CVPR 2025 论文列表

Vision Transformers (ViTs) have been widely applied in various computer vision and vision-language tasks. To gain insights into their robustness in practical scenarios, transferable adversarial examples on ViTs have been extensively studied. A typical approach to improving adversarial transferability is by refining the surrogate model. However, existing work on ViTs has restricted their surrogate refinement to backward propagation. In this work, we instead focus on Forward Propagation Refinement (FPR) and specifically refine two key modules of ViTs: attention maps and token embeddings. For attention maps, we propose Attention Map Diversification (AMD), which diversifies certain attention maps and also implicitly imposes beneficial gradient vanishing during backward propagation. For token embeddings, we propose Momentum Token Embedding (MTE), which accumulates historical token embeddings to stabilize the forward updates in both the Attention and MLP blocks. We conduct extensive experiments with adversarial examples transferred from ViTs to various CNNs and ViTs, demonstrating that our FPR outperforms the current best (backward) surrogate refinement by up to 7.0% on average. We also validate its superiority against popular defenses and its compatibility with other transfer methods. Codes and appendix are available at https://github.com/RYC-98/FPR.

With the widespread deployment of Multimodal Large Language Models (MLLMs) for visual-reasoning tasks, improving their safety has become crucial. Recent research indicates that despite training-time safety alignment, these models remain vulnerable to jailbreak attacks--carefully crafted image-prompt pairs that compel the model to generate harmful content. In this work, we first highlight a critical safety gap, demonstrating that alignment achieved solely through safety training may be insufficient against jailbreak attacks. To address this vulnerability, we propose Immune, an inference-time defense framework that leverages a safe reward model during decoding to defend against jailbreak attacks. Additionally, we provide a mathematical characterization of Immune, offering insights on why it improves safety against jailbreak. Extensive evaluations on diverse jailbreak benchmarks using recent MLLMs reveal that Immune effectively enhances model safety while preserving the model's original capabilities. For instance, against text-based jailbreak attacks on LLaVA-1.6, Immune reduces the attack success rate by 57.82% and 16.78% compared to the base MLLM and state-of-the-art defense strategy, respectively.

High-performance Multimodal Large Language Models (MLLMs) rely heavily on data quality. This study introduces a novel data synthesis method, leveraging insights from contrastive learning and image difference captioning to enhance fine-grained image recognition in MLLMs. By analyzing object differences in detailed regions between similar images, we challenge the model to identify both matching and distinct components. Specifically, our method initially create pairs of similar images that highlight object variations. After that, we introduce a Difference Area Generator for object differences identifying, followed by a Difference Captions Generator for differences describing. The outcome is a high-quality dataset of "object replacement" samples, named Img-Diff, which can be expanded as needed due to its automation. We use the generated dataset to finetune state-of-the-art (SOTA) MLLMs such as InternVL2, yielding comprehensive improvements across numerous image difference and Visual Question Answering tasks. For instance, the trained models notably surpass the SOTA models GPT-4V and Gemini on the MMVP benchmark. Additionally, we conduct thorough evaluations to confirm the dataset's diversity, quality, and robustness, presenting several insights on the synthesis of such a contrastive dataset. We release our codes and dataset to encourage further research on multimodal data synthesis and MLLMs' fundamental capabilities for image understanding.

Adapting CLIP models for few-shot recognition has recently attracted significant attention. Despite considerable progress, these adaptations remain hindered by the pervasive challenge of data scarcity. Text-to-image models, capable of generating abundant photorealistic labeled images, offer a promising solution. However, existing approaches treat synthetic images merely as complements to real images, rather than as standalone knowledge repositories stemming from distinct foundation models. To overcome this limitation, we reconceptualize synthetic images as an *imagined base set*, i.e., a unique, large-scale synthetic dataset encompassing diverse concepts. We introduce a novel CLIP adaptation methodology called *ImagineFSL*, involving pretraining on the imagined base set followed by fine-tuning on downstream few-shot tasks. We find that, compared to no pretraining, both supervised and self-supervised pretraining are beneficial, with the latter providing better performance. Building on this finding, we propose an improved self-supervised method tailored for few-shot scenarios, enhancing the transferability of representations from synthetic to real image domains. Additionally, we present an image generation pipeline that employs chain-of-thought and in-context learning techniques, harnessing foundation models to automatically generate diverse, realistic images. Our methods are validated across eleven datasets, consistently outperforming state-of-the-art methods by substantial margins.

