CVPR 2026 Papers

Occluded visible-infrared person re-identification (Occluded VI-ReID) remains difficult due to modality heterogeneity and occlusions, both of which break structural consistency and weaken cross-modality feature alignment. Existing methods rely mainly on spatial-domain cues (such as local body parts and salient patches), but their discriminability degrades severely under varying imaging conditions or partial visibility. To address these issues, we introduce a spatial-frequency collaborative perspective that offers global perception and cross-location consistency. Specifically, we propose a Spatial-Frequency Collaborative Learning (SFCL) framework that uses frequency information to complement spatial representations. SFCL comprises a Cross-Modality Frequency Alignment Module (CFAM), a Spatial-Frequency Interaction Module (SFIM), and a Frequency-Aware Discriminative (FAD) loss. The CFAM models the spectral features of visible/infrared images in the frequency domain, establishing modality-consistent spectral priors. The SFIM injects these priors into spatial features, promoting dual-domain interaction and complementary representations of spatial and frequency semantics. In addition, the FAD loss jointly enforces cross-modality frequency alignment and semantic consistency, thus enhancing robustness and discriminability under occlusions. For real-occlusion evaluation, we construct two occluded datasets, Occ-SYSU-MM01 and Occ-RegDB, on which SFCL outperforms the state-of-the-art.

Bundle adjustment is a long-standing problem in computer vision that solves for camera parameters and 3D point coordinates from 2D image observations. While there has been much work on various aspects, like adaptation to different camera models and sensors, and considerations for solving the optimization problem, in this paper, we deal with a fundamental and distinct question of the uniqueness of its solution. In particular, we examine the unique solvability of the 3D reconstruction problem using parallel rigidity theory. We design an algorithm to ensure that the topology of the bipartite graph formed by the camera-3D point relations in bundle adjustment does not result in independent scaling of the edges in its subgraphs. To tackle the generally large-sized bipartite graph, we leverage camera-camera relationships in 3D reconstruction problems for efficiency. We demonstrate the benefits of our analysis for a global structure-from-motion pipeline. Applying our proposed algorithm results in significantly cleaner reconstructions by removing misplaced cameras and 3D points.

A key barrier to the real-world deployment of humanoid robots is the lack of autonomous loco-manipulation skills. We introduce VIRAL, a visual sim-to-real framework that learns humanoid loco-manipulation entirely in simulation and deploys it zero-shot to real hardware. VIRAL follows a teacher-student design: a privileged RL teacher, operating on full state, learns long-horizon loco-manipulation using a delta action space and reference state initialization. A vision-based student policy is then distilled from the teacher via large-scale simulation with tiled rendering, trained with a mixture of online DAgger and behavior cloning. We find that compute scale is critical: scaling simulation to tens of GPUs (up to 64) makes both teacher and student training reliable, while low-compute regimes often fail. To bridge the sim-to-real gap, VIRAL combines large-scale visual domain randomization--over lighting, materials, camera parameters, image quality, and sensor delays--with real-to-sim alignment of the dexterous hand and cameras. Deployed on a Unitree G1 humanoid, the resulting RGB-based policy performs continuous loco-manipulation for up to 54 cycles, generalizing to diverse spatial and appearance variations without any real-world fine-tuning, and approaching expert-level teleoperation performance. Extensive ablations dissect the key design choices required to make RGB-based humanoid loco-manipulation work in practice.

We introduce SketchDeco, a training-free approach to sketch colourisation that bridges the gap between professional design needs and intuitive, region-based control. Our method empowers artists to use simple masks and colour palettes for precise spatial and chromatic specification, avoiding both the tediousness of manual assignment and the ambiguity of text-based prompts. We reformulate this task as a novel, training-free composition problem. Our core technical contribution is a guided latent-space blending process: we first leverage diffusion inversion to precisely "paint" user-defined colours into specified regions, and then use a custom self-attention mechanism to harmoniously blend these local edits with a globally consistent base image. This ensures both local colour fidelity and global harmony without requiring any model fine-tuning. Our system produces high-quality results in 15--20 inference steps on consumer GPUs, making professional-quality, controllable colourisation accessible.

