CVPR 2026 论文列表

Raw images taken in low-light conditions are very noisy due to low photon count and sensor noise. Learning-based denoisers have the potential to reconstruct high-quality images. For training, however, these denoisers require large paired datasets of clean and noisy images, which are difficult to collect. Noise synthesis is an alternative to large-scale data acquisition: given a clean image, we can synthesize a realistic noisy counterpart. In this work, we propose a general and practical noise synthesis method that requires only one single noisy image and one single dark frame per ISO setting. We represent signal-dependent noise with a Poisson distribution and introduce a Fourier-domain spectral sampling algorithm to accurately model signal-independent noise. The latter generates diverse noise realizations that maintain the spatial and statistical properties of real sensor noise. As opposed to concurrent approaches, our method neither relies on simplified parametric models nor on large sets of clean-noisy image pairs. It is accurate and practical. Moreover, our synthesis method leads to state-of-the-art performances on multiple low-light denoising benchmarks.

3D pose transfer aims to transfer the pose-style of a source mesh to a target character while preserving both the target's geometry and the source's pose characteristic. Existing methods are largely restricted to characters with similar structures and fail to generalize to category-free settings (e.g., transferring a humanoid's pose to a quadruped). The key challenge lies in the structural and transformation diversity inherent in distinct character types, which often leads to mismatched regions and poor transfer quality. To address these issues, we first construct a million-scale pose dataset across hundreds of distinct characters. We further propose MimiCAT, a cascade-transformer model designed for category-free 3D pose transfer. Instead of relying on strict one-to-one correspondence mappings, MimiCAT leverages semantic keypoint labels to learn a novel soft correspondence that enables flexible many-to-many matching across characters. The pose transfer is then formulated as a conditional generation process, in which the source transformations are first projected onto the target through soft correspondence matching and subsequently refined using shape-conditioned representations. Extensive qualitative and quantitative experiments demonstrate that MimiCAT generalizes plausible poses across diverse character morphologies, surpassing prior approaches restricted to narrow-category transfer (e.g., humanoid-to-humanoid).

Visual Speaker Authentication (VSA) verifies identity by analyzing lip dynamics during prompted speech, offering enhanced privacy compared to full-face methods while maintaining discriminability for high-security applications. However, recent advances in talking face generation (TFG) have enabled realistic forgeries that closely mimic lip dynamics in sync with speech, posing severe threats to VSA systems. Prevailing defenses rely heavily on supervised classifiers trained on known forgeries via empirical risk minimization, resulting in poor generalization to unseen attacks, dependency on continuously updated fake data, and complete failure in the absence of effective forgery priors. In this paper, we revisit the design of forgery detectors and argue that over-reliance on fake priors hinders the exploitation of rich authenticity signals inherently present in real videos. We propose a novel detector trained exclusively on authentic data, learning forgery-aware representations through three key components: (1) lightweight modules that capture forgery-indicative statistics from real videos; (2) an asymmetric contrastive objective that compacts real samples while repelling potential forgeries in representation space; and (3) a theoretically grounded regularizer that shapes real representations into a tractable, isotropic Gaussian. To support rigorous evaluation, we introduce a benchmark suite spanning diverse TFG forgeries. Across eight modern forgery attacks and ten state-of-the-art (SOTA) detectors, our method achieves over a 10% reduction in error rates while preserving identity-verification capability with minimal overhead, and demonstrates consistent gains on datasets that better emulate real-world scenarios.

Vision-Language-Action (VLA) models are promising for embodied intelligence, yet they often overlook the predictive and temporal-causal structure underlying visual dynamics. World-model VLAs address this by predicting future frames, but waste capacity reconstructing redundant backgrounds. To overcome these limitations, we introduce CoWVLA, a new "Chain of World" paradigm that unifies world-model temporal reasoning with a disentangled latent motion representation. First, a pretrained video VAE serves as a latent motion extractor, explicitly factorizing video segments into structure and motion latents. Then, during pre-training, the VLA learns from an instruction and an initial frame to infer a continuous latent motion chain and predict the segment's terminal frame. Finally, during co-fine-tuning, this latent dynamic is aligned with discrete action prediction by jointly modeling sparse keyframes and action sequences in a unified autoregressive decoder. This design preserves the world-model benefits of temporal reasoning and world knowledge while retaining the compactness and interpretability of latent actions, enabling efficient visuomotor learning. Extensive experiments on robotic simulation benchmarks show that CoWVLA outperforms existing world-model and latent-action approaches and achieves moderate computational efficiency, highlighting its potential as a more effective VLA pretraining paradigm. Project website: https://fx-hit.github.io/cowvla-io.

