arXiv subjects

Computer vision

Explore arXiv computer-vision papers and cs.CV metadata. Search within the index for image segmentation, object detection and visual representation learning.

Follow this topic with RSS

At least 397 records · Page 22Linked to original sources

arXiv2025

TDM: Temporally-Consistent Diffusion Model for All-in-One Real-World Video Restoration

In this paper, we propose the first diffusion-based all-in-one video restoration method that utilizes the power of a pre-trained Stable Diffusion and a fine-tuned ControlNet. Our method can restore various types of video degradation with a single unified model, overcoming the limitation of standard methods that require specific models for each restoration task. Our contributions include an efficient training strategy with Task Prompt Guidance (TPG) for diverse restoration tasks, an inference strategy that combines Denoising Diffusion Implicit Models~(DDIM) inversion with a novel Sliding Window Cross-Frame Attention (SW-CFA) mechanism for enhanced content preservation and temporal consistency, and a scalable pipeline that makes our method all-in-one to adapt to different video restoration tasks. Through extensive experiments on five video restoration tasks, we demonstrate the superiority of our method in generalization capability to real-world videos and temporal consistency preservation over existing state-of-the-art methods. Our method advances the video restoration task by providing a unified solution that enhances video quality across multiple applications.

cs.CV↗

arXiv2026

SMDDFNet: State-space Modeling and Dynamic Dual Fusion Network for Traffic Sign Detection

Traffic sign detection is a challenging visual signal processing task for advanced driver assistance, where small objects, scale variation, and occlusion limit conventional detectors with fixed receptive fields. This paper proposes State-space Modeling and Dynamic Dual Fusion Network (SMDDFNet), a deep learning detector for traffic sign images. SMDDFNet integrates a Dynamic Dual Fusion (DDF) module and a state-space modeling backbone to enhance multi-scale feature representation. DDF combines efficient multi-scale attention with content-aware dynamic filtering in the frequency domain, while the backbone captures long-range dependencies with linear computational complexity. A multi-scale feature fusion neck further aggregates pyramid features for robust localization of small signs. Experiments on TT100K, GTSDB, PASCAL VOC, and the Roboflow~100 \emph{vehicle} subset show that SMDDFNet achieves competitive accuracy against recent detectors while retaining real-time throughput. The source code is available at https://github.com/rainbowyuyu/SMDDFNet

cs.CV↗

arXiv2025

Enhancing UAV Search under Occlusion using Next Best View Planning

Search and rescue missions are often critical following sudden natural disasters or in high-risk environmental situations. The most challenging search and rescue missions involve difficult-to-access terrains, such as dense forests with high occlusion. Deploying unmanned aerial vehicles for exploration can significantly enhance search effectiveness, facilitate access to challenging environments, and reduce search time. However, in dense forests, the effectiveness of unmanned aerial vehicles depends on their ability to capture clear views of the ground, necessitating a robust search strategy to optimize camera positioning and perspective. This work presents an optimized planning strategy and an efficient algorithm for the next best view problem in occluded environments. Two novel optimization heuristics, a geometry heuristic, and a visibility heuristic, are proposed to enhance search performance by selecting optimal camera viewpoints. Comparative evaluations in both simulated and real-world settings reveal that the visibility heuristic achieves greater performance, identifying over 90% of hidden objects in simulated forests and offering 10% better detection rates than the geometry heuristic. Additionally, real-world experiments demonstrate that the visibility heuristic provides better coverage under the canopy, highlighting its potential for improving search and rescue missions in occluded environments.

cs.RO↗

arXiv2025

Alias-free 4D Gaussian Splatting

Existing dynamic scene reconstruction methods based on Gaussian Splatting enable real-time rendering and generate realistic images. However, adjusting the camera's focal length or the distance between Gaussian primitives and the camera to modify rendering resolution often introduces strong artifacts, stemming from the frequency constraints of 4D Gaussians and Gaussian scale mismatch induced by the 2D dilated filter. To address this, we derive a maximum sampling frequency formulation for 4D Gaussian Splatting and introduce a 4D scale-adaptive filter and scale loss, which flexibly regulates the sampling frequency of 4D Gaussian Splatting. Our approach eliminates high-frequency artifacts under increased rendering frequencies while effectively reducing redundant Gaussians in multi-view video reconstruction. We validate the proposed method through monocular and multi-view video reconstruction experiments.Ours project page: https://4d-alias-free.github.io/4D-Alias-free/