Weight regularization methods in continual learning (CL) alleviate catastrophic forgetting by assessing and penalizing changes to important model weights. Elastic Weight Consolidation (EWC) is a foundational and widely used approach within this framework that estimates weight importance based on gradients. However, it has consistently shown suboptimal performance.In this paper, we conduct a systematic analysis of importance estimation in EWC from a gradient-based perspective. For the first time, we find that EWC's reliance on the Fisher Information Matrix (FIM) results in gradient vanishing and inaccurate importance estimation in certain scenarios.Our analysis also reveals that Memory Aware Synapses (MAS), a variant of EWC, imposes unnecessary constraints on parameters irrelevant to prior tasks, termed the redundant protection. Consequently, both EWC and its variant exhibit fundamental misalignments in estimating the importance of weights, leading to inferior performance. To tackle these issues, we propose the Logits Reversal (LR) operation, a simple yet effective modification that rectifies the importance estimation of EWC. Specifically, reversing the logit values during the calculation of the FIM can effectively prevent both the gradient vanishing and the redundant protection. Extensive experiments across various CL tasks and datasets show that the proposed method significantly outperforms existing EWC and its variants. Therefore, we refer to it as EWC Done Right (EWC-DR).

Modern multimodal large language models (MLLMs) typically keep the language model fixed and train a visual projector that maps the pixels into a sequence of tokens in its embedding space, so that images can be presented in essentially the same form as text. However, the language model has been optimized to operate on discrete, semantically meaningful tokens, while prevailing visual projectors transform an image into a long stream of continuous and highly correlated embeddings. This causes the visual tokens to behave differently from the word-like units that LLMs are originally trained to understand. We propose a novel Disentangled Visual Tokenization (DiVT) that clusters patch embeddings into coherent semantic units, so each token corresponds to a distinct visual concept instead of a rigid grid cell. DiVT further adapts its token budget to image complexity, providing an explicit accuracy compute trade-off modifying neither the vision encoder nor the language model. Across diverse multimodal benchmarks, DiVT matches or surpasses baselines with significantly fewer visual tokens, demonstrating robustness under limited token budgets, significantly reducing memory cost and latency while making visual inputs more compatible with LLMs. Our code is available at https://github.com/snuviplab/DiVT.

Accurate 3D object detection in autonomous driving relies on effectively combining complementary information from multiple sensors. 4D millimeter-wave radar provides sparse yet physically reliable measurements, whose potential for enhancing sensor fusion has not been fully utilized. In this work, we propose Radar Prior Guided Fusion(RPGFusion), a practical 4D radar-camera fusion framework. We first generate radar prior maps that encode spatial confidence and depth cues. These priors guide image feature sampling while preventing the uneven BEV feature distribution (near-dense, far-sparse) caused by Lift-Splat-Shoot view transformation. To address the sparsity and noise inherent in point clouds, we adopt a hybrid robust encoding and sparse-to-dense feature propagation. We further introduce spatial alignment and semantic fusion modules to reconcile geometric and semantic differences between modalities, yielding more consistent and complementary BEV representations. Extensive experiments on the public View-of-Delft and TJ4DRadSet show that RPGFusion outperforms prior radar-camera fusion methods, achieving SOTA performance. Our work not only uses 4D radar signals to guide image BEV queries, but also enables robust radar feature encoding and densification for 3D perception, demonstrating the strong potential of 4D radar.

Image abstraction, a fundamental component of non-photorealistic rendering (NPR), aims to simplify photographs into stylized depictions while preserving perceptually important structures. A central difficulty is selectivity: removing fine textures while preserving semantically meaningful boundaries. Existing approaches often expose only a few entangled controls, so the amount of smoothing and the set of preserved structures cannot be adjusted in a clear and intuitive manner. We propose the Semantic Scale Space (SSS), a two-dimensional abstraction framework parameterized by smoothing strength and semantic granularity. SSS externalizes the stopping set by using a controllable semantic boundary detector to specify which structures act as barriers to smoothing, independently of how far simplification proceeds within homogeneous regions. We further introduce Adaptive Granularity Scheduling Smoothing (AGSS), a concrete traversal policy in this space that combines a donor-gated diffusion operator with a fine-to-coarse granularity schedule. To evaluate selectivity fairly, we also introduce an effect-matched protocol based on a Region Homogeneity Index, which compares methods under matched smoothing levels. On the SBD benchmark, AGSS achieves higher boundary preservation and lower geometric drift than strong baselines at the same degree of smoothing. On DIV2K, a user study further shows that downstream NPR results built on AGSS-based abstractions are consistently preferred. These results demonstrate that SSS and AGSS provide a practical framework and method for controllable image abstraction in modern creative applications.

