Adversarial robustness generally relies on large-scale architectures and datasets alongside extensive adversary generation, hindering resource-efficient deployment. For scalable solutions, adversarially robust knowledge distillation has emerged as a principle strategy, facilitating the transfer of adversarial robustness from a large-scale teacher model to a lightweight student model. However, existing works focus solely on sample-to-sample alignment of features or predictions between the teacher and student models, overlooking the vital role of their statistical alignment. Thus, we propose a novel adversarially robust knowledge distillation method that integrates the alignment of feature distributions between the teacher and student backbones under adversarial and clean sample sets. To motivate our idea, for an adversarially trained model (e.g., student or teacher), we show that the adversarially robust accuracy (evaluated on testing adversarial samples under an increasing perturbation radius) correlates negatively with the gap between the feature variance evaluated on testing adversarial samples and testing clean samples. Such a negative correlation exhibits a strong linear trend, suggesting that aligning the feature covariance of the student model toward the feature covariance of the teacher model should improve the adversarial robustness of the student model by reducing the variance gap. A similar trend is observed by reducing the variance gap between the gram matrices of the student and teacher models. Extensive evaluations highlight the state-of-the-art adversarial robustness and natural performance of our method across diverse datasets and distillation scenarios.
论文检索
输入标题、作者或关键词，从 2,387 篇学术成果中精准定位
Object counting methods typically rely on manually annotated datasets. The cost of creating such datasets has restricted the versatility of these networks to count objects from specific classes (such as humans or penguins), and counting objects from diverse categories remains a challenge. The availability of robust text-to-image latent diffusion models (LDMs) raises the question of whether these models can be utilized to generate counting datasets. However, LDMs struggle to create images with an exact number of objects based solely on text prompts but they can be used to offer a dependable \textit{sorting} signal by adding and removing objects within an image. Leveraging this data, we initially introduce an unsupervised sorting methodology to learn object-related features that are subsequently refined and anchored for counting purposes using counting data generated by LDMs. Further, we present a density classifier-guided method for dividing an image into patches containing objects that can be reliably counted. Consequently, we can generate counting data for any type of object and count them in an unsupervised manner. Our approach outperforms other unsupervised and few-shot alternatives and is not restricted to specific object classes for which counting data is available. Code to be released upon acceptance.
Novel view synthesis has achieved remarkable quality and efficiency by the paradigm of 3D Gaussian Splatting (3D-GS), but still faces two challenges: 1) significant performance degradation when trained with only few-shot samples due to a lack of geometry constraint, and 2) incapability of rendering at a higher resolution that is beyond the input resolution of training samples. In this paper, we propose Dual-Lens 3D-GS (DL-GS) to achieve high-resolution (HR) and few-shot view synthesis, by leveraging the characteristics of the asymmetric dual-lens system commonly equipped on mobile devices. This kind of system captures the same scene with different focal lengths (\textit{i.e.}, wide-angle and telephoto) under an asymmetric stereo configuration, which naturally provides geometric hints for few-shot training and HR guidance for resolution improvement. Nevertheless, there remain two major technical problems to achieving this goal. First, how to effectively exploit the geometry information from the asymmetric stereo configuration? To this end, we propose a consistency-aware training strategy, which integrates a dual-lens-consistent loss to regularize the 3D-GS optimization. Second, how to make the best use of the dual-lens training samples to effectively improve the resolution of newly synthesized views? To this end, we design a multi-reference-guided refinement module to select proper telephoto and wide-angle guided images from training samples based on the camera pose distances, and then exploit their information for high-frequency detail enhancement. Extensive experiments on simulated and real-captured datasets validate the distinct superiority of our DL-GS over various competitors on the task of HR and few-shot view synthesis.
MapDistill: Boosting Efficient Camera-based HD Map Construction via Camera-LiDAR Fusion Model Distillation
PDF ↗Online high-definition (HD) map construction is an important and challenging task in autonomous driving. Recently, there has been a growing interest in cost-effective multi-view camera-based methods without relying on other sensors like LiDAR. However, these methods suffer from a lack of explicit depth information, necessitating the use of large models to achieve satisfactory performance. To address this, we employ the Knowledge Distillation (KD) idea for efficient HD map construction for the first time and introduce a novel KD-based approach called MapDistill to transfer knowledge from a high-performance Camera-LiDAR fusion model to a lightweight camera-only model. Specifically, we adopt the teacher-student architecture, i.e., a Camera-LiDAR fusion model as the teacher and a lightweight camera model as the student, and devise a dual BEV transform module to facilitate cross-modal knowledge distillation while maintaining cost-effective camera-only deployment. Additionally, we present a comprehensive distillation scheme encompassing cross-modal relation distillation, dual-level feature distillation, and map head distillation. This approach alleviates knowledge transfer challenges between modalities, enabling the student model to learn improved feature representations for HD map construction. Experimental results on the challenging nuScenes dataset demonstrate the effectiveness of MapDistill, surpassing existing competitors by over 7.7 mAP or 4.5x speedup. The source code will be released.