Readout multiplexing is a promising solution to overcome hardware limitations and data bottlenecks in imaging with single-photon detectors. Conventional multiplexed readout processing creates an upper bound on photon counts at a very fine time scale, where frames with multiple detected photons must either be discarded or allowed to introduce significant bias. We formulate multiphoton coincidence resolution as an inverse imaging problem and introduce a solution framework to probabilistically resolve the spatial locations of photon incidences. Specifically, we develop a theoretical abstraction of row--column multiplexing and a model of photon events that make readouts ambiguous. Using this, we propose a novel estimator that spatially resolves up to four coincident photons. Monte Carlo simulations show that our proposed method increases the peak signal-to-noise ratio (PSNR) of reconstruction by 3 to 4 dB compared to conventional methods under optimal incident flux conditions. Additionally, this method reduces the required number of readout frames to achieve the same mean-squared error as other methods by a factor of 4. Finally, our method matches the Cramer-Rao bound for detection probability estimation for a wider range of incident flux values compared to conventional methods. While demonstrated for a specific detector type and readout architecture, this method can be extended to more general multiplexing with different detector models.

User engagement is greatly enhanced by fully immersive multimodal experiences that combine visual and auditory stimuli. Consequently, the next frontier in VR/AR technologies lies in immersive volumetric videos with complete scene capture, large 6-DoF interactive space, Multi-modal feedback, and high resolution & frame-rate contents. To stimulate the reconstruction of immersive volumetric videos, we introduce ImViD, a multi-view, multi-modal dataset featuring complete space-oriented data capture and various indoor/outdoor scenarios. Our capture rig supports multi-view video-audio capture while on the move, a capability absent in existing datasets, which significantly enhances the completeness, flexibility, and efficiency of data capture. The captured multi-view videos (with synchronized audios) are in 5K resolution at 60FPS, lasting from 1-5 minutes, and include rich foreground-background elements, and complex dynamics. We benchmark existing methods using our dataset and establish a base pipeline for constructing immersive volumetric videos from multi-view audiovisual inputs for 6-DoF multimodal immersive VR experiences. The benchmark and the reconstruction and interaction results demonstrate the effectiveness of our dataset and baseline method, which we believe will stimulate future research on immersive volumetric video production.

The evolution of Large Vision-Language Models (LVLMs) has progressed from single-image understanding to multi-image reasoning. Despite this advancement, our findings indicate that LVLMs struggle to robustly utilize information across multiple images, with predictions significantly affected by the alteration of image positions. To further explore this issue, we introduce Position-wise Question Answering (PQA), a meticulously designed task to quantify reasoning capabilities at each position. Our analysis reveals a pronounced position bias in LVLMs: open-source models excel in reasoning with images positioned later but underperform with those in the middle or at the beginning, while proprietary models like GPT-4o show improved comprehension for images at the beginning and end but struggle with those in the middle. Motivated by these insights, we propose SoFt Attention (SoFA), a simple, training-free approach that mitigates this bias by employing linear interpolation between inter-image causal attention and bidirectional counterparts. Experimental results demonstrate that SoFA effectively reduces position bias and significantly enhances the reasoning performance of existing LVLMs.

