CVPR 2025 Papers

Learning self-supervised visual correspondence is a long-studied task fundamental to visual understanding and human perception. However, existing correspondence methods largely focus on small image transformations, such as object tracking in high-framerate videos or learning pixel-to-pixel mappings between images with high view overlap. This severely limits their application in dynamic multi-view settings such as robot imitation learning. In this work, we introduce Predictive Cycle Consistency for learning object correspondence between extremely disjoint views of a scene without paired segmentation data. Our technique bootstraps object correspondence pseudolabels from raw image segmentations using conditional grayscale colorization and a cycle-consistency refinement prior. We then train deep ViTs on these pseudolabels, which we use to generate higher-quality pseudolabels and iteratively train better correspondence models. We demonstrate the performance of our method under both extreme in-the-wild camera view changes and across large temporal gaps in video. Our approach beats all prior supervised and prior SoTA self-supervised correspondence models on the EgoExo4D correspondence benchmark (+6.7 IoU Exo Query) and the prior SoTA self-supervised methods SiamMAE and DINO V1&V2 on the DAVIS-2017 and LVOS datasets across large frame gaps.

Continual learning (CL) aims to continually accumulate knowledge from a non-stationary data stream without catastrophic forgetting of learned knowledge, requiring a balance between stability and adaptability. Relying on the generalizable representation in pre-trained models (PTMs), PTM-based CL methods perform effective continual adaptation on downstream tasks by adding learnable adapters or prompts upon the frozen PTMs. However, many existing PTM-based CL methods use restricted adaptation on a fixed set of these modules to avoid forgetting, suffering from limited CL ability. Periodically adding task-specific modules results in linear model growth rate and impaired knowledge reuse. We propose Self-Expansion of pre-trained models with Modularized Adaptation (SEMA), a novel approach to enhance the control of stability-plasticity balance in PTM-based CL. SEMA automatically decides to reuse or add adapter modules on demand in CL, depending on whether significant distribution shift that cannot be handled is detected at different representation levels. We design modular adapter consisting of a functional adapter and a representation descriptor. The representation descriptors are trained as a distribution shift indicator and used to trigger self-expansion signals. For better composing the adapters, an expandable weighting router is learned jointly for mixture of adapter outputs. SEMA enables better knowledge reuse and sub-linear expansion rate. Extensive experiments demonstrate the effectiveness of the proposed self-expansion method, achieving state-of-the-art performance compared to PTM-based CL methods without memory rehearsal. Code is available at https://github.com/huiyiwang01/SEMA-CL.

Underwater (UW) robotics applications require depth and restored images simultaneously in real-time, irrespective of whether the UW images are captured in good lighting conditions or not. Most of the UW image restoration and depth estimation methods have been devised for images under normal lighting. Consequently, they struggle to perform on poorly lit images. Even though artificial illumination can be used when there is insufficient ambient light, it can introduce non-uniform lighting artifacts in the restored images. Hence, the recovery of depth and restored images directly from Low-Light UW (LLUW) images is a critical requirement in marine applications. While a few works have attempted LLUW image restoration, there are no reported works on joint recovery of depth and clean image from LLUW images. We propose a Self-supervised Low-light Underwater Image and Depth recovery network (SelfLUID-Net) for joint estimation of depth and restored image in real-time from a single LLUW image. We have collected an Underwater Low-light Stereo Video (ULVStereo) dataset which is the first-ever UW dataset with stereo pairs of low-light and normally-lit UW images. For the dual tasks of image and depth recovery from a LLUW image, we effectively utilize the stereo data from ULVStereo that provides cues for both depth and illumination-independent clean image. We harness a combination of the UW image formation process, the Retinex model, and constraints enforced by the scene geometry for our self-supervised training. To handle occlusions, we additionally utilize monocular frames from our video dataset and propose a masking scheme to prevent dynamic transients, that do not respect the underlying scene geometry, from misguiding the learning process. Evaluations on five LLUW datasets demonstrate the superiority and generalization ability of our proposed SelfLUID-Net over existing state-of-the-art methods. The dataset ULVStereo is available at https://github.com/nishavarghese15/ULVStereo.