Depth completion is the task of generating a dense depth map given an image and a sparse depth map as inputs. It has important applications in various downstream tasks. In this paper, we present OGNI-DC, a novel framework for depth completion. The key to our method is "Optimization-Guided Neural Iterations" (OGNI). It consists of a recurrent unit that refines a depth gradient field and a differentiable depth integrator that integrates the depth gradients into a depth map. OGNI-DC exhibits strong generalization, outperforming baselines by a large margin on unseen datasets and across various sparsity levels. Moreover, OGNI-DC has high accuracy, achieving state-of-the-art performance on the NYUv2 and the KITTI benchmarks. Code is attached for reviewing and will be released if the paper is accepted.
CPT-VR: Improving Surface Rendering via Closest Point Transform with View-Reflection Appearance
PDF ↗Differentiable surface rendering has significantly advanced 3D reconstruction. Existing surface rendering methods assume that the local surface is planar, and thus employ linear approximation based on the Singed Distance Field (SDF) values to predict the point on the surface. However, this assumption overlooks the inherently irregular and non-planar nature of object surfaces in the real world. Consequently, the approximate points tend to deviate from the zero-level set, affecting the fidelity of the reconstructed shape. In this paper, we propose a novel surface rendering method termed CPT-VR, which leverages the Closet Point Transform (CPT) and View and Reflection direction vectors to enhance the quality of reconstruction. Specifically, leveraging the physical property of CPT that accurately projects points near the surface onto the zero-level set, we correct the deviated points, thus achieving an accurate geometry representation. Based on our accurate geometry representation, incorporating the reflection vector into our method can facilitate the appearance modeling of specular regions. Moreover, to enable our method to no longer be dependent on any prior knowledge of the background, we present a background model to learn the background appearance. Compared to previous state-of-the-art methods, CPT-VR achieves better surface reconstruction quality, even for cases with complex structures and specular highlights.
Digitizing static scenes and dynamic events from multi-view images has long been a challenge in the fields of computer vision and graphics. Recently, 3D Gaussian Splatting has emerged as a practical and scalable method for reconstruction, gaining popularity due to its impressive quality of reconstruction, real-time rendering speeds, and compatibility with widely used visualization tools. However, the method requires a substantial number of input views to achieve high-quality scene reconstruction, introducing a significant practical bottleneck. This challenge is especially pronounced in capturing dynamic scenes, where deploying an extensive camera array can be prohibitively costly. In this work, we identify the lack of spatial autocorrelation as one of the factors contributing to the suboptimal performance of the 3DGS technique in sparse reconstruction settings. To address the issue, we propose an optimization strategy that effectively regularizes splat features by modeling them as the outputs of a corresponding implicit neural field. This results in a consistent enhancement of reconstruction quality across various scenarios. Our approach adeptly manages both static and dynamic cases, as demonstrated by extensive testing across different setups and scene complexities.
Current optical flow and point-tracking methods rely heavily on synthetic datasets. Event cameras are novel vision sensors with advantages in challenging visual conditions, but state-of-the-art frame-based methods cannot be easily adapted to event data due to the limitations of current event simulators. We introduce a novel self-supervised loss combining the Contrast Maximization framework with a non-linear motion prior in the form of pixel-level trajectories and propose an efficient solution to solve the high-dimensional assignment problem between non-linear trajectories and events. Their effectiveness is demonstrated in two scenarios: In dense continuous-time motion estimation, our method improves the zero-shot performance of a synthetically trained model on the real-world dataset EVIMO2 by 29%. In optical flow estimation, our method elevates a simple UNet to achieve state-of-the-art performance among self-supervised methods on the DSEC optical flow benchmark.