Recent advances in video generation have shown remarkable potential for constructing world simulators. However, current models still struggle to produce physically consistent results, particularly when handling large-scale or complex dynamics. This limitation arises primarily because existing approaches respond isotropically to physical prompts and neglect the fine-grained alignment between generated content and localized physical cues. To address these challenges, we propose ProPhy, a Progressive Physical Alignment Framework that enables explicit physics-aware conditioning and anisotropic generation. ProPhy employs a two-stage Mixture-of-Physics-Experts (MoPE) mechanism for discriminative physical prior extraction, where Semantic Experts infer semantic-level physical principles from textual descriptions, and Refinement Experts capture token-level physical dynamics. This mechanism allows the model to learn fine-grained, physics-aware video representations that better reflect underlying physical laws. Furthermore, we introduce a physical alignment strategy that transfers the physical reasoning capabilities of vision-language models (VLMs) into the Refinement Experts, facilitating a more accurate representation of dynamic physical phenomena. Extensive experiments on physics-aware video generation benchmarks demonstrate that ProPhy produces more realistic, dynamic, and physically coherent results than existing state-of-the-art methods.

Autonomous driving needs fast, scalable 4D reconstruction and re-simulation for training and evaluation, yet most methods for dynamic driving scenes still rely on per-scene optimization, known camera calibration, or short frame windows, making them slow and impractical. We revisit this problem from a feedforward perspective and introduce Driving Gaussian Grounded Transformer (DGGT), a unified framework for pose-free dynamic scene reconstruction. We note that the existing formulations, treating camera pose as a required input, limit flexibility and scalability. Instead, we reformulate pose as an output of the model, enabling reconstruction directly from sparse, unposed images and supporting an arbitrary number of views for long sequences. Our approach jointly predicts per-frame 3D Gaussian maps and camera parameters, disentangles dynamics with a lightweight dynamic head, and preserves temporal consistency with a lifespan head that modulates visibility over time. A diffusion-based rendering refinement further reduces motion/interpolation artifacts and improves novel-view quality under sparse inputs. The result is a single-pass, pose-free algorithm that achieves state-of-the-art performance and speed. Trained and evaluated on large-scale driving benchmarks (Waymo, nuScenes, Argoverse2), our method outperforms prior work both when trained on each dataset and in zero-shot transfer across datasets, and it scales well as the number of input frames increases.

Latent diffusion models have emerged as the dominant framework for high-fidelity and efficient image generation, owing to their ability to learn diffusion processes in compact latent spaces. However, while previous research has focused primarily on reconstruction accuracy and semantic alignment of the latent space, we observe that another critical factor, robustness to sampling perturbations, also plays a crucial role in determining generation quality. Through empirical and theoretical analyses, we show that the commonly used b-VAE-based tokenizers in latent diffusion models, tend to produce overly compact latent manifolds that are highly sensitive to stochastic perturbations during diffusion sampling, leading to visual degradation. To address this issue, we propose a simple yet effective solution that constructs a latent space robust to sampling perturbations while maintaining strong reconstruction fidelity. This is achieved by introducing a Variance Expansion loss that counteracts variance collapse and leverages the adversarial interplay between reconstruction and variance expansion to achieve an adaptive balance that preserves reconstruction accuracy while improving robustness to stochastic sampling. Extensive experiments demonstrate that our approach consistently enhances generation quality across different latent diffusion architectures, confirming that robustness in latent space is a key missing ingredient for stable and faithful diffusion sampling. Our project page: https://github.com/CVL-UESTC/VE-Loss .

Training machine learning models on massive datasets is expensive and time-consuming. Dataset distillation addresses this by creating a small synthetic dataset that achieves the same performance as the full dataset. Recent methods use diffusion models to generate distilled data, either by promoting diversity or matching training gradients. However, existing approaches produce redundant training signals, where samples convey overlapping information. Empirically, disjoint subsets of distilled datasets capture 80-90% overlapping signals. This redundancy stems from optimizing visual diversity or average training dynamics without accounting for similarity across samples, leading to datasets where multiple samples share similar information rather than complementary knowledge. We propose learnability-driven dataset distillation, which constructs synthetic datasets incrementally through successive stages. Starting from a small set, we train a model and generate new samples guided by learnability scores that identify what the current model can learn from, creating an adaptive curriculum. We introduce Learnability-Guided Diffusion (LGD), which balances training utility for the current model with validity under a reference model to generate curriculum-aligned samples. Our approach reduces redundancy by 39.1%, promotes specialization across training stages, and achieves state-of-the-art results on ImageNet-1K (60.1%), ImageNette (87.2%), and ImageWoof (72.9%).

