cs.CV（2026-07-01）

📊 共 55 篇论文 | 🔗 13 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (19 🔗5) 支柱三：空间感知与语义 (Perception & Semantics) (14 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (12 🔗3) 支柱七：动作重定向 (Motion Retargeting) (4) 支柱一：机器人控制 (Robot Control) (4) 支柱六：视频提取与匹配 (Video Extraction) (2 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (19 篇)

#题目一句话要点标签🔗⭐
1 GeoSearcher: Anchor-Guided Progressive Reasoning for Remote Sensing Visual Grounding with Process Supervision 提出GeoSearcher以解决遥感视觉定位中的小目标识别问题 large language model multimodal visual grounding ✅
2 Multimodal Continuous Reasoning via Asymmetric Mutual Variational Learning 提出非对称互变分布学习以解决多模态推理瓶颈问题 large language model multimodal
3 Rosetta: Composable Native Multimodal Pretraining 提出Rosetta框架以解决多模态知识遗忘问题 foundation model multimodal ✅
4 Foundation Models vs. Radiomics for Lung Computed Tomography: A Benchmark of Feature Extractors, Classification Heads, and Segmentation Choices 基于基础模型与放射组学的肺部CT特征提取与分类比较研究 foundation model
5 QuaMoE-DRF: Proactive Beam and Rate Adaptation via Multimodal Dynamic Radio Map Forecasting in ISAC Networks 提出QuaMoE-DRF以解决ISAC网络中的主动波束和速率适应问题 multimodal
6 Retrieved Images as Visual Thought: Training-Free Multimodal In-Context Learning for the Open-vs-Closed Gap 提出ReVisIT框架以解决视觉推理中的生成依赖问题 multimodal
7 Closed-loop coupling of personalised and foundation models for real-time treatment guidance with MRI 提出闭环耦合框架以解决MRI实时治疗指导中的延迟问题 foundation model
8 Autonomous Scientific Discovery via Iterative Meta-Reflection 提出DiscoPER以解决自主科学发现中的开放性探索问题 large language model multimodal
9 Towards High-Resolution Visual Perception via Hierarchical Entity Exploration 提出层次实体探索以解决高分辨率视觉感知问题 large language model multimodal
10 Spotted: Location-informed Reidentification of Hyenas and Leopards in Camera Trap Surveys 提出Spotted框架以解决相机陷阱动物重识别问题 foundation model TAMP
11 DroneIQA-VLE: Multi-Task Drone Image Quality Assessment via Vision-Language Ensemble 提出DroneIQA-VLE以解决低空无人机图像质量评估问题 large language model multimodal ✅
12 Wake up for Touch! Mask-isolated Tactile Alignment Learning in MLLMs 提出Splash框架以解决多模态LLMs触觉对齐问题 multimodal
13 GenAU: Language-Grounded Industrial Anomaly Understanding with Vision-Language Models 提出GenAU以解决工业异常检测的多任务理解问题 instruction following
14 Stitched Embeddings: A Unified Latent Space for 3D Garments and 2D Patterns 提出Stitched Embeddings以解决3D服装与2D图案建模问题 foundation model ✅
15 SpiralFovea: Input-Adaptive Foveated Tokenization as a Third Lever of Resource-Adaptive Inference 提出SpiralFovea以解决固定网格输入适应性不足问题 foundation model
16 GaussianFusion: Unified 3D Gaussian Representation for Multi-Modal Fusion Perception 提出GaussianFusion以解决多模态融合感知中的细节损失问题 multimodal
17 AV-SyncBench: Decoupled Benchmarking of Temporal and Semantic Audio-Visual Synchronization 提出AV-SyncBench以解决音视频同步评估的偏差问题 multimodal ✅
18 StochasT: Learning with Stochastic Turn Depth for Visual Instruction Tuning 提出StochasT以解决视觉指令调优中的多轮训练问题 multimodal
19 MindAU: EEG-Conditioned Facial Action Unit Editing via Dual-Stream Manifold Alignment 提出MindAU以解决EEG引导的面部表情编辑问题 multimodal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (14 篇)