Vision-language models (VLMs) are typically composed of a vision encoder, e.g. CLIP, and a language model (LM) that interprets the encoded features to solve downstream tasks. Despite remarkable progress, VLMs are subject to several shortcomings due to the limited capabilities of vision encoders, e.g. ``blindness'' to certain image features, visual hallucination, etc. To address these issues, we study broadening of the visual encoding capabilities of VLMs. We first comprehensively benchmark several vision encoders with different inductive biases for solving VLM tasks. We observe that there is no single encoding configuration that consistently achieves top performance across different tasks, and encoders with different biases can perform surprisingly similarly. Motivated by this, we introduce a method, named BRAVE, that consolidates features from multiple frozen encoders into a more versatile representation that can be directly fed as the input to a frozen LM. BRAVE achieves state-of-the-art performance on a broad range of captioning and VQA benchmarks and significantly reduces the aforementioned issues of VLMs, while requiring a smaller number of trainable parameters than existing methods and having a more compressed representation. Our results highlight the potential of incorporating different visual biases for a more broad and contextualized visual understanding of VLMs.
Recent foundation models trained on a tremendous scale of data have shown great promise in a wide range of computer vision tasks and application domains. However, less attention has been paid to the marine realms, which in contrast cover the majority of our blue planet. The scarcity of labeled data is the most hindering issue, and marine photographs illustrate significantly different appearances and contents from general in-air images. Using existing foundation models for marine visual analysis does not yield satisfactory performance, due to not only the data distribution shift, but also the intrinsic limitations of the existing foundation models (e.g., lacking semantics, redundant mask generation, or restricted to image-level scene understanding). In this work, we emphasize both model and data approaches for understanding marine ecosystems. We introduce MarineInst, a foundation model for the analysis of the marine realms with instance visual description, which outputs instance masks and captions for marine object instances. To train MarineInst, we acquire MarineInst20M, the largest marine image dataset to date, which contains a wide spectrum of marine images with high-quality semantic instance masks constructed by a mixture of human-annotated instance masks and model-generated instance masks from our automatic procedure of binary instance filtering. To generate informative and detailed semantic instance captions, we use vision-language models to produce semantic richness with various granularities. Our model and dataset support a wide range of marine visual analysis tasks, from image-level scene understanding to regional mask-level instance understanding. More significantly, MarineInst exhibits strong generalization ability and flexibility to support a wide range of downstream tasks with state-of-the-art performance.
Optimizing Factorized Encoder Models: Time and Memory Reduction for Scalable and Efficient Action Recognition
PDF ↗In this paper, we address the challenges posed by the substantial training time and memory consumption associated with video transformers, focusing on the ViViT (Video Vision Transformer) model, in particular the Factorised Encoder version, as our baseline for action recognition tasks. The factorised encoder variant follows the late-fusion approach that is adopted by many state of the art approaches. Despite standing out for its favorable speed/accuracy tradeoffs among the different variants of ViViT, its considerable training time and memory requirements still pose a significant barrier to entry. Our method is designed to lower this barrier and is based on the idea of freezing the spatial transformer during training. This leads to a low accuracy model if naively done. But we show that by (1) appropriately initializing the temporal transformer (a module responsible for processing temporal information) (2) introducing a compact adapter model connecting frozen spatial representations (a module that selectively focuses on regions of the input image) to the temporal transformer, we can enjoy the benefits of freezing the spatial transformer without sacrificing accuracy. Through extensive experimentation over 6 benchmarks, we demonstrate that our proposed training strategy significantly reduces training costs (by ) and memory consumption while maintaining or slightly improving performance by up to 1.79\% compared to the baseline model. Our approach additionally unlocks the capability to utilize larger image transformer models as our spatial transformer and access more frames with the same memory consumption. We also show the generalization of this approach to other factorized encoder models. The advancements made in this work have the potential to advance research in the video understanding domain and provide valuable insights for researchers and practitioners with limited resources, paving the way for more efficient and scalable alternatives in the action recognition field.
BAM-DETR: Boundary-Aligned Moment Detection Transformer for Temporal Sentence Grounding in Videos
PDF ↗Temporal sentence grounding aims to localize moments relevant to a language description. Recently, DETR-like approaches achieved notable progress by predicting the center and length of a target moment. However, they suffer from the issue of center misalignment raised by the inherent ambiguity of moment centers, leading to inaccurate predictions. To remedy this problem, we propose a novel boundary-oriented moment formulation. In our paradigm, the model no longer needs to find the precise center but instead suffices to predict any anchor point within the interval, from which the boundaries are directly estimated. Based on this idea, we design a boundary-aligned moment detection transformer, equipped with a dual-pathway decoding process. Specifically, it refines the anchor and boundaries within parallel pathways using global and boundary-focused attention, respectively. This separate design allows the model to focus on desirable regions, enabling precise refinement of moment predictions. Further, we propose a quality-based ranking method, ensuring that proposals with high localization qualities are prioritized over incomplete ones. Experiments on three benchmarks validate the effectiveness of the proposed methods. The code will be publicly available.