Model fusion is a key strategy for robust recognition in unconstrained scenarios, as different models provide complementary strengths. This is especially important for whole-body human recognition, where biometric cues such as face, gait, and body shape vary across samples and are typically integrated via score-fusion. However, existing score-fusion strategies are usually static, invoking all models for every test sample regardless of sample quality or modality reliability. To overcome these limitations, we propose FusionAgent, a novel agentic framework that leverages a Multimodal Large Language Model (MLLM) to perform dynamic, sample-specific model selection. Each expert model is treated as a tool, and through Reinforcement Fine-Tuning (RFT) with a metric-based reward, the agent learns to adaptively determine the optimal model combination for each test input. To address the model score misalignment and embedding heterogeneity, we introduce Anchor-based Confidence Top-k (ACT) score-fusion, which anchors on the most confident model and integrates complementary predictions in a confidence-aware manner. Extensive experiments on multiple whole-body biometric benchmarks demonstrate that FusionAgent significantly outperforms SoTA methods while achieving higher efficiency through fewer model invocations, underscoring the critical role of dynamic, explainable, and robust model fusion in real-world recognition systems. Project page: https://fusionagent.github.io/.

Diffusion Transformers (DiTs) have greatly advanced text-to-image generation, but models still struggle to generate the correct spatial relations between objects as specified in the text prompt. In this study, we adopt a mechanistic interpretability approach to investigate how a DiT can generate correct spatial relations between objects. We train, from scratch, DiTs of different sizes with different text encoders to learn to generate images containing two objects whose attributes and spatial relations are specified in the text prompt. We find that, although all the models can learn this task to near-perfect accuracy, the underlying mechanisms differ drastically depending on the choice of text encoder. When using random text embeddings, we find that the spatial-relation information is passed to image tokens through a two-stage circuit, involving two cross-attention heads that separately read the spatial relation and single-object attributes in the text prompt. When using a pretrained text encoder (T5), we find that the DiT uses a different circuit that leverages information fusion in the text tokens, reading spatial-relation and single-object information together from a single text token. We further show that, although the in-domain performance is similar for the two settings, their robustness to out-of-domain perturbations differs, potentially suggesting the difficulty of generating correct relations in real-world scenarios.

The Segment Anything Model 2 (SAM2) has emerged as a foundation model for universal segmentation. Owing to its generalizable visual representations, SAM2 has been successfully applied to various downstream tasks. However, extending SAM2 to the RGB-D video salient object detection (RGB-D VSOD) task encounters three challenges including limited spatial modeling of linear LoRA, insufficient employment of SAM's multi-scale features, and dependence of initialization on explicit prompts. To address the issues, we present Multi-Modal Mixture-of-Experts with Memory-Augmented SAM (M4-SAM), which equips SAM2 with modality-related PEFT, hierarchical feature fusion, and prompt-free memory initialization. Firstly, we inject Modality-Aware MoE-LoRA, which employs convolutional experts to encode local spatial priors and introduces a modality dispatcher for efficient multi-modal fine-tuning, into SAM2's encoder. Secondly, we deploy Gated Multi-Level Feature Fusion, which hierarchically aggregates multi-scale encoder features with an adaptive gating mechanism, to balance spatial details and semantic context. Finally, to conduct zero-shot VSOD without manual prompts, we utilize a Pseudo-Guided Initialization, where a coarse mask is regarded as a pseudo prior and used to bootstrap the memory bank. Extensive experiments demonstrate that M4-SAM achieves the state-of-the-art performance across all evaluation metrics on three public RGB-D VSOD datasets.