#题目一句话要点标签🔗⭐
20 GaussianEmoTalker: Real-Time Emotional Talking Head Synthesis with Audio-Driven and Blendshape-Based 3D Gaussian Splatting 提出GaussianEmoTalker以解决实时情感化虚拟人合成问题 3D gaussian splatting gaussian splatting splatting ✅
21 Privacy-Preserving Depth-Only Open-Vocabulary 3D Semantic Segmentation Via Uncertainty-Guided Test-Time Optimization 提出UTTO框架以解决隐私保护下的3D语义分割问题 scene understanding open-vocabulary open vocabulary
22 EO-VGGT: Orbital Ray-Conditioned 3D Foundation Models for Satellite Multi-View Reconstruction 提出EO-VGGT以解决卫星多视图重建中的几何不一致问题 3D reconstruction VGGT foundation model
23 Relation-Centric Open-Vocabulary 3D Gaussian Segmentation 提出PairGS框架以解决开放词汇3D高斯分割问题 open-vocabulary open vocabulary
24 Improving Sparse-View 3DGS Generalization via Flat Minima Optimization 提出平坦极小值优化以改善稀疏视图3D高斯点云泛化能力 3D gaussian splatting 3DGS gaussian splatting
25 Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation 提出FlexDepth以解决复杂驾驶环境中的单目深度估计问题 depth estimation monocular depth ✅
26 GADA: Geometry-Aware Deformable Aggregation for Image-Based Gaussian Splatting 提出几何感知可变聚合方法以解决高频细节失真问题 gaussian splatting splatting
27 EgoGapBench: Benchmarking Egocentric Action Selection in Multi-Agent Scenes 提出EgoGapBench以解决多智能体场景中的自我中心动作选择问题 scene understanding egocentric first-person view
28 LIST3R: Long-sequence Instance-aware 3D Reconstruction 提出LIST3R框架以解决长序列3D重建问题 3D reconstruction ✅
29 Active Spatial Guidance: Eliminating Injected Positional Mechanisms in Vision Transformers 提出主动空间引导以消除视觉变换器中的位置注入机制 depth estimation monocular depth ✅
30 Caption Bottleneck Models 提出Caption Bottleneck Models以解决概念集定义挑战 open-vocabulary open vocabulary
31 World from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video 提出动态高斯重建方法以解决单目视频生成问题 3DGS
32 Robust 3D Alignment of Generative Reconstructions via Partial Monocular Observations 提出一种无训练几何对齐框架以解决生成3D重建与单目观测对齐问题 3D reconstruction
33 CORGI: Consistency-Aware 3D Dog Reconstruction from a Single Image in the Wild 提出CORGI框架以解决单图像下的3D犬重建问题 3DGS

🔬 支柱二：RL算法与架构 (RL & Architecture) (12 篇)