Multimodal image-tabular learning is gaining attention, yet it faces challenges due to limited labeled data. While earlier work has applied self-supervised learning (SSL) to unlabeled data, its task-agnostic nature often results in learning suboptimal features for downstream tasks. Semi-supervised learning (SemiSL), which combines labeled and unlabeled data, offers a promising solution. However, existing multimodal SemiSL methods typically focus on unimodal or modality-shared features, ignoring valuable task-relevant modality-specific information, leading to a Modality Information Gap. In this paper, we propose STiL, a novel SemiSL tabular-image framework that addresses this gap by comprehensively exploring task-relevant information. STiL features a new disentangled contrastive consistency module to learn cross-modal invariant representations of shared information while retaining modality-specific information via disentanglement. We also propose a novel consensus-guided pseudo-labeling strategy to generate reliable pseudo-labels based on classifier consensus, along with a new prototype-guided label smoothing technique to refine pseudo-label quality with prototype embeddings, thereby enhancing task-relevant information learning in unlabeled data. Experiments on natural and medical image datasets show that STiL outperforms the state-of-the-art supervised/SSL/SemiSL image/multimodal approaches. Our code is available at https://github.com/siyi-wind/STiL.

Recently, sparsely-supervised 3D object detection has gained great attention, achieving performance close to fully-supervised 3D detectors while requiring only a few annotated instances. Nevertheless, these methods suffer challenges when accurate labels are extremely absent. In this paper, we propose a boosting strategy, termed SP3D, explicitly utilizing the cross-modal semantic prompts generated from Large Multimodal Models (LMMs) to boost the 3D detector with robust feature discrimination capability under sparse annotation settings. Specifically, we first develop a Confident Points Semantic Transfer (CPST) module that generates accurate cross-modal semantic prompts through boundary-constrained center cluster selection. Based on these accurate semantic prompts, which we treat as seed points, we introduce a Dynamic Cluster Pseudo-label Generation (DCPG) module to yield pseudo-supervision signals from the geometry shape of multi-scale neighbor points. Additionally, we design a Distribution Shape score (DS score) that chooses high-quality supervision signals for the initial training of the 3D detector. Experiments on the KITTI dataset and Waymo Open Dataset (WOD) have validated that SP3D can enhance the performance of sparsely supervised detectors by a large margin under meager labeling conditions. Moreover, we verified SP3D in the zero-shot setting, where its performance exceeded that of the state-of-the-art methods. The code is available at https://github.com/xmuqimingxia/SP3D.

Video-based Large Language Models (Video-LLMs) have witnessed substantial advancements in recent years, propelled by the advancement in multi-modal LLMs. Although these models have demonstrated proficiency in providing the overall description of videos, they struggle with fine-grained understanding, particularly in aspects such as visual dynamics and video details inquiries. To tackle these shortcomings, we find that fine-tuning Video-LLMs on self-supervised fragment tasks, greatly improve their fine-grained video understanding abilities. Hence we propose two key contributions:(1) Self-Supervised Fragment Fine-Tuning (SF^2T), a novel effortless fine-tuning method, employs the rich inherent characteristics of videos for training, while unlocking more fine-grained understanding ability of Video-LLMs. Moreover, it relieves researchers from labor-intensive annotations and smartly circumvents the limitations of natural language, which often fails to capture the complex spatiotemporal variations in videos;(2) A novel benchmark dataset, namely FineVidBench, for rigorously assessing Video-LLMs' performance at both the scene and fragment levels, offering a comprehensive evaluation of their capabilities.We assessed multiple models and validated the effectiveness of SF^2T on them. Experimental results reveal that our approach improves their ability to capture and interpret spatiotemporal details.