This work introduces ILIAS, a new test dataset for Instance-Level Image retrieval At Scale. It is designed to evaluate the ability of current and future foundation models and retrieval techniques to recognize particular objects. The key benefits over existing datasets include large scale, domain diversity, accurate ground truth, and a performance that is far from saturated. ILIAS includes query and positive images for 1,000 object instances, manually collected to capture challenging conditions and diverse domains. Large-scale retrieval is conducted against 100 million distractor images from YFCC100M. To avoid false negatives without extra annotation effort, we include only query objects confirmed to have emerged after 2014, i.e. the compilation date of YFCC100M. An extensive benchmarking is performed with the following observations: i) models fine-tuned on specific domains, such as landmarks or products, excel in that domain but fail on ILIAS ii) learning a linear adaptation layer using multi-domain class supervision results in performance improvements, especially for vision-language models iii) local descriptors in retrieval re-ranking are still a key ingredient, especially in the presence of severe background clutter iv) the text-to-image performance of the vision-language foundation models is surprisingly close to the corresponding image-to-image case. website: https://vrg.fel.cvut.cz/ilias/

The ability to synthesize personalized group photos and specify the positions of each identity offers immense creative potential. While such imagery can be visually appealing, it presents significant challenges for existing technologies. A persistent issue is identity (ID) leakage, where injected facial features interfere with one another, resulting in low face resemblance, incorrect positioning, and visual artifacts. Existing methods suffer from limitations such as the reliance on segmentation models, increased runtime, or a high probability of ID leakage. To address these challenges, we propose ID-Patch, a novel method that provides robust association between identities and 2D positions. Our approach generates an ID patch and ID embeddings from the same facial features: the ID patch is positioned on the conditional image for precise spatial control, while the ID embeddings integrate with text embeddings to ensure high resemblance. Experimental results demonstrate that ID-Patch surpasses baseline methods across metrics, such as face ID resemblance, ID-position association accuracy, and generation efficiency. Project Page is: https://byteaigc.github.io/ID-Patch/

Despite the recent breakthroughs achieved by Large Vision Language Models (LVLMs) in understanding and responding to complex visual-textual contexts, their inherent hallucination tendencies limit their practical application in real-world scenarios that demand high levels of precision. Existing methods typically either fine-tune the LVLMs using additional data, which incurs extra costs in manual annotation and computational resources or perform comparisons at the decoding stage, which may eliminate useful language priors for reasoning while introducing inference time overhead. Therefore, we propose ICT, a lightweight, training-free method that calculates an intervention direction to shift the model's focus towards different levels of visual information, enhancing its attention to high-level and fine-grained visual details. During the forward pass stage, the intervention is applied to the attention heads that encode the overall image information and the fine-grained object details, effectively mitigating the phenomenon of overly language priors, and thereby alleviating hallucinations. Extensive experiments demonstrate that ICT achieves strong performance with a small amount of data and generalizes well across different datasets and models. Our code will be public.

The increasing adoption of large-scale models under 7 billion parameters in both language and vision domains enables inference tasks on a single consumer-grade GPU but makes fine-tuning models of this scale, especially 7B models, challenging. This limits the applicability of pruning methods that require full fine-tuning. Meanwhile, pruning methods that do not require fine-tuning perform well at low sparsity levels (10%-50%) but struggle at mid-to-high sparsity levels (50%-70%), where the error behaves equivalently to that of semi-structured pruning. To address these issues, this paper introduces ICP, which finds a balance between full fine-tuning and zero fine-tuning. First, Sparsity Rearrange is used to reorganize the predefined sparsity levels, followed by Block-wise Compensate Pruning, which alternates pruning and compensation on the model's backbone, fully utilizing inference results while avoiding full model fine-tuning. Experiments show that ICP improves performance at mid-to-high sparsity levels compared to baselines, with only a slight increase in pruning time and no additional peak memory overhead.

