cs.CV（2026-07-02）

📊 共 54 篇论文 | 🔗 11 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (16 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (15 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (12 🔗3) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (3) 支柱四：生成式动作 (Generative Motion) (2) 支柱五：交互与反应 (Interaction & Reaction) (2) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱三：空间感知与语义 (Perception & Semantics) (16 篇)

#题目一句话要点标签🔗⭐
1 Consistent Scene Understanding in 3D Gaussian Splatting via Multi-Cue Mask Refinement 提出多线索掩膜优化框架以解决3D高斯点云一致性问题 3D gaussian splatting 3DGS gaussian splatting
2 Structure-Aware Gaussian Splatting for Large-Scale Scene Reconstruction 提出结构感知高斯点云以解决大规模场景重建问题 3D gaussian splatting gaussian splatting splatting ✅
3 VisionAId: An Offline-First Multimodal Android Assistant for People with Visual Impairment, Featuring Personalized Object Retrieval 提出VisionAId以解决视觉障碍者日常生活中的物体识别问题 depth estimation monocular depth metric depth
4 MVFusion-GS: Motion-Variance Guided Temporal Attention for High-Quality Dynamic Gaussian Splatting 提出MVFusion-GS以解决动态场景重建中的运动感知不足问题 3D gaussian splatting 3DGS gaussian splatting
5 The Turning Point of 3D Plant Phenotyping: 3D Foundation Models Enable Minute-to-Second Cross-Crop Reconstruction and Beyond 提出3D基础模型以解决3D植物表型高通量重建问题 3D gaussian splatting 3D reconstruction gaussian splatting
6 Mind the Gap: Standard 3DGS Evaluation Primarily Measures Near-Trajectory Interpolation 提出公平匹配计数协议以解决3D高斯点云评估中的插值问题 3D gaussian splatting 3DGS gaussian splatting
7 X-Splat: Gaussian Splatting for 3D CBCT Generation from Single Panoramic Radiograph 提出X-Splat框架以从单张全景X光片生成3D CBCT图像 gaussian splatting splatting NeRF ✅
8 FoundDP: Revisiting Weak Disparity Observability in Dual-Pixel Depth Estimation 提出FoundDP以解决双像素深度估计中的弱视差可观测性问题 depth estimation monocular depth metric depth ✅
9 ProCal: Inference-Time Proposal Calibration for Open-Vocabulary Object Detection 提出ProCal以解决开放词汇目标检测中的定位精度问题 open-vocabulary open vocabulary
10 Diversity-aware View Partitioning for Scalable VGGT 提出多样性感知视图分割以解决VGGT扩展问题 3D reconstruction VGGT spatial relationship
11 InvSplat: Inverse Feed-Forward Scene Splatting 提出Inverse Feed-Forward场景点云以解决逆渲染问题 3D reconstruction splatting ✅
12 Search-based Testing of Vision Language Models for In-Car Scene Understanding 提出ISU-Test以解决汽车场景理解中的测试问题 scene understanding
13 ICDepth: Taming Video Diffusion Models for Video Depth Estimation via In-Context Conditioning 提出ICDepth以解决单目视频深度估计中的时序一致性问题 depth estimation
14 QWERTY: Training-Free Motion Control via Query-Warped Video Diffusion Transformers 提出QWERTY框架以实现无训练的运动控制 optical flow latent optimization
15 DL-VINS-Factory: A Modular Framework for Learned Visual Front-Ends in Visual-Inertial SLAM 提出DL-VINS-Factory以解决视觉惯性SLAM中的特征提取问题 optical flow ✅
16 PointDiT: Pixel-Space Diffusion for Monocular Geometry Estimation 提出PointDiT以简化单图像3D重建方法 3D reconstruction

🔬 支柱二：RL算法与架构 (RL & Architecture) (15 篇)

#题目一句话要点标签🔗⭐
17 Geometric Foundation Model Distillation for Efficient Lunar 3D Reconstruction 提出几何基础模型蒸馏以解决月球3D重建效率问题 distillation 3D reconstruction foundation model
18 Bridging 3D Gaussians and Semantic Occupancy for Comprehensive Open-Vocabulary Scene Understanding from Unposed Images 提出COVScene以解决无姿态图像的全面3D场景理解问题 distillation scene understanding open-vocabulary
19 Teaching Vision-Language-Action Models What to See and Where to Look 提出DriveTeach-VLA以解决VLA模型空间依赖性不足问题 distillation vision-language-action VLA ✅
20 Multimodal Fusion for Fine-Grained Classification of Breast Fibroadenoma and Phyllodes Tumors 提出多模态融合方法以解决乳腺纤维腺瘤与叶状肿瘤的细粒度分类问题 representation learning multimodal
21 DetailAnywhere: Fashion Detail Generation via Cross-Modal Feature Alignment Distillation 提出DetailAnywhere以解决时尚细节生成问题 reinforcement learning distillation multimodal
22 PWM-ArtGen: Part World Model for Articulated Object Generation 提出PWM-ArtGen以解决关节物体生成中的运动结构预测问题 world model world models
23 Multi-Resolution Flow Matching: Training-Free Diffusion Acceleration via Staged Sampling 提出MrFlow以解决文本到图像扩散加速问题 flow matching distillation
24 Seek to Segment: Active Perception for Panoramic Referring Segmentation 提出PanoSeeker以解决主动全景指代分割问题 reinforcement learning embodied AI
25 Dual-Selective Network for Domain-Incremental Change Detection 提出双选择网络以解决领域增量变化检测问题 Mamba state space model distillation
26 FlowCIR: Semantic Transport via Flow Matching for Zero-Shot Composed Image Retrieval 提出FlowCIR以解决零-shot组合图像检索中的语义传输问题 flow matching
27 Efficient PEFT Methods with Adaptive Checkpointing for Vision Models and VLMs on Resource Constrained Consumer-GPUs 提出自适应检查点的高效PEFT方法以解决资源受限GPU的调优问题 Mamba foundation model
28 SFKD: Spatial--Frequency Joint-Aware Heterogeneous Knowledge Distillation via Multi-Level Wavelet Spectral Interaction 提出SFKD框架以解决异构模型知识蒸馏中的空间信息损失问题 distillation
29 RTE-FM-Dehazer: Radiative Transfer Equation Inspired Flow Matching for Real-World Image Dehazing 提出RTE-FM-Dehazer以解决图像去雾中的残余雾霾问题 flow matching
30 WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory 提出WorldDirector以解决动态物体记忆与视角控制问题 world model world models
31 Understanding Geometric Representations in Self-Supervised Vision Transformers via Subspace Intervention 提出子空间干预框架以研究自监督视觉变换器中的几何表示 masked autoencoder MAE ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (12 篇)