Generating natural human grasps necessitates consideration of not just object geometry but also semantic information. Solely depending on object shape for grasp generation confines the applications of prior methods in downstream tasks. This paper presents a novel semantic-based grasp generation method, termed SemGrasp, which generates a static human grasp pose by incorporating semantic information into the grasp representation. We introduce a discrete representation that aligns the grasp space with semantic space, enabling the generation of grasp postures in accordance with language instructions. A Multimodal Large Language Model (MLLM) is subsequently fine-tuned, integrating object, grasp, and language within a unified semantic space. To facilitate the training of SemGrasp, we compile a large-scale, grasp-text-aligned dataset named CapGrasp, featuring over 300k detailed captions and 50k diverse grasps. Experimental findings demonstrate that SemGrasp efficiently generates natural human grasps in alignment with linguistic intentions. Our code, models, and dataset are available publicly at: https://kailinli.github.io/SemGrasp.
Domain Generalization (DG), designed to enhance out-of-distribution (OOD) generalization, is all about learning invariance against domain shifts utilizing sufficient supervision signals. Yet, the scarcity of such labeled data has led to the rise of unsupervised domain generalization (UDG) — a more important yet challenging task in that models are trained across diverse domains in an unsupervised manner and eventually tested on unseen domains. UDG is fast gaining attention but is still far from well-studied. To close the research gap, we propose a novel learning framework designed for UDG, termed the Disentangled Masked AutoEncoder (DisMAE), aiming to discover the disentangled representations that faithfully reveal the intrinsic features and superficial variations without access to the class label. At its core is the distillation of domain-invariant semantic features, which cannot be distinguished by domain classifier, while filtering out the domain-specific variations (for example, color schemes and texture patterns) that are unstable and redundant. Notably, DisMAE co-trains the asymmetric dual-branch architecture with semantic and lightweight variation encoders, offering dynamic data manipulation and representation level augmentation capabilities. Extensive experiments on four benchmark datasets (i.e., DomainNet, PACS, VLCS, and Colored MNIST) with both DG and UDG tasks demonstrate that DisMAE can achieve competitive OOD performance compared with the state-of-the-art DG and UDG baselines, which shed light on potential research lines for improving generalization ability with large-scale unlabeled data.
Enhancing Perceptual Quality in Video Super-Resolution through Temporally-Consistent Detail Synthesis using Diffusion Models
PDF ↗In this paper, we address the problem of enhancing perceptual quality in video super-resolution (VSR) using Diffusion Models (DMs) while ensuring temporal consistency among frames. We present StableVSR, a VSR method based on DMs that can significantly enhance the perceptual quality of upscaled videos by synthesizing realistic and temporally-consistent details. We introduce the Temporal Conditioning Module (TCM) into a pre-trained DM for single image super-resolution to turn it into a VSR method. TCM uses the novel Temporal Texture Guidance, which provides it with spatially-aligned and detail-rich texture information synthesized in adjacent frames. This guides the generative process of the current frame toward high-quality and temporally-consistent results. In addition, we introduce the novel Frame-wise Bidirectional Sampling strategy to encourage the use of information from past to future and vice-versa. This strategy improves the perceptual quality of the results and the temporal consistency across frames. We demonstrate the effectiveness of StableVSR in enhancing the perceptual quality of upscaled videos while achieving better temporal consistency compared to existing state-of-the-art methods for VSR.
This work focuses on generating realistic, physically-based human behaviors from multi-modal inputs, which may only partially specify the desired motion. For example, the input may come from a VR controller providing arm motion and body velocity, partial key-point animation, computer vision applied to videos, or even higher-level motion goals. This requires a versatile low-level humanoid controller that can handle such sparse, under-specified guidance, seamlessly switch between skills, and recover from failures. Current approaches for learning humanoid controllers from demonstration data capture some of these characteristics, but none achieve them all. To this end, we introduce the Masked Humanoid Controller (MHC), a novel approach that applies multi-objective imitation learning on augmented and selectively masked motion demonstrations. The training methodology results in an MHC that exhibits the key capabilities of catch-up to out-of-sync input commands, combining elements from multiple motion sequences, and completing unspecified parts of motions from sparse multimodal input. We demonstrate these key capabilities for an MHC learned over a dataset of 87 diverse skills and showcase different multi-modal use cases, including integration with planning frameworks to highlight MHC’s ability to solve new user-defined tasks without any finetuning.