Most Gaussian Splatting techniques that provide a 3D semantic representation of the scene don't optimize the underlying 3D geometry of the scene. This makes object-level editing or asset extraction challenging. Recent methods, like COBGS, Trace3D, and ObjectGS, acknowledge this limitation and propose approaches that modify the geometry of the scene to represent the underlying semantics. We go a step further and propose a novel solution that provides near perfect boundaries in object extraction. We do so by introducing two new losses in the optimization that take care of: 1. Modifying the geometry of visible Gaussians to respect semantic boundaries, and, 2. Modifying the geometry of non-visible Gaussians that appear once the object is extracted. Our first loss propagates gradients directly through the rasterization to allow for seamless integration within the optimization of the Gaussian parameters. Our second loss also propagates gradients to the Gaussian parameters, but does so without passing through the rasterization. This allows it to modify the geometry of the scene, even if not much transmittance arrives to a Gaussian (partial or non-visible). Exhaustive comparisons to 12 state of the art methods over 4 datasets, using six metrics, demonstrate that our approach produces overall the best boundary segmentation to date.

Video Novel View Synthesis (VNVS) aims to render arbitrary novel viewpoints of dynamic scenes from a single-view video, but its algorithmic training faces a major challenge: the lack of large-scale multi-view video datasets. Prior methods often train on monocular data by framing it as an inpainting task, which typically leads to a training-inference gap and visual artifacts. While synthetic multi-view data can partially alleviate the data scarcity issue, its high acquisition costs and limited diversity restrict scalability. To address these problems, we propose Scaling4D, a novel strategy that theoretically bridges the training-inference gap while leveraging large-scale monocular videos for training. Specifically, we take a higher-level perspective on the problem, reformulating VNVS into a general correspondence-guided generation task. Furthermore, in conjunction with extensive real-world data, we establish a synthetic data pipeline integrated with our training strategy to enhance precision. Qualitative and quantitative results demonstrate a positive correlation between performance and training data volume, confirming the scalability.

Aligning Diffusion models has achieved remarkable breakthroughs in generating high-quality, human preference-aligned images. Existing techniques, such as supervised fine-tuning (SFT) and DPO-style preference optimization, have become principled tools for fine-tuning diffusion models. However, SFT relies on high-quality images that are costly to obtain, while DPO-style methods depend on large-scale preference datasets, which are often inconsistent in quality. Beyond data dependency, these methods are further constrained by computational inefficiency. To address these two challenges, we propose Composite Reward Assisted Fine-Tuning (CRAFT), a lightweight yet powerful fine-tuning paradigm that requires significantly reduced training data while maintaining computational efficiency. It first leverages a Composite Reward Filtering (CRF) technique to construct a high-quality and consistent training dataset and then perform an enhanced variant of SFT. We also theoretically prove that CRAFT actually optimizes the lower bound of group-based reinforcement learning, establishing a principled connection between SFT with selected data and reinforcement learning. Our extensive empirical results demonstrate that CRAFT with only 100 samples can easily outperform recent SOTA preference optimization methods with thousands of preference-paired samples. Moreover, CRAFT can even achieve 11-220xfaster convergences than the baseline preference optimization methods, highlighting its extremely high efficiency.

Pre-trained video models learn powerful priors for generating high-quality, temporally coherent content. While these models excel at temporal coherence, their dynamics are often constrained by the continuous nature of their training data. We hypothesize that by injecting the rich and unconstrained content diversity from image data into this coherent temporal framework, we can generate image sets that feature both natural transitions and a far more expansive dynamic range. To this end, we introduce iMontage, a unified framework designed to repurpose a powerful video model into an all-in-one image generator. The framework consumes and produces variable-length image sets, unifying a wide array of image generation and editing tasks. To achieve this, we propose an elegant and minimally invasive adaptation strategy, complemented by a tailored data curation process and training paradigm. This approach allows the model to acquire broad image manipulation capabilities without corrupting its invaluable original motion priors. iMontage excels across several mainstream many-in-many-out tasks, not only maintaining strong cross-image contextual consistency but also generating scenes with extraordinary dynamics that surpass conventional scopes. Our code and model weights will be made publicly available.

