resources and 7 to 9$\times$ fewer FLOPs than the baseline. Compared to the state-of-the-art contrastive learning methods, RECLIP demonstrates 5 to 59$\times$ training resource savings while maintaining highly competitive zero-shot classification and retrieval performance. We hope this work will pave the path for the broader research community to explore language supervised pretraining in more resource-friendly settings.