The latest advancements in multi-modal large language models (MLLMs) have spurred a strong renewed interest in end-to-end motion planning approaches for autonomous driving. Many end-to-end approaches rely on human annotations to learn intermediate perception and prediction tasks, while purely self-supervised approaches--which directly learn from sensor inputs to generate planning trajectories without human annotations--often underperform the state of the art. We observe a key gap in the input representation space: end-to-end approaches built on MLLMs are often pretrained with reasoning tasks in 2D image space rather than the native 3D space in which autonomous vehicles plan. To this end, we propose S4-Driver, a scalable self-supervised motion planning algorithm with spatio-temporal visual representation, based on the popular PaLI multimodal large language model. S4-Driver uses a novel sparse volume strategy to seamlessly transform the strong visual representation of MLLMs from perspective view to 3D space without the need to finetune the vision encoder. This representation aggregates multi-view and multi-frame visual inputs and enables better prediction of planning trajectories in 3D space. To validate our method, we run experiments on both nuScenes and Waymo Open Motion Dataset (with in-house camera data). Results show that S4-Driver performs favorably against existing supervised multi-task approaches while requiring no human annotations. It also demonstrates great scalability when pretrained on large volumes of unannotated driving logs.

Self-supervised image denoising methods have garnered significant research attention in recent years, for this kind of method reduces the requirement of large training datasets.Compared to supervised methods, self-supervised methods rely more on the prior embedded in deep networks themselves. As a result, most of the self-supervised methods are designed with Convolution Neural Networks(CNNs) architectures, which well capture one of the most important image prior, translation equivariant prior. Inspired by the great success achieved by the introduction of translational equivariance, in this paper, we explore the way to further incorporate another important image prior. Specifically, we first apply high-accuracy rotation equivariant convolution to self-supervised image denoising. Through rigorous theoretical analysis, we have proved that simply replacing all the convolution layers with rotation equivariant convolution layers would modify the network into its rotation equivariant version.To the best of our knowledge, this is the first time that rotation equivariant image prior is introduced to self-supervised image denoising at the network architecture level with a comprehensive theoretical analysis of equivariance errors, whichoffers a new perspective to the field of self-supervised image denoising.Moreover, to further improve the performance, we design a new mask mechanism to fusion the output of rotation equivariant network and vanilla CNN-based network, and construct an adaptive rotation equivariant framework. Through extensive experiments on three typical methods, we have demonstrated the effectiveness of the proposed method.

Transformers significantly raise the performance limits across various tasks, spurring research into integrating them into spiking neural networks. However, a notable performance gap remains between existing spiking Transformers and their artificial neural network counterparts. Here, we first analyze the cause of this gap and attribute it to the dot product's ineffectiveness in measuring similarity between spiking queries and keys, due to numerous non-spiking events. To address this, we propose a novel a-XNOR similarity measure tailored for spike trains. It redefines the correlation between non-spike pairs as a specific value a, effectively overcoming the limitations of dot-product similarity. Furthermore, considering the sparse nature of spike trains where spikes carry more information than non-spikes, the a-XNOR similarity correspondingly highlights the distinct importance of spikes over non-spikes. Extensive experiments demonstrate that a-XNOR similarity significantly improves performance across different spiking Transformer architectures on various static and neuromorphic datasets, further revealing the potential of spiking Transformers.

Mid-level vision capabilities -- such as generic object localization and 3D geometric understanding -- are not only fundamental to human vision but are also crucial for many real-world applications of computer vision. These abilities emerge with minimal supervision during the early stages of human visual development. Despite their significance, current self-supervised learning (SSL) approaches are primarily designed and evaluated for high-level recognition tasks, leaving their mid-level vision capabilities largely unexamined. In this study, we introduce a suite of benchmark protocols to systematically assess mid-level vision capabilities and present a comprehensive, controlled evaluation of 22 prominent SSL models across 8 mid-level vision tasks. Our experiments reveal a weak correlation between mid-level and high-level task performance. We also identify several SSL methods with highly imbalanced performance across mid-level and high-level capabilities, as well as some that excel in both. Additionally, we investigate key factors contributing to mid-level vision performance, such as pretraining objectives and network architectures. Our study provides a holistic and timely view of what SSL models have learned, complementing existing research that primarily focuses on high-level vision tasks. We hope our findings guide future SSL research to benchmark models not only on high-level vision tasks but on mid-level as well.