Text-to-Image (T2I) models have achieved significant progress in generating high-quality images, with compositional visual generation emerging as an important capability that enables them to synthesize coherent, natural scenes from multiple discrete concepts. However, this powerful compositionality, while enhancing creativity, also introduces new safety risks: combinations of different concepts can produce high-risk images without explicitly expressing harmful content. Motivated by this, we propose CoRA (Composable Reassembly Attack): an attack method that preserves the original semantics while bypassing safety filters. Unlike traditional compositional generation approaches that rely on modifying the sampling process, CoRA operates solely in the text space under a black-box setting, iteratively rewriting and guiding prompts through interactive steps. Specifically, CoRA decomposes a potentially harmful intent into a set of fine-grained, superficially benign but semantically complete visual elements, and then uses iterative selection and reassembly to guide the target T2I model to recombine these elements without triggering safety checks, thereby recovering the original malicious semantics. Experimental results show that CoRA significantly improves attack success rates, producing higher-risk outputs while maintaining semantic consistency. Warning: This paper contains offensive or disturbing content.

Recent advancements in Unified Multimodal Models (UMMs) have enabled remarkable image understanding and generation capabilities. However, while models like Gemini-2.5-Flash-Image show emerging abilities to reason over multiple related images, existing benchmarks rarely address the challenges of multi-image context generation, focusing mainly on text-to-image or single-image editing tasks. In this work, we introduce MICON-Bench, a comprehensive benchmark covering six tasks that evaluate cross-image composition, contextual reasoning, and identity preservation. We further propose an MLLM-driven Evaluation-by-Checkpoint framework for automatic verification of semantic and visual consistency, where multimodal large language model (MLLM) serves as a verifier. Additionally, we present Dynamic Attention Rebalancing (DAR), a training-free, plug-and-play mechanism that dynamically adjusts attention during inference to enhance coherence and reduce hallucinations. Extensive experiments on various state-of-the-art open-source models demonstrate both the rigor of MICON-Bench in exposing multi-image reasoning challenges and the efficacy of DAR in improving generation quality and cross-image coherence.

Few-shot learning aims to enable rapid adaptation to unseen tasks using limited data. Optimization-based meta-learning addresses this challenge by acquiring shared prior knowledge across diverse tasks. However, its effectiveness degrades in cross-domain scenarios where unseen tasks differ significantly from training tasks. We identify this degradation as a failure to acquire generalizable prior knowledge, which is fundamentally caused by gradient discrepancies--conflicting update directions arising in the meta-training environment with diverse task distributions. To achieve robust few-shot generalization, we propose Data-Centric Meta-Learning (DCML), a novel framework that mitigates gradient discrepancies by aligning task-specific input distributions with shared prior knowledge. DCML accomplishes this alignment through a meta-learnable visual prompt that is integrated into the entire meta-learning process--unlike previous prompt-based methods restricted solely to test-time adaptation. During meta-training, the prompt transforms each task's inputs to induce more consistent gradients, thereby facilitating the learning of generalizable prior knowledge. Leveraging this robust knowledge, DCML enables rapid and parameter-efficient test-time adaptation by updating only the lightweight prompt and classifier while keeping the backbone frozen.Extensive experiments demonstrate that DCML consistently outperforms baselines, particularly in challenging few-shot cross-domain scenarios, establishing a data-centric perspective for robust meta-learning.

While neural video codecs (NVCs) have demonstrated superior compression ratio, their prohibitive computational complexity remains a critical barrier to real-world deployment. This paper introduces a chunk-based coding framework designed to significantly improve the rate-distortion-complexity trade-off. Instead of processing frames sequentially, our approach encodes a chunk of multiple frames into a single compact latent representation and decodes them simultaneously. This is enabled by cross-frame interaction modules for joint spatial-temporal modeling and frame-specific decoders for parallel reconstruction. This paradigm not only dramatically enhances coding throughput but also facilitates more effective modeling of long-term temporal correlations. To further boost speed, we propose a streamlined entropy coding mechanism that consolidates bit-stream interactions into a single step, substantially reducing decoding overhead. Building on these innovations, we present DCVC-UF (Ultra-Fast), a new NVC that sets a new SOTA in performance. Our experiments show that DCVC-UF can achieve ultra-fast encoding and decoding speeds, significantly outperforming previous leading codecs. DCVC-UF serves as a notable landmark in the journey of NVC evolution. The code is at https://github.com/microsoft/DCVC.