cs.CV↗

arXiv2025

Synthetic Curriculum Reinforces Compositional Text-to-Image Generation

Text-to-Image (T2I) generation has long been an open problem, with compositional synthesis remaining particularly challenging. This task requires accurate rendering of complex scenes containing multiple objects that exhibit diverse attributes as well as intricate spatial and semantic relationships, demanding both precise object placement and coherent inter-object interactions. In this paper, we propose a novel compositional curriculum reinforcement learning framework named CompGen that addresses compositional weakness in existing T2I models. Specifically, we leverage scene graphs to establish a novel difficulty criterion for compositional ability and develop a corresponding adaptive Markov Chain Monte Carlo graph sampling algorithm. This difficulty-aware approach enables the synthesis of training curriculum data that progressively optimize T2I models through reinforcement learning. We integrate our curriculum learning approach into Group Relative Policy Optimization (GRPO) and investigate different curriculum scheduling strategies. Our experiments reveal that CompGen exhibits distinct scaling curves under different curriculum scheduling strategies, with easy-to-hard and Gaussian sampling strategies yielding superior scaling performance compared to random sampling. Extensive experiments demonstrate that CompGen significantly enhances compositional generation capabilities for both diffusion-based and auto-regressive T2I models, highlighting its effectiveness in improving the compositional T2I generation systems.

cs.CV↗

arXiv2025

RNN as Linear Transformer: A Closer Investigation into Representational Potentials of Visual Mamba Models

Mamba has recently garnered attention as an effective backbone for vision tasks. However, its underlying mechanism in visual domains remains poorly understood. In this work, we systematically investigate Mamba's representational properties and make three primary contributions. First, we theoretically analyze Mamba's relationship to Softmax and Linear Attention, confirming that it can be viewed as a low-rank approximation of Softmax Attention and thereby bridging the representational gap between Softmax and Linear forms. Second, we introduce a novel binary segmentation metric for activation map evaluation, extending qualitative assessments to a quantitative measure that demonstrates Mamba's capacity to model long-range dependencies. Third, by leveraging DINO for self-supervised pretraining, we obtain clearer activation maps than those produced by standard supervised approaches, highlighting Mamba's potential for interpretability. Notably, our model also achieves a 78.5 percent linear probing accuracy on ImageNet, underscoring its strong performance. We hope this work can provide valuable insights for future investigations of Mamba-based vision architectures.

cs.CV↗

arXiv2025

ViMix-14M: A Curated Multi-Source Video-Text Dataset with Long-Form, High-Quality Captions and Crawl-Free Access

Text-to-video generation has surged in interest since Sora, yet open-source models still face a data bottleneck: there is no large, high-quality, easily obtainable video-text corpus. Existing public datasets typically require manual YouTube crawling, which yields low usable volume due to link rot and access limits, and raises licensing uncertainty. This work addresses this challenge by introducing ViMix-14M, a curated multi-source video-text dataset of around 14 million pairs that provides crawl-free, download-ready access and long-form, high-quality captions tightly aligned to video. ViMix-14M is built by merging diverse open video sources, followed by unified de-duplication and quality filtering, and a multi-granularity, ground-truth-guided re-captioning pipeline that refines descriptions to better match actions, scenes, and temporal structure. We evaluate the dataset by multimodal retrieval, text-to-video generation, and video question answering tasks, observing consistent improvements over counterpart datasets. We hope this work can help removing the key barrier to training and fine-tuning open-source video foundation models, and provide insights of building high-quality and generalizable video-text datasets.

cs.CV↗

arXiv2025

Can a Second-View Image Be a Language? Geometric and Semantic Cross-Modal Reasoning for X-ray Prohibited Item Detection

Automatic X-ray prohibited items detection is vital for security inspection and has been widely studied. Traditional methods rely on visual modality, often struggling with complex threats. While recent studies incorporate language to guide single-view images, human inspectors typically use dual-view images in practice. This raises the question: can the second view provide constraints similar to a language modality? In this work, we introduce DualXrayBench, the first comprehensive benchmark for X-ray inspection that includes multiple views and modalities. It supports eight tasks designed to test cross-view reasoning. In DualXrayBench, we introduce a caption corpus consisting of 45,613 dual-view image pairs across 12 categories with corresponding captions. Building upon these data, we propose the Geometric (cross-view)-Semantic (cross-modality) Reasoner (GSR), a multimodal model that jointly learns correspondences between cross-view geometry and cross-modal semantics, treating the second-view images as a "language-like modality". To enable this, we construct the GSXray dataset, with structured Chain-of-Thought sequences: , , . Comprehensive evaluations on DualXrayBench demonstrate that GSR achieves significant improvements across all X-ray tasks, offering a new perspective for real-world X-ray inspection.