Vision-based end-to-end (E2E) driving has garnered interest in the research community due to its scalability and synergy with multimodal large language models (MLLMs). However, current E2E driving benchmarks primarily feature nominal scenarios paired with existing open-loop evaluation metrics that fall short in capturing the multimodal nature of driving or effectively evaluating performance in long-tail scenarios. To address these gaps, we introduce the Waymo Open Dataset for End-to-End Driving (WOD-E2E). WOD-E2E contains 4,021 driving segments (approximately 12 hours), specifically curated for challenging long-tail scenarios that are rare in daily life with an occurring frequency of less than 0.03%. Each segment in WOD-E2E includes the high-level routing information, ego states, and 360-degree camera views from 8 surrounding cameras. To evaluate E2E driving performance on these long-tail situations, we propose a novel open-loop evaluation metric: Rater Feedback Score (RFS). Unlike conventional distance-based metrics, RFS measures how closely a predicted trajectory matches rater-annotated trajectory preference labels. WOD-E2E includes rater preference labels for validation, and a separate held out test set is used for the 2025 WOD-E2E Challenge benchmark leaderboard.

Latent inpainting in diffusion models still relies almost universally on linearly interpolating VAE latents under a downsampled mask. We propose a key principle for compositing image latents: Pixel-Equivalent Latent Compositing (PELC). An equivalent latent compositor should be the same as compositing in pixel space. This principle enables full-resolution mask control and true soft-edge alpha compositing, even though VAEs compress images 8x spatially. Modern VAEs capture global context beyond patch-aligned local structure, so linear latent blending cannot be pixel-equivalent: it produces large artifacts at mask seams and global degradation and color shifts. We introduce DecFormer, a 7.7M-parameter transformer that predicts per-channel blend weights and an off-manifold residual correction to realize mask-consistent latent fusion. DecFormer is trained so that decoding after fusion matches pixel-space alpha compositing, is plug-compatible with existing diffusion pipelines, requires no backbone finetuning and adds only 0.07% of FLUX.1-Dev's parameters and 3.5% FLOP overhead. On the FLUX.1 family, DecFormer restores global color consistency, soft-mask support, sharp boundaries, and high-fidelity masking, reducing error metrics around edges by up to 53% over standard mask interpolation. Used as an inpainting prior, a lightweight LoRA on FLUX.1-Dev with DecFormer achieves fidelity comparable to FLUX.1-Fill, a fully finetuned inpainting model. While we focus on inpainting, PELC is a general recipe for pixel-equivalent latent editing, as we demonstrate on a complex color-correction task.

Although multi-modal large language models (MLLMs) have shown strong capabilities across diverse domains, their application in generating fine-grained 3D perception and prediction outputs in autonomous driving remains underexplored. In this paper, we propose DrivePI, a novel spatial-aware 4D MLLM that serves as a unified Vision-Language-Action (VLA) framework that is also compatible with vision-action (VA) models. Our method jointly performs spatial understanding, 3D perception (i.e., 3D occupancy), prediction (i.e., occupancy flow), and planning (i.e., action outputs) in parallel through end-to-end optimization. To obtain both precise geometric information and rich visual appearance, our approach integrates point clouds, multi-view images, and language instructions within a unified MLLM architecture. We further develop a data engine to generate text-occupancy and text-flow QA pairs for 4D spatial understanding. Remarkably, with only a 0.5B Qwen2.5 model as MLLM backbone, DrivePI as a single unified model matches or exceeds both existing VLA models and specialized VA models. Specifically, compared to VLA models, DrivePI outperforms OpenDriveVLA-7B by 2.5% mean accuracy on nuScenes-QA and reduces collision rate by 70% over ORION (from 0.37% to 0.11%) on nuScenes. Against specialized VA models, DrivePI surpasses FB-OCC by 10.3 RayIoU for 3D occupancy on OpenOcc, reduces the mAVE from 0.591 to 0.509 for occupancy flow on OpenOcc, and achieves 32% lower L2 error than VAD (from 0.72m to 0.49m) for planning on nuScenes. Code will be available at https://github.com/happinesslz/DrivePI.