#题目一句话要点标签🔗⭐
34 Cross4D-JEPA: Dense Cross-modal Correspondence Distillation for 4D Point Cloud Representation Learning 提出Cross4D-JEPA以解决动态4D点云表示学习问题 JEPA representation learning teacher-student
35 RetailSMV: Exocentric vs. Egocentric Adaptation of Foundation Video World Models in Retail 提出RetailSMV以优化零售场景下视频模型的适应性 world model world models egocentric
36 BrainFIBRE: A Foundation Model via Information Decomposition for Brain Microstructure 提出BrainFIBRE以解决脑微结构信息整合问题 representation learning foundation model multimodal
37 OmniView-Space: Reinforcing Spatial Reasoning via Multi-Perspective Spatial Mapping 提出OmniView-Space以解决多模态大语言模型的空间推理问题 distillation 3D reconstruction egocentric
38 Foundation Model-driven Key Anatomy Frame Selection for Blind-sweep Ultrasound Fetal Birth Weight Estimation 提出基础模型驱动的关键解剖帧选择框架以解决盲扫超声胎儿出生体重估计问题 MAE foundation model ✅
39 ClinRAG-GRAPH: Clinical-prior Retrieval-Augmented Graph Model with Domain Adversarial Learning for Breast pCR Prediction 提出ClinRAG-GRAPH以解决乳腺癌pCR预测中的多模态建模挑战 representation learning large language model multimodal ✅
40 Information-Regularized Attention for Visual-Centric Reasoning 提出信息正则化注意力机制以解决视觉语言模型不稳定问题 representation learning multimodal visual grounding
41 FrameONE: Hierarchical Motion Modeling for Universal Multi-View Echocardiographic Keyframe Detection 提出FrameONE以解决多视角超声心动图关键帧检测问题 representation learning motion representation ✅
42 Mirror-Fusion Attention for Reflection-Aware Self-Supervised Representation Learning 提出镜像融合注意力以解决自监督学习中的反射感知问题 representation learning MAE
43 Perceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning 提出Perceive-to-Reason框架以解决细粒度视觉推理问题 reinforcement learning multimodal
44 Attribute-Prompted Kernel Hashing for Unsupervised Data-Efficient Cross-Modal Retrieval 提出属性引导的核哈希方法以解决无监督跨模态检索问题 contrastive learning foundation model
45 OnPoint: Offline-to-Online Multi-Level Distillation for Point-Supervised Online Temporal Action Localization 提出OnPoint框架以解决点监督在线时序动作定位问题 distillation

🔬 支柱七：动作重定向 (Motion Retargeting) (4 篇)

#题目一句话要点标签🔗⭐
46 Learning When to Listen: Gated Affect Fusion for Human Motion Prediction 提出门控情感融合方法以改善人类动作预测精度 human motion human motion prediction motion prediction
47 MoHallBench: A Benchmark for Motion Hallucination in Video Large Language Models 提出MoHallBench以解决视频大语言模型中的运动幻觉问题 human motion large language model
48 EPO: Boosting 3D Foundation Models with Edge-based Pose Optimization 提出边缘基础姿态优化方法以提升3D基础模型的重建精度 motion reconstruction foundation model
49 Learning Cardiac Motion Priors for Implicit Neural Representations 提出心脏运动先验以优化隐式神经表示的适应性 motion estimation

🔬 支柱一：机器人控制 (Robot Control) (4 篇)

#题目一句话要点标签🔗⭐
50 ABot-M0.5: Unified Mobility-and-Manipulation World Action Model 提出ABot-M0.5以解决移动操控中的世界动作建模问题 manipulation mobile manipulation world model
51 DeWorldSG: Depth-Aware 3D Semantic Scene Graph Generation via World-Model Priors 提出DeWorldSG以解决3D场景图生成中的不稳定性问题 manipulation world model world models
52 Linguistic Relative Policy Optimization for Video Anomaly Reasoning 提出语言相对策略优化以解决视频异常检测问题 trajectory optimization large language model multimodal
53 GEAR-Seg: A Grounded Explainable Agent for Reasoning Segmentation and Data Engine 提出GEAR-Seg以解决推理分割的可解释性问题 manipulation distillation large language model

🔬 支柱六：视频提取与匹配 (Video Extraction) (2 篇)

#题目一句话要点标签🔗⭐
54 Imprint: Online Memory Compression for Long-Horizon Egocentric QA 提出Imprint以解决长时间跨度自我中心问答中的记忆压缩问题 egocentric
55 DART: Difficulty-Adaptive Routing for Zero-Shot Video Temporal Grounding 提出DART以解决零-shot视频时间定位中的推理差距问题 feature matching ✅

⬅️ 返回 cs.CV 首页 · 🏠 返回主页