cs.CV↗

arXiv2025

SegSplat: Feed-forward Gaussian Splatting and Open-Set Semantic Segmentation

We have introduced SegSplat, a novel framework designed to bridge the gap between rapid, feed-forward 3D reconstruction and rich, open-vocabulary semantic understanding. By constructing a compact semantic memory bank from multi-view 2D foundation model features and predicting discrete semantic indices alongside geometric and appearance attributes for each 3D Gaussian in a single pass, SegSplat efficiently imbues scenes with queryable semantics. Our experiments demonstrate that SegSplat achieves geometric fidelity comparable to state-of-the-art feed-forward 3D Gaussian Splatting methods while simultaneously enabling robust open-set semantic segmentation, crucially \textit{without} requiring any per-scene optimization for semantic feature integration. This work represents a significant step towards practical, on-the-fly generation of semantically aware 3D environments, vital for advancing robotic interaction, augmented reality, and other intelligent systems.

cs.CV↗

arXiv2025

Exploring Weak-to-Strong Generalization for CLIP-based Classification

Aligning large-scale commercial models with user intent is crucial to preventing harmful outputs. Current methods rely on human supervision but become impractical as model complexity increases. When models surpass human knowledge, providing accurate feedback becomes challenging and inefficient. A novel solution proposed recently is using a weaker model to supervise a stronger model. This concept leverages the ability of weaker models to perform evaluations, thereby reducing the workload on human supervisors. Previous work has shown the effectiveness of weak-to-strong generalization in the context of language-only models. Extending this concept to vision-language models leverages these insights, adapting the proven benefits to a multi-modal context. In our study, we explore weak-to-strong generalization for CLIP-based classification. We propose a method, class prototype learning (CPL), which aims to enhance the classification capabilities of the CLIP model, by learning more representative prototypes for each category. Our findings indicate that, despite using a simple loss function under weak supervision, CPL yields robust improvements in targeted scenarios, particularly when pretraining is limited. Extensive experiments demonstrate that our approach is effective under these settings, achieving a 3.67% improvement over strong baseline methods.

cs.CV↗

arXiv2025

ChineseVideoBench: Benchmarking Multi-modal Large Models for Chinese Video Question Answering

This paper introduces ChineseVideoBench, a pioneering benchmark specifically designed for evaluating Multimodal Large Language Models (MLLMs) in Chinese Video Question Answering. The growing demand for sophisticated video analysis capabilities highlights the critical need for comprehensive, culturally-aware evaluation frameworks. ChineseVideoBench addresses this gap by providing a robust dataset and tailored evaluation metrics, enabling rigorous assessment of state-of-the-art MLLMs on complex Chinese video content. Specifically, ChineseVideoBench comprises 8 main classes and 12 sub-classes, encompassing tasks that demand both deep video understanding and nuanced Chinese linguistic and cultural awareness. Our empirical evaluations reveal that ChineseVideoBench presents a significant challenge to current MLLMs. Among the models assessed, Gemini 2.5 Pro achieves the highest performance with an overall score of 77.9%, while InternVL-38B emerges as the most competitive open-source model.

cs.CV↗

arXiv2025

4D-VGGT: A General Foundation Model with SpatioTemporal Awareness for Dynamic Scene Geometry Estimation

We investigate a challenging task of dynamic scene geometry estimation, which requires representing both spatial and temporal features. Typically, existing methods align the two features into a unified latent space to model scene geometry. However, this unified paradigm suffers from potential mismatched representation due to the heterogeneous nature between spatial and temporal features. In this work, we propose 4D-VGGT, a general foundation model with divide-and-conquer spatiotemporal representation for dynamic scene geometry. Our model is divided into three aspects: 1) Multi-setting input. We design an adaptive visual grid that supports input sequences with arbitrary numbers of views and time steps. 2) Multi-level representation. We propose a cross-view global fusion for spatial representation and a cross-time local fusion for temporal representation. 3) Multi-task prediction. We append multiple task-specific heads to spatiotemporal representations, enabling a comprehensive visual geometry estimation for dynamic scenes. Under this unified framework, these components enhance the feature discriminability and application universality of our model for dynamic scenes. In addition, we integrate multiple geometry datasets to train our model and conduct extensive experiments to verify the effectiveness of our method across various tasks on multiple dynamic scene geometry benchmarks.

