cs.CV（2025-06-13）

📊 共 29 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (11 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (9 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (3 🔗1) 支柱七：动作重定向 (Motion Retargeting) (2) 支柱八：物理动画 (Physics-based Animation) (2) 支柱六：视频提取与匹配 (Video Extraction) (1 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (11 篇)

#题目一句话要点标签🔗⭐
1 Dynamic Mixture of Curriculum LoRA Experts for Continual Multimodal Instruction Tuning 提出动态混合课程LoRA专家以解决持续多模态指令调优问题 large language model multimodal
2 DaMO: A Data-Efficient Multimodal Orchestrator for Temporal Reasoning with Video LLMs 提出DaMO以解决视频语言模型中的时序推理问题 large language model multimodal
3 TAViS: Text-bridged Audio-Visual Segmentation with Foundation Models 提出TAViS以解决音视频分割中的跨模态对齐问题 foundation model multimodal
4 VGR: Visual Grounded Reasoning 提出VGR以解决多模态推理中的语言偏见问题 large language model multimodal chain-of-thought
5 Exploring the Effectiveness of Deep Features from Domain-Specific Foundation Models in Retinal Image Synthesis 提出基于深度特征的损失函数以改进视网膜图像合成 foundation model
6 VFaith: Do Large Multimodal Models Really Reason on Seen Images Rather than Previous Memories? 提出VFaith以评估多模态模型的视觉推理能力 multimodal
7 Enhance Multimodal Consistency and Coherence for Text-Image Plan Generation 提出多模态一致性与连贯性增强框架以解决文本-图像计划生成问题 multimodal ✅
8 Manager: Aggregating Insights from Unimodal Experts in Two-Tower VLMs and MLLMs 提出Manager插件以解决两塔VLMs和MLLMs中的单模态专家聚合问题 large language model multimodal ✅
9 CLIP Meets Diffusion: A Synergistic Approach to Anomaly Detection 提出CLIPFUSION以解决异常检测中的多模态融合问题 foundation model
10 Quizzard@INOVA Challenge 2025 -- Track A: Plug-and-Play Technique in Interleaved Multi-Image Model 提出LLaVA-NeXT-Interleave以解决多图像推理问题 foundation model ✅
11 A$^2$LC: Active and Automated Label Correction for Semantic Segmentation 提出A$^2$LC框架以解决语义分割中的标签纠错问题 foundation model

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

#题目一句话要点标签🔗⭐
12 MambaVSR: Content-Aware Scanning State Space Model for Video Super-Resolution 提出MambaVSR以解决视频超分辨率中的非局部依赖建模问题 Mamba SSM state space model
13 How Visual Representations Map to Language Feature Space in Multimodal LLMs 提出冻结模型与线性适配器以解决视觉与语言对齐问题 representation learning large language model multimodal
14 InceptionMamba: Efficient Multi-Stage Feature Enhancement with Selective State Space Model for Microscopic Medical Image Segmentation 提出InceptionMamba以解决显微医学图像分割效率问题 Mamba state space model
15 AgentSense: Virtual Sensor Data Generation Using LLM Agents in Simulated Home Environments 提出AgentSense以解决智能家居中缺乏多样化标注数据的问题 world model embodied AI large language model ✅
16 Stop learning it all to mitigate visual hallucination, Focus on the hallucination target 提出偏好学习方法以缓解多模态大语言模型的视觉幻觉问题 preference learning large language model multimodal
17 Self-supervised Learning of Echocardiographic Video Representations via Online Cluster Distillation 提出DISCOVR以解决心脏超声视频表示学习问题 representation learning distillation ✅
18 DAVID-XR1: Detecting AI-Generated Videos with Explainable Reasoning 提出DAVID-XR1以解决AI生成视频检测的可解释性问题 distillation chain-of-thought
19 EasyARC: Evaluating Vision Language Models on True Visual Reasoning 提出EasyARC以解决多模态视觉推理评估问题 reinforcement learning multimodal
20 Auto-Connect: Connectivity-Preserving RigFormer with Direct Preference Optimization 提出Auto-Connect以解决自动绑定中骨骼连通性问题 direct preference optimization

🔬 支柱三：空间感知与语义 (Perception & Semantics) (3 篇)

#题目一句话要点标签🔗⭐
21 GraphGSOcc: Semantic-Geometric Graph Transformer with Dynamic-Static Decoupling for 3D Gaussian Splatting-based Occupancy Prediction 提出GraphGSOcc以解决3D语义占用预测中的动态静态耦合问题 3D gaussian splatting 3DGS gaussian splatting
22 Affogato: Learning Open-Vocabulary Affordance Grounding with Automated Data Generation at Scale 提出Affogato以解决开放词汇的可用性定位问题 open-vocabulary open vocabulary affordance ✅
23 OV-MAP : Open-Vocabulary Zero-Shot 3D Instance Segmentation Map for Robots 提出OV-MAP以解决开放世界3D实例分割问题 open-vocabulary open vocabulary

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

#题目一句话要点标签🔗⭐
24 Dynamic Double Space Tower 提出动态双空间塔以解决视觉问答中的推理不足问题 spatial relationship multimodal
25 SphereDrag: Spherical Geometry-Aware Panoramic Image Editing 提出SphereDrag以解决全景图像编辑中的几何问题 geometric consistency

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

#题目一句话要点标签🔗⭐
26 SignAligner: Harmonizing Complementary Pose Modalities for Coherent Sign Language Generation 提出SignAligner以解决手语生成中的多模态协调问题 spatiotemporal multimodal
27 EyeSim-VQA: A Free-Energy-Guided Eye Simulation Framework for Video Quality Assessment 提出EyeSim-VQA以解决视频质量评估中的自适应修复问题 spatiotemporal

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
28 EgoPrivacy: What Your First-Person Camera Says About You? 提出EgoPrivacy以评估第一人称视频的隐私风险 egocentric egocentric vision first-person view ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
29 Efficient Multi-Camera Tokenization with Triplanes for End-to-End Driving 提出基于三平面的多摄像头高效标记方法以提升自动驾驶性能 motion planning

⬅️ 返回 cs.CV 首页 · 🏠 返回主页