Image denoising enhances image quality, serving as a foundational technique across various computational photography applications. The obstacle to clean image acquisition in real scenarios necessitates the development of self-supervised image denoising methods only depending on noisy images, especially a single noisy image. Existing self-supervised image denoising paradigms (Noise2Noise and Noise2Void) rely heavily on information-lossy operations, such as downsampling and masking, culminating in low-quality denoising performance.In this paper, we propose a novel self-supervised single image denoising paradigm, Positive2Negative, to break the information-lossy barrier.Our paradigm involves two key steps: Renoised Data Construction (RDC) and Denoised Consistency Supervision (DCS). RDC renoises the predicted denoised image by the predicted noise to construct multiple noisy images, preserving all the information of the original image. DCS ensures consistency across the multiple denoised images, supervising the network to learn robust denoising. Our Positive2Negative paradigm achieves state-of-the-art performance in self-supervised single image denoising with significant speed improvements. The code is released to the public at https://github.com/Li-Tong-621/P2N.

Point-Supervised Object Detection (PSOD) in a discriminative style has recently gained significant attention for its impressive detection performance and cost-effectiveness. However, accurately predicting high-quality pseudo-box labels for drone-view images, which often feature densely packed small objects, remains a challenge. This difficulty arises primarily from the limitation of rigid sampling strategies, which hinder the pseudo-box optimization process. To address this, we propose PointSR, an effective and robust point-supervised object detection framework with self-regularized sampling that integrates temporal and informative constraints throughout the pseudo-box generation process. Specifically, the framework comprises three key components: Temporal-Ensembling Encoder (TE Encoder), Coarse Pseudo-box Prediction, and Pseudo-box Refinement. The TE Encoder builds an anchor prototype library by aggregating temporal information for dynamic anchor adjustment. In Coarse Pseudo-box Prediction, anchors are refined using the prototype library, and a set of informative samples is collected for subsequent refinement. During Pseudo-box Refinement, these informative negative samples are used to suppress low-confidence candidate positive samples, thereby improving the quality of the pseudo-boxes. Experimental results on benchmark datasets demonstrate that PointSR significantly outperforms state-of-the-art methods, achieving up to 2.6%~\mathbf 7.2% higher AP_ 50 using only point supervision. Additionally, it exhibits strong robustness to perturbation in human-labeled points.

With the rapidly increasing demand for oriented object detection (OOD), recent research involving weakly-supervised detectors for learning OOD from point annotations has gained great attention. In this paper, we rethink this challenging task setting with the layout among instances and present Point2RBox-v2. At the core are three principles: 1) Gaussian overlap loss. It learns an upper bound for each instance by treating objects as 2D Gaussian distributions and minimizing their overlap. 2) Voronoi watershed loss. It learns a lower bound for each instance through watershed on Voronoi tessellation. 3) Consistency loss. It learns the size/rotation variation between two output sets with respect to an input image and its augmented view. Supplemented by a few devised techniques, e.g. edge loss and copy-paste, the detector is further enhanced. To our best knowledge, Point2RBox-v2 is the first approach to explore the spatial layout among instances for learning point-supervised OOD. Our solution is elegant and lightweight, yet it is expected to give a competitive performance especially in densely packed scenes: 62.61%/86.15%/34.71% on DOTA/HRSC/FAIR1M.