Despite significant progress in text-to-video generation, current models still suffer from unrealistic dynamics, temporal inconsistency, and unstable semantic alignment. Existing preference alignment approaches rely on costly and often ambiguous human or VLM-based video preference annotation, which has become a major bottleneck for scaling data. To address this challenge, we propose an annotation-free preference alignment method that constructs accurate preference pairs through video continuation.We extend a pretrained video generation model into a continuation model and apply continuation with different amounts of reference frames while keeping the total video length fixed. As generated segments are inferior to ground-truth frames and and fixed-length continuations conditioned on more reference frames contain less generated content, they exhibit higher fidelity than those with fewer references, naturally inducing a preference order.We further introduce Asymmetrical DPO, which computes preference loss on all continuation regions except the shared prefix conditioning frames and normalizes it by their length, preventing spurious preference signals from leaking into the conditioned portion.Experiments across multiple benchmarks show that our method delivers significant improvements in dynamics realism, temporal coherence, and semantic alignment over existing DPO-based approaches, while fully eliminating the need for human preference labeling or auxiliary reward models.

Event-based Action Recognition (EAR) provides a promising pathway for understanding dynamic behaviors under challenging conditions. Recent progress in vision-language models has introduced a cross-modal learning paradigm into EAR, enabling models to associate event streams with textual semantics for enhancing conceptual understanding. However, existing methods typically overlook the intrinsic polarity-driven motion cues that are fundamental to event data, leading to suboptimal spatiotemporal representations. To address this limitation, we propose a POlarity Knowledge Enhanced framework (POKER), which explicitly incorporates event polarity-aware motion knowledge across visual and textual modalities.POKER consists of two synergistic components: Polarity Motion Capturer (PMC) and Polarity Motion Reasoner (PMR). Specifically, PMC decouples positive and negative polarities to capture polarity-sensitive motion cues, while PMR semantically analyzes polarity-induced motion dynamics via large language models. Through the polarity alignment, POKER couples semantic reasoning with visual dynamics, achieving more discriminative representations. Extensive experiments on multiple benchmarks demonstrate that POKER enhances performance across diverse event representations.

We address the problem of tactile localization, where the goal is to identify image regions that share the same material properties as a tactile input. Existing visuo-tactile methods rely on global alignment and thus fail to capture the fine-grained local correspondences required for this task. The challenge is amplified by existing datasets, which predominantly contain close-up, low-diversity images. We propose a model that learns local visuo-tactile alignment via dense cross-modal feature interactions, producing tactile saliency maps for touch-conditioned material segmentation. To overcome dataset constraints, we introduce: (i) in-the-wild multi-material scene images that expand visual diversity, and (ii) a material-diversity pairing strategy that aligns each tactile sample with visually varied yet tactilely consistent images, improving contextual localization and robustness to weak signals. We also construct two new tactile-grounded material segmentation datasets for quantitative evaluation. Experiments on both new and existing benchmarks show that our approach substantially outperforms prior visuo-tactile methods in tactile localization.

Membership inference attacks (MIAs) aim to determine whether a specific data point was part of a model's training set, serving as effective tools for evaluating privacy leakage of vision models. However, existing MIAs implicitly assume honest query inputs, and their adversarial robustness remains unexplored. We show that MIAs for vision models expose a previously overlooked adversarial surface: adversarial membership manipulation, where imperceptible perturbations can reliably push non-member images into the member region of state-of-the-art MIAs. In this paper, we provide the first unified perspective on this phenomenon by analyzing its mechanism and implications. We begin by demonstrating that adversarial membership fabrication is consistently effective across diverse architectures and datasets. We then reveal a distinctive geometric signature--a characteristic gradient-norm collapse trajectory--that reliably separates fabricated from true members despite their nearly identical semantic representations. Building on this insight, we introduce a principled detection strategy grounded in gradient-geometry signals and develop a robust inference framework that substantially mitigates adversarial manipulation. Extensive experiments show that fabrication is broadly effective, while our detection and robust inference strategies significantly enhance resilience. This work establishes the first comprehensive framework for adversarial membership manipulation in vision models.