Vision Transformers (ViTs) have achieved state-of-the-art results across various vision tasks. To enable a practical deployment of ViTs on modern hardware systems, post-training quantization (PTQ) has been actively studied in recent years. In particular, Hessian-based block reconstruction approaches have demonstrated promising results in quantizing ViT models to ultra-low bitwidths (e.g., 4-bit). However, finding a representative approximate Hessian, a fundamental step in recent approaches such as APHQ-ViT and FIMA-Q, remains underexplored in terms of the quantization-induced error and estimation cost. To address these shortcomings, we first reveal that the sample independence assumption used in recent works, which ignores the covariance term, can lead to a significant approximation error, especially for sub-four-bits. Inspired by least-squares regression, we propose LS-ViT, a block reconstruction framework that effectively estimates a representative Hessian by explicitly minimizing this approximation error across all samples. Extensive experiments with various ViT models across different vision tasks demonstrate that LS-ViT achieves new state-of-the-art performance. In addition, LS-ViT reduces quantization time compared to prior work, enabling a practical, plug-and-play, quantization-aware deployment for ViTs. The source code is available.

We present SGS-Intrinsic, an indoor inverse rendering framework that works well for sparse-view images. Unlike existing 3D Gaussian Splatting (3DGS) based methods that focus on object-centric reconstruction and fail to work under sparse view settings, our method allows to achieve high-quality geometry reconstruction and accurate disentanglement of material and illumination. The core idea is to construct a dense and geometry-consistent Gaussian semantic field guided by semantic and geometric priors, providing a reliable foundation for subsequent inverse rendering. Building upon this, we perform material-illumination disentanglement by combining a hybrid illumination model and material prior to effectively capture illumination-material interactions. To mitigate the impact of cast shadows and enhance the robustness of material recovery, we introduce illumination-invariant material constraint together with a deshadowing model. Extensive experiments on benchmark datasets show that our method consistently improves both reconstruction fidelity and inverse rendering quality over existing 3DGS-based inverse rendering approaches. Our code is available at https://github.com/GrumpySloths/SGS_Intrinsic.github.io.

Table recognition (TR) aims to transform table images into semi-structured representations such as HTML or Markdown.As a core component of document parsing, TR has long relied on supervised learning, with recent efforts dominated by fine-tuning vision-language models (VLMs) using labeled data.While VLMs have brought TR to the next level, pushing performance further demands large-scale labeled data that is costly to obtain.Consequently, although proprietary models have continuously pushed the performance boundary, open-source models, often trained with limited resources and, in practice, the only viable option for many due to privacy regulations, still lag far behind.To bridge this gap, we introduce TRivia, a self-supervised fine-tuning method that enables pretrained VLMs to learn TR directly from unlabeled table images in the wild. Built upon Group Relative Policy Optimization, TRivia automatically identifies unlabeled samples that most effectively facilitate learning and eliminates the need for human annotations through a question-answering-based reward mechanism. An attention-guided module generates diverse questions for each table image, and the ability to interpret the recognition results and answer them correctly provides feedback to optimize the TR model.This closed-loop process allows the TR model to autonomously learn to recognize, structure, and reason over tables without labeled data. Leveraging this pipeline, we present TRivia-3B, an open-sourced, compact, and state-of-the-art TR model that surpasses existing systems (e.g., Gemini 2.5 Pro, MinerU2.5) on three popular benchmarks.

Contrastive learning (CL) has become a powerful approach for learning representations from unlabeled images. However, existing CL methods focus predominantly on visual appearance features while neglecting topological characteristics (e.g., connectivity patterns, boundary configurations, cavity formations) that provide valuable cues for medical image analysis. To address this limitation, we propose a new topological CL framework (TopoCL) that explicitly exploits topological structures during contrastive learning for medical imaging. Specifically, we first introduce topology-aware augmentations that control topological perturbations using a relative bottleneck distance between persistence diagrams, preserving medically relevant topological properties while enabling controlled structural variations. We then design a Hierarchical Topology Encoder that captures topological features through self-attention and cross-attention mechanisms. Finally, we develop an adaptive mixture-of-experts (MoE) module to dynamically integrate visual and topological representations. TopoCL can be seamlessly integrated with existing CL methods. We evaluate TopoCL on five representative CL methods (SimCLR, MoCo-v3, BYOL, DINO, and Barlow Twins) and five diverse medical image classification datasets. The experimental results show that TopoCL achieves consistent improvements: an average gain of 3.26% in linear probe classification accuracy with strong statistical significance, verifying its effectiveness.