cs.CV（2026-06-11）

📊 共 24 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (9 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (6 🔗1) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱七：动作重定向 (Motion Retargeting) (2 🔗1) 支柱八：物理动画 (Physics-based Animation) (2)

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
1 Masked and Predictive Self-Supervised Foundation Models for 3D Brain MRI 提出自监督基础模型以提升3D脑MRI疾病检测效果 JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
2 Avatar V: Scaling Video-Reference Avatar Video Generation 提出Avatar V以解决视频参考头像生成中的动态识别问题 flow matching RLHF distillation
3 VISA: VLM-Guided Instance Semantic Auditing for 3D Occupancy World Models 提出VISA以解决3D占用世界模型中的语义审计问题 world model world models
4 MoVerse: Real-Time Video World Modeling with Panoramic Gaussian Scaffold 提出MoVerse以解决实时视频世界建模问题 world model world models
5 ReFree: Towards Realistic Co-Speech Video Generation via Reward-Free RL and Multilevel Speech Guidance 提出ReFree-S2V以解决真实感对话视频生成问题 reinforcement learning flow matching character animation
6 Visual Place Recognition in Forests with Depth-Aware Distillation 提出深度感知蒸馏框架以解决森林环境中的视觉位置识别问题 distillation
7 OR-Action: Multi-Role Video Understanding with Fine-Grained Actions 提出OR-Action以解决手术室活动细粒度理解问题 distillation egocentric
8 Objects Before Words: Object-First Inductive Biases for Grounding Language in Child-View Video 提出BabyMind以解决儿童视角视频中的语言基础问题 contrastive learning egocentric ✅
9 MaskWAM: Unifying Mask Prompting and Prediction for World-Action Models 提出MaskWAM以解决WAM中的空间瓶颈问题 world action model world action models

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#题目一句话要点标签🔗⭐
10 LaME: Learning to Think in Latent Space for Multimodal Embedding via Information Bottleneck 提出LaME以解决多模态嵌入中的推理效率问题 multimodal chain-of-thought
11 How do Self-Supervised Remote Sensing Vision Models Transfer to Downstream Tasks? 研究自监督遥感视觉模型在下游任务中的迁移能力 foundation model multimodal
12 Cross-Modal Masked Compositional Concept Modeling for Enhancing Visio-Linguistic Compositionality 提出MACCO框架以解决视觉语言模型的组合理解问题 large language model multimodal ✅
13 Mirage Probes: How Vision Models Fake Visual Understanding 提出Mirage Probes以解决视觉语言模型的虚假理解问题 visual grounding
14 Temporal Backtracking Search for Test-time Generative Video Reasoning 提出时间回溯搜索以解决生成视频推理中的瓶颈问题 large language model
15 Perceive, Interact, Reason: Building Tool-Augmented Visual Agents for Spatial Reasoning 提出PERIA以解决空间推理任务中的工具增强问题 multimodal

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
16 RepWAM: World Action Modeling with Representation Visual-Action Tokenizers 提出RepWAM以解决现有世界动作模型的局限性 manipulation world action model world action models ✅
17 InterleaveThinker: Reinforcing Agentic Interleaved Generation 提出InterleaveThinker以解决图像生成中的交错生成问题 manipulation multimodal instruction following
18 Measurement-Calibrated Multi-Camera Fusion for Vision-Based Indoor Localization 提出测量校准的多摄像头融合方法以解决室内定位问题 motion tracking

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
19 SemanticXR: Low Power and Real-time Queryable Semantic Mapping with an Object-Level Device-Cloud Architecture 提出SemanticXR以解决移动XR设备的语义映射问题 semantic mapping semantic map open-vocabulary
20 Flex4DHuman: Flexible Multi-view Video Diffusion for 4D Human Reconstruction 提出Flex4DHuman以解决动态人类重建问题 gaussian splatting splatting 4DHumans

🔬 支柱七：动作重定向 (Motion Retargeting) (2 篇)

#题目一句话要点标签🔗⭐
21 Dual-Domain Equivariant Generative Adversarial Network for Multimodal CT-PET Synthesis 提出双域等变生成对抗网络以解决多模态CT-PET合成问题 geometric consistency multimodal
22 OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data 提出OmniDirector以解决多镜头视频生成中的相机运动克隆问题 motion representation multimodal ✅

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗⭐
23 HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers 提出HYDRA-X以解决多模态模型中图像与视频统一表示问题 spatiotemporal multimodal
24 CineOrchestra: Unified Entity-Centric Conditioning for Cinematic Video Generation 提出CineOrchestra以解决多维度视频生成控制问题 spatiotemporal

⬅️ 返回 cs.CV 首页 · 🏠 返回主页