Recent advancements in computer vision have highlighted the scalability of Vision Transformers (ViTs) across various tasks, yet challenges remain in balancing adaptability, computational efficiency, and the ability to model higher-order relationships. Vision Graph Neural Networks (ViGs) offer an alternative by leveraging graph-based methodologies but are hindered by the computational bottlenecks of clustering algorithms used for edge generation. To address these issues, we propose the Hypergraph Vision Transformer (HgVT), which incorporates a hierarchical bipartite hypergraph structure into the vision transformer framework to capture higher-order semantic relationships while maintaining computational efficiency. HgVT leverages population and diversity regularization for dynamic hypergraph construction without clustering, and expert edge pooling to enhance semantic extraction and facilitate graph-based image retrieval. Empirical results demonstrate that HgVT achieves strong performance on image classification and retrieval, positioning it as an efficient framework for semantic-based vision tasks.

Uncertainty Quantification (UQ) is crucial for ensuring the reliability of machine learning models deployed in real-world autonomous systems. However, existing approaches typically quantify task-level output prediction uncertainty without considering epistemic uncertainty at the multimodal feature fusion level, leading to sub-optimal outcomes. Additionally, popular uncertainty quantification methods, e.g., Bayesian approximations, remain challenging to deploy in practice due to high computational costs in training and inference. In this paper, we propose HyperDUM, a novel deterministic uncertainty method (DUM) that efficiently quantifies feature-level epistemic uncertainty by leveraging hyperdimensional computing. Our method captures the channel and spatial uncertainties through channel and patch -wise projection and bundling techniques respectively. Multimodal sensor features are then adaptively weighted to mitigate uncertainty propagation and improve feature fusion. Our evaluations show that HyperDUM on average outperforms the state-of-the-art (SOTA) algorithms by up to 2.01%/1.27% in 3D Object Detection and up to 1.29% improvement over baselines in semantic segmentation tasks under various types of uncertainties. Notably, HyperDUM requires 2.36x less Floating Point Operations and up to 38.30x less parameters than SOTA methods, providing an efficient solution for real-world autonomous systems.

Addressing the retrieval of unsafe content from vision-language models such as CLIP is an important step towards real-world integration. Current efforts have relied on unlearning techniques that try to erase the model's knowledge of unsafe concepts. While effective in reducing unwanted outputs, unlearning limits the model's capacity to discern between safe and unsafe content. In this work, we introduce a novel approach that shifts from unlearning to an awareness paradigm by leveraging the inherent hierarchical properties of the hyperbolic space. We propose to encode safe and unsafe content as an entailment hierarchy, where both are placed in different regions of hyperbolic space. Our HySAC, Hyperbolic Safety-Aware CLIP, employs entailment loss functions to model the hierarchical and asymmetrical relations between safe and unsafe image-text pairs. This modelling - ineffective in standard vision-language models due to their reliance on Euclidean embeddings - endows the model with awareness of unsafe content, enabling it to serve as both a multimodal unsafe classifier and a flexible content retriever, with the option to dynamically redirect unsafe queries toward safer alternatives or retain the original output. Extensive experiments show that our approach not only enhances safety recognition, but also establishes a more adaptable and interpretable framework for content moderation in vision-language models. Our source code is available at: https://github.com/aimagelab/HySAC

This paper aims to address universal segmentation for image and video perception with the strong reasoning ability empowered by Visual Large Language Models (VLLMs). Despite significant progress in current unified segmentation methods, limitations in adaptation to both image and video scenarios, as well as the complex reasoning segmentation, make it difficult for them to handle various challenging instructions and achieve an accurate understanding of fine-grained vision-language correlations. We propose HyperSeg, the first VLLM-based universal segmentation model for pixel-level image and video perception, encompassing generic segmentation tasks and more complex reasoning perception tasks requiring powerful reasoning abilities and world knowledge. Besides, to fully leverage the recognition capabilities of VLLMs and the fine-grained visual information, HyperSeg incorporates hybrid entity recognition and fine-grained visual perceiver modules for various segmentation tasks. Combined with the temporal adapter, HyperSeg achieves a comprehensive understanding of temporal information. Experimental results validate the effectiveness of our insights in resolving universal image and video segmentation tasks, including the more complex reasoning perception tasks. Our code is available at https://github.com/congvvc/HyperSeg.