#题目一句话要点标签🔗⭐
32 LLM-Empowered Multimodal Fusion Framework for Autonomous Driving: Semantic Enhancement and Channel-Adaptive Design 提出LM-SCIP框架以解决动态输入质量影响的多模态融合问题 large language model multimodal
33 Learning Spectral and Polarimetric Clues for One-to-Multimodal Novel View Synthesis 提出SPoILeR以解决多模态新视图合成问题 multimodal
34 Descriptor: LYNRED Mobility Dataset Multimodal Detection Subset (LYNRED-MDS) 提出LYNRED-MDS数据集以解决低能见度条件下的早期碰撞预测问题 multimodal
35 MMBench-Live: A Continuously Evolving Benchmark for Multimodal Models 提出MMBench-Live以解决多模态基准评估的静态性问题 multimodal ✅
36 EAGLE-360: Embodied Active Global-to-Local Exploration in 360$^\circ$ 提出EAGLE-360以解决360°全景环境中的主动视觉搜索问题 large language model multimodal
37 ReQuest: Rethinking-based Question-Aware Frame Selection for Long-Form Video QA 提出ReQuest以解决长视频问答中的关键帧选择问题 large language model multimodal
38 Temporal and Cross-Modal Alignment for Enhanced Audiovisual Video Captioning 提出TCA-Captioner以解决视听视频字幕中的时间与跨模态对齐问题 large language model multimodal
39 Alignment Is All You Need For X-to-4D Generation 提出Align4D以解决X到4D生成的多模态转换问题 multimodal ✅
40 Combating Textual Noise and Redundancy: Entropy-Aware Dense Visual Token Pruning 提出熵感知密集视觉标记修剪以解决文本噪声问题 multimodal
41 AdaCount: Training-Free Similarity-Guided Spatial and Feature Adaptation for Zero-Shot Object Counting 提出AdaCount以解决零-shot目标计数中的实例分离问题 foundation model
42 SAB-LVLM: Significance-Aware Binarization for Large Vision-Language Models 提出SAB-LVLM以解决大规模视觉语言模型的二值化问题 multimodal ✅
43 LASER: A Corrective Lens for LVLMs via Visual Attention Preservation and Sink Suppression 提出LASER以解决长时间解码中的视觉遗忘问题 visual grounding

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
44 ComplexMimic: Human-Scene Interaction Imitation in Complex 3D Environments 提出ComplexMimic以解决复杂环境中的人场交互模仿问题 motion tracking imitation learning distillation ✅
45 VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment 提出VLAFlow框架以统一比较视觉-语言-动作模型的训练效果 manipulation flow matching vision-language-action
46 Online Segment 3D Gaussians via Launching Virtual Drones 提出SAGO框架以解决3D高斯在线分割问题 manipulation distillation 3D gaussian splatting

🔬 支柱六：视频提取与匹配 (Video Extraction) (3 篇)

#题目一句话要点标签🔗⭐
47 Multi-THuMBS: Multi-person Tracking of 3D Human Meshes Beyond Video Shots 提出Multi-THuMBS以解决多人物体3D网格追踪中的镜头切换问题 human mesh recovery motion reconstruction
48 HandsOnWorld: Unconstrained Egocentric Video Generation with Camera-Disentangled Hand Control 提出HandsOnWorld以解决无标记手控制视频生成问题 egocentric
49 Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs 提出SG-Ego和GLEN以解决人类活动理解问题 Ego4D embodied AI

🔬 支柱四：生成式动作 (Generative Motion) (2 篇)

#题目一句话要点标签🔗⭐
50 Training-free Controllable Human Motion Generation under Heterogeneous Constraints 提出MIC框架以解决异构约束下的无训练可控人类运动生成问题 motion generation human motion human motion generation
51 JointHOI: Jointly Generating Contact Maps Enhances Hand Object Interaction Generation 提出JointHOI以解决手部与物体交互生成中的接触问题 physically plausible penetration HOI

🔬 支柱五：交互与反应 (Interaction & Reaction) (2 篇)

#题目一句话要点标签🔗⭐
52 LongEgoRefer: A Benchmark for Long-Form Egocentric Video Referring Expression Comprehension 提出LongEgoRefer基准以解决长视频中的指代表达理解问题 human-object interaction egocentric Ego4D
53 InterCMDM: Block-Causal Diffusion for Autoregressive Human Interaction Generation 提出InterCMDM以解决人际互动生成中的因果性问题 two-person interaction

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
54 SAMoR: Motion Modelling for Articulated Objects of Any Skeleton and Topology 提出SAMoR以解决任意骨架拓扑的运动建模问题 motion representation

⬅️ 返回 cs.CV 首页 · 🏠 返回主页