Point detection has been developed to locate pedestrians in crowded scenes by training a counter through a point-to-point (P2P) supervision scheme. Despite its excellent localization and counting performance, training a point-based counter still faces challenges concerning annotation labor: hundreds to thousands of points are required to annotate a single sample capturing a dense crowd. In this paper, we integrate point-based methods into a semi-supervised counting framework based on pseudo-labeling, enabling the training of a counter with only a few annotated samples supplemented by a large volume of pseudo-labeled data. However, during implementation, the training encounters issues as the confidence for pseudo-labels fails to be propagated to background pixels via the P2P. To tackle this challenge, we devise a point-specific activation map (PSAM) to visually interpret the phenomena occurring during the ill-posed training. Observations from the PSAM suggest that the feature map is excessively activated by the loss for unlabeled data, causing the decoder to misinterpret these over-activations as pedestrians. To mitigate this issue, we propose a point-to-region (P2R) scheme to substitute P2P, which segments out local regions rather than detects a point corresponding to a pedestrian for supervision. Consequently, pixels in the local region can share the same confidence with the corresponding pseudo points. Experimental results in both semi-supervised counting and unsupervised domain adaptation highlight the advantages of our method, illustrating P2R can resolve issues identified in PSAM. The code is available at https://github.com/Elin24/P2RLoss.

Self-supervised learning (SSL) on 3D point clouds has the potential to learn feature representations that can transfer to diverse sensors and multiple downstream perception tasks. However, recent SSL approaches fail to define pretext tasks that retain geometric information such as object pose and scale, which can be detrimental to the performance of downstream localization and geometry-sensitive 3D scene understanding tasks, such as 3D semantic segmentation and 3D object detection. We propose PSA-SSL, a novel extension to point cloud SSL that learns object pose and size-aware (PSA) features. Our approach defines a self-supervised bounding box regression pretext task, which retains object pose and size information. Furthermore, we incorporate LiDAR beam pattern augmentation on input point clouds, which encourages learning sensor-agnostic features. Our experiments demonstrate that with a single pretrained model, our light-weight yet effective extensions achieve significant improvements on 3D semantic segmentation with limited labels across popular autonomous driving datasets (Waymo, nuScenes, SemanticKITTI). Moreover, our approach outperforms other state-of-the-art SSL methods on 3D semantic segmentation (using up to 10 times less labels), as well as on 3D object detection. Our code will be released on https://github.com/TRAILab/PSA-SSL.

Event cameras provide low-latency perception for only milliwatts of power. This makes them highly suitable for resource-restricted, agile robots such as small flying drones. Self-supervised learning based on contrast maximization holds great potential for event-based robot vision, as it foregoes the need for high-frequency ground truth and allows for online learning in the robot's operational environment. However, online, on-board learning raises the major challenge of achieving sufficient computational efficiency for real-time learning, while maintaining competitive visual perception performance. In this work, we improve the time and memory efficiency of the contrast maximization pipeline, making on-device learning of low-latency monocular depth possible. We demonstrate that online learning on board a small drone yields more accurate depth estimates and more successful obstacle avoidance behavior compared to only pre-training. Benchmarking experiments show that the proposed pipeline is not only efficient, but also achieves state-of-the-art depth estimation performance among self-supervised approaches. Our work taps into the unused potential of online, on-device robot learning, promising smaller reality gaps and better performance.

Recently, virtual staining has emerged as a promising alternative to revolutionize histological staining by digitally generating stains. However, most existing methods suffer from the curse of staining unreality and unreliability. In this paper, we propose the Orthogonal Decoupling Alignment Generative Adversarial Network (ODA-GAN) for unpaired virtual immunohistochemistry (IHC) staining. Our approach is based on the assumption that an image consists of IHC staining-related features, which influence staining distribution and intensity, and staining-unrelated features, such as tissue morphology. Leveraging a pathology foundation model, we first develop a weakly-supervised segmentation pipeline as an alternative to expert annotations. We introduce an Orthogonal MLP (O-MLP) module to project image features into an orthogonal space, decoupling them into staining-related and unrelated components. Additionally, we propose a Dual-stream PatchNCE (DPNCE) loss to resolve contrastive learning contradictions in the staining-related space, thereby enhancing staining accuracy. To further improve realism, we introduce a Multi-layer Domain Alignment (MDA) module to bridge the domain gap between generated and real IHC images. Evaluations on three benchmark datasets show that our ODA-GAN reaches state-of-the-art (SOTA) performance. Our source code is available at https://github.com/ittong/ODA-GAN.