Decomposing a video into a layer-based representation is crucial for easy video editing for the creative industries, as it enables independent editing of specific layers. Existing video-layer decomposition models rely on implicit neural representations (INRs) trained independently for each video, making the process time-consuming when applied to new videos. Noticing this limitation, we propose a meta-learning strategy to learn a generic video decomposition model to speed up the training on new videos. Our model is based on a hypernetwork architecture which, given a video-encoder embedding, generates the parameters for a compact INR-based neural video decomposition model. Our strategy mitigates the problem of single-video overfitting and, importantly, shortens the convergence of video decomposition on new, unseen videos. Our code is available at: https://hypernvd.github.io/

Multimodal LLMs have advanced vision-language tasks but still struggle with understanding video scenes. To bridge this gap, Video Scene Graph Generation (VidSGG) has emerged to capture multi-object relationships across video frames. However, prior methods rely on pairwise connections, limiting their ability to handle complex multi-object interactions and reasoning. To this end, we propose Multimodal LLMs on a Scene HyperGraph (HyperGLM), promoting reasoning about multi-way interactions and higher-order relationships. Our approach uniquely integrates entity scene graphs, which capture spatial relationships between objects, with a procedural graph that models their causal transitions, forming a unified HyperGraph. Significantly, HyperGLM enables reasoning by injecting this unified HyperGraph into LLMs. Additionally, we introduce a new Video Scene Graph Reasoning (VSGR) dataset featuring 1.9M frames from third-person, egocentric, and drone views and supports five tasks: Scene Graph Generation, Scene Graph Anticipation, Video Question Answering, Video Captioning, and Relation Reasoning. Empirically, HyperGLM consistently outperforms state-of-the-art methods across five tasks, effectively modeling and reasoning complex relationships in diverse video scenes.

Advanced interpretation of hyperspectral remote sensing images benefits many precise Earth observation tasks. Recently, visual foundation models have promoted the remote sensing interpretation but concentrating on RGB and multispectral images. Due to the varied hyperspectral channels, existing foundation models would face image-by-image tuning situation, imposing great pressure on hardware and time resources. In this paper, we propose a tuning-free hyperspectral foundation model called HyperFree, by adapting the existing visual prompt engineering. To process varied channel numbers, we design a learned weight dictionary covering full-spectrum from 0.4 2.5um, supporting to build the embedding layer dynamically. To make the prompt design more tractable, HyperFree can generate multiple semantic-aware masks for one prompt by treating feature distance as semantic-similarity. After pre-training HyperFree on constructed large-scale high-resolution hyperspectral images, HyperFree (1 prompt) has shown comparable results with specialized models (5 shots) on 5 tasks and 11 datasets. Code and dataset would be accessible at https://rsidea.whu.edu.cn/hyperfree.htm#.

Mesh quality assessment (MQA) models play a critical role in the design, optimization, and evaluation of mesh operation systems in a wide variety of applications. Current MQA models, whether model-based methods using topology-aware features or projection-based approaches working on rendered 2D projections, often fail to capture the intricate interactions between texture and 3D geometry. We introduce HybridMQA, a first-of-its-kind hybrid full-reference colored MQA framework that integrates model-based and projection-based approaches, capturing complex interactions between textural information and 3D structures for enriched quality representations. Our method employs graph learning to extract detailed 3D representations, which are then projected to 2D using a novel feature rendering process that precisely aligns them with colored projections. This enables the exploration of geometry-texture interactions via cross-attention, producing comprehensive mesh quality representations. Extensive experiments demonstrate HybridMQA's superior performance across diverse datasets, highlighting its ability to effectively leverage geometry-texture interactions for a thorough understanding of mesh quality. Our project website is available at https://arshafiee.github.io/hybridmqa/.