cs.CV↗

arXiv2025

NeuroVascU-Net: A Unified Multi-Scale and Cross-Domain Adaptive Feature Fusion U-Net for Precise 3D Segmentation of Brain Vessels in Contrast-Enhanced T1 MRI

Precise 3D segmentation of cerebral vasculature from T1-weighted contrast-enhanced (T1CE) MRI is crucial for safe neurosurgical planning. Manual delineation is time-consuming and prone to inter-observer variability, while current automated methods often trade accuracy for computational cost, limiting clinical use. We present NeuroVascU-Net, the first deep learning architecture specifically designed to segment cerebrovascular structures directly from clinically standard T1CE MRI in neuro-oncology patients, addressing a gap in prior work dominated by TOF-MRA-based approaches. NeuroVascU-Net builds on a dilated U-Net and integrates two specialized modules: a Multi-Scale Contextual Feature Fusion ($MSC^2F$) module at the bottleneck and a Cross-Domain Adaptive Feature Fusion ($CDA^2F$) module at deeper hierarchical layers. $MSC^2F$ captures both local and global information via multi-scale dilated convolutions, while $CDA^2F$ dynamically integrates domain-specific features, enhancing representation while keeping computation low. The model was trained and validated on a curated dataset of T1CE scans from 137 brain tumor biopsy patients, annotated by a board-certified functional neurosurgeon. NeuroVascU-Net achieved a Dice score of 0.8609 and precision of 0.8841, accurately segmenting both major and fine vascular structures. Notably, it requires only 12.4M parameters, significantly fewer than transformer-based models such as Swin U-NetR. This balance of accuracy and efficiency positions NeuroVascU-Net as a practical solution for computer-assisted neurosurgical planning.

cs.CV↗

arXiv2025

CrossJEPA: Cross-Modal Joint-Embedding Predictive Architecture for Efficient 3D Representation Learning from 2D Images

Image-to-point cross-modal learning has emerged to address the scarcity of large-scale 3D datasets in 3D representation learning. However, current methods that leverage 2D data often result in large, slow-to-train models, making them computationally expensive and difficult to deploy in resource-constrained environments. The architecture design of such models is therefore critical, determining their performance, memory footprint, and compute efficiency. The Joint-embedding Predictive Architecture (JEPA) has gained wide popularity in self-supervised learning for its simplicity and efficiency, but has been under-explored in cross-modal settings, partly due to the misconception that masking is intrinsic to JEPA. In this light, we propose CrossJEPA, a simple Cross-modal Joint Embedding Predictive Architecture that harnesses the knowledge of an image foundation model and trains a predictor to infer embeddings of specific rendered 2D views from corresponding 3D point clouds, thereby introducing a JEPA-style pretraining strategy beyond masking. By conditioning the predictor on cross-domain projection information, CrossJEPA purifies the supervision signal from semantics exclusive to the target domain. We further exploit the frozen teacher design with a one-time target embedding caching mechanism, yielding amortized efficiency. CrossJEPA achieves a new state-of-the-art in linear probing on the synthetic ModelNet40 (94.2%) and the real-world ScanObjectNN (88.3%) benchmarks, using only 14.1M pretraining parameters (8.5M in the point encoder), and about 6 pretraining hours on a standard single GPU. These results position CrossJEPA as a performant, memory-efficient, and fast-to-train framework for 3D representation learning via knowledge distillation. We analyze CrossJEPA intuitively, theoretically, and empirically, and extensively ablate our design choices. Code will be made available.

cs.CV↗

arXiv2025

DocPTBench: Benchmarking End-to-End Photographed Document Parsing and Translation