Developing gaze estimation models that generalize well to unseen domains and in-the-wild conditions remains a challenge with no known best solution. This is mostly due to the difficulty of acquiring ground truth data that cover the distribution of faces, head poses, and environments that exist in the real world. Most recent methods attempt to close the gap between specific source and target domains using domain adaptation. In this work, we propose to train general gaze estimation models which can be directly employed in novel environments without adaptation. To do so, we leverage the observation that head, body, and hand pose estimation benefit from revising them as dense 3D coordinate prediction, and similarly express gaze estimation as regression of dense 3D eye meshes. To close the gap between image domains, we create a large-scale dataset of diverse faces with gaze pseudo-annotations, which we extract based on the 3D geometry of the scene, and design a multi-view supervision framework to balance their effect during training. We test our method in the task of gaze generalization, in which we demonstrate improvement of up to 30% compared to state-of-the-art when no ground truth data are available, and up to 10% when they are. The project material will become available for research purposes.
Despite recent advances in large-scale text-to-image generative models, manipulating real images with these models remains a challenging problem. The main limitations of existing editing methods are that they either fail to perform with consistent quality on a wide range of image edits, or require time-consuming hyperparameter tuning or fine-tuning of the diffusion model to preserve the image-specific appearance of the input image. Most of these approaches utilize source image information via intermediate feature caching which is inserted in generation process as itself. However, such technique produce feature misalignment of the model that leads to inconsistent results. We propose a novel approach that is built upon modified diffusion sampling process via guidance mechanism. In this work, we explore self-guidance technique to preserve the overall structure of the input image and its local regions appearance that should not be edited. In particular, we explicitly introduce layout preserving energy functions that are aimed to save local and global structures of the source image. Additionally, we propose a noise rescaling mechanism that allows to preserve noise distribution by balancing the norms of classifier-free guidance and our proposed guiders during generation. It leads to more consistent and better editing results. Such guiding approach does not require fine-tuning diffusion model and exact inversion process. As a result, the proposed method provides a fast and high quality editing mechanism. In our experiments, we show through human evaluation and quantitative analysis that the proposed method allows to produce desired editing which is more preferable by the human and also achieves a better trade-off between editing quality and preservation of the original image.
Diverse and realistic floor plan data are essential for the development of useful computer-aided methods in architectural design. Today's large-scale floor plan datasets predominantly feature simple floor plan layouts, typically representing single-apartment dwellings only. To compensate for the mismatch between current datasets and the real world, we develop \textbf{Modified Swiss Dwellings} (MSD) -- the first large-scale floor plan dataset that contains a significant share of layouts of multi-apartment dwellings. MSD features over 5.3K floor plans of medium- to large-scale building complexes, covering over 18.9K distinct apartments. We validate that existing approaches for floor plan generation, while effective in simpler scenarios, cannot yet seamlessly address the challenges posed by MSD. Our benchmark calls for new research in floor plan machine understanding. Code and data are open.
Following the advent of NeRFs, 3D Gaussian Splatting (3D-GS) has paved the way to real-time neural rendering overcoming the computational burden of volumetric methods. Following the pioneering work of 3D-GS, several methods have attempted to achieve compressible and high-fidelity performance. However, by employing a geometry-agnostic optimization scheme, these methods neglect the inherent 3D structure of the scene, thereby restricting the expressivity and the quality of the representation, resulting in various floating points and artifacts. In this work, we propose a structure-aware Gaussian Splatting method (SAGS) that implicitly encodes the geometry of the scene, which reflects to state-of-the-art rendering performance and reduced storage requirements on benchmark novel-view synthesis datasets. SAGS is founded on a local-global graph representation that facilitates the learning of complex scenes and enforces meaningful point displacements that preserve the scene's geometry. Additionally, we introduce a lightweight version of SAGS, using a simple yet effective mid-point interpolation scheme, which showcases a compact representation of the scene with up to 20$\times$ size reduction without the reliance on any compression strategies. Extensive experiments across multiple benchmark datasets demonstrate the superiority of SAGS compared to state-of-the-art 3D-GS methods under both rendering quality and model size. Besides, we demonstrate that our structure-aware method can effectively mitigate floating artifacts and irregular distortions of previous methods while obtaining precise depth maps. Code and models will be publicly available.