The advent of Multimodal Large Language Models (MLLMs) has unlocked the potential for end-to-end document parsing and translation. However, prevailing benchmarks such as OmniDocBench and DITrans are dominated by pristine scanned or digital-born documents, and thus fail to adequately represent the intricate challenges of real-world capture conditions, such as geometric distortions and photometric variations. To fill this gap, we introduce DocPTBench, a comprehensive benchmark specifically designed for Photographed Document Parsing and Translation. DocPTBench comprises over 1,300 high-resolution photographed documents from multiple domains, includes eight translation scenarios, and provides meticulously human-verified annotations for both parsing and translation. Our experiments demonstrate that transitioning from digital-born to photographed documents results in a substantial performance decline: popular MLLMs exhibit an average accuracy drop of 18% in end-to-end parsing and 12% in translation, while specialized document parsing models show significant average decrease of 25%. This substantial performance gap underscores the unique challenges posed by documents captured in real-world conditions and reveals the limited robustness of existing models. Dataset and code are available at https://github.com/Topdu/DocPTBench.

cs.CV↗

arXiv2025

Perceptual-Evidence Anchored Reinforced Learning for Multimodal Reasoning

Reinforcement Learning with Verifiable Rewards (RLVR) has significantly advanced the reasoning capabilities of Large Language Models (LLMs) and is now being applied to Vision-Language Models (VLMs). However, vanilla RLVR for VLMs verifies only the final textual output, critically neglecting the foundational step of visual perception. This oversight leads to visual hallucinations and reward hacking, as reasoning built upon flawed perception is inherently unreliable. To address this, we propose PEARL (Perceptual-Evidence Anchored Reinforced Learning), a dual-branch, perception-reasoning synergistic that strengthens multimodal reasoning by explicitly anchoring it to verified visual evidence. For each reasoning-oriented QA instance, PEARL first derive a perception checklist -- a set of perception-oriented sub-questions with verifiable answers that probe the model's understanding of key visual evidence. During training, auxiliary rollouts on this checklist yield a perceptual reward that both directly reinforces the model's perception ability and acts as a fidelity gate for reasoning. If the model passes the perception check, its policy update is biased towards evidence-anchored reasoning. Otherwise, the process is halted to prevent reasoning from flawed premises. PEARL can be seamlessly integrated with popular RL methods like GRPO and DAPO. Comprehensive experiments show PEARL achieves substantial gains on multimodal reasoning benchmarks, e.g., a +9.7% improvement over the baseline and +6.6% over GRPO on MathVerse.

cs.CV↗

arXiv2025

ReCoGS: Real-time ReColoring for Gaussian Splatting scenes

Gaussian Splatting has emerged as a leading method for novel view synthesis, offering superior training efficiency and real-time inference compared to NeRF approaches, while still delivering high-quality reconstructions. Beyond view synthesis, this 3D representation has also been explored for editing tasks. Many existing methods leverage 2D diffusion models to generate multi-view datasets for training, but they often suffer from limitations such as view inconsistencies, lack of fine-grained control, and high computational demand. In this work, we focus specifically on the editing task of recoloring. We introduce a user-friendly pipeline that enables precise selection and recoloring of regions within a pre-trained Gaussian Splatting scene. To demonstrate the real-time performance of our method, we also present an interactive tool that allows users to experiment with the pipeline in practice. Code is available at https://github.com/loryruta/recogs.

cs.CV↗

arXiv2025

EventBench: Towards Comprehensive Benchmarking of Event-based MLLMs

Multimodal large language models (MLLMs) have made significant advancements in event-based vision, yet the comprehensive evaluation of their capabilities within a unified benchmark remains largely unexplored. In this work, we introduce EventBench, a benchmark that offers eight diverse task metrics together with a large-scale event stream dataset. EventBench differs from existing event-based benchmarks in four key aspects: (1) openness in accessibility, releasing all raw event streams and task instructions across eight evaluation metrics; (2) diversity in task coverage, spanning understanding, recognition, and spatial reasoning tasks for comprehensive capability assessment; (3) integration in spatial dimensions, pioneering the design of 3D spatial reasoning tasks for event-based MLLMs; and (4) scale in data volume, with an accompanying training set of over one million event-text pairs supporting large-scale training and evaluation. Using EventBench, we evaluate state-of-the-art closed-source models such as GPT-5 and Gemini-2.5 Pro, leading open-source models including Qwen2.5-VL and InternVL3, and event-based MLLMs such as EventGPT that directly process raw event streams. Extensive evaluation reveals that while current event-based MLLMs demonstrate strong performance in event stream understanding, they continue to struggle with fine-grained recognition and spatial reasoning.

cs.CV↗