cs.CV（2026-06-15）

📊 共 47 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) (14 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (13 🔗4) 支柱三：空间感知与语义 (Perception & Semantics) (11 🔗1) 支柱一：机器人控制 (Robot Control) (8 🔗1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#题目一句话要点标签🔗⭐
1 Gen-VCoT: Generative Visual Chain-of-Thought Reasoning via Diffusion-Based RGB Intermediate Representations 提出Gen-VCoT以解决多模态推理中的可解释性问题 large language model multimodal visual grounding
2 LOCUS: Local Visual Cue Search for Enhancing Fine-Grained Perception in Multimodal Large Language Models 提出LOCUS以解决多模态大语言模型的细粒度视觉感知问题 large language model multimodal
3 Unified Multimodal Model for Brain MRI Imputation and Understanding 提出UniBrain以解决脑MRI模态缺失问题 large language model multimodal
4 Learned Image Compression for Vision-Language-Action Models 提出SPARC以解决视觉-语言-动作模型中的图像压缩瓶颈问题 vision-language-action VLA
5 Cascaded Sparse Autoencoders Learn Multi-Level Visual Concepts in Multimodal LLMs 提出级联稀疏自编码器以学习多层次视觉概念 large language model multimodal
6 Multimodal LLM-Empowered Re-Ranking for Generalizable Person Re-Identification 提出多模态LLM增强重排序以解决领域泛化行人重识别问题 large language model multimodal ✅
7 VinQA: Visual Elements Interleaved Long-form Answer Generation for Real-World Multimodal Document QA 提出VinQA以解决多模态文档问答中的视觉元素利用不足问题 large language model multimodal
8 Transformation-driven generation of comparable projection images from multimodal anatomical scenes 提出基于变换驱动的框架以生成可比的投影图像 multimodal
9 UniDDT: Unifying Multimodal Understanding and Generation with Decoupled Diffusion Transformer 提出UniDDT以解决多模态理解与生成的统一问题 multimodal
10 GRACE: Boosting Video MLLMs with Grounded Action-Centric Evidence for Viewer Sentiment Prediction 提出基于事件证据增强的框架以提升视频情感预测 large language model multimodal
11 MeshLoom: Feed-Forward Non-Rigid Registration of Mesh Sequences 提出MeshLoom以解决非刚性网格序列配准问题 TAMP
12 WaveDINO: Learning-Based Atmospheric Correction of Unwrapped InSAR Interferograms Validated by GNSS: Results at Laguna del Maule and Campi Flegrei Volcanoes 提出WaveDINO以解决InSAR干涉图大气校正问题 foundation model
13 Chronological Blindness: Benchmarking Temporal Reasoning in Vision-Language Models with CHRONOSIGHT 提出CHRONOSIGHT以评估视觉语言模型的时间推理能力 instruction following
14 VisualClaw: A Real-Time, Personalized Agent for the Physical World 提出VisualClaw以解决多模态代理在视频处理中的高延迟与静态性问题 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (13 篇)

#题目一句话要点标签🔗⭐
15 SUP-MCRL: Subject-aware Unified Pseudo-feature Coded Multimodal Contrastive Representation Learning for EEG Visual Decoding 提出SUP-MCRL以解决脑机接口视觉解码中的语义一致性问题 representation learning multimodal ✅
16 DifferAD-R1: A Difference-Guided IndustrialAnomaly Localization with Multimodal LargeLanguage Models 提出DifferAD-R1以解决工业异常定位中的未见缺陷检测问题 reinforcement learning large language model multimodal ✅
17 Instance-Aware Knowledge Distillation for Semi-Supervised Learning of an On-Board Multi-Task Dense Prediction Model for Collision Avoidance System 提出实例感知知识蒸馏以解决半监督学习中的碰撞避免问题 distillation depth estimation monocular depth
18 DreamX-World 1.0: A General-Purpose Interactive World Model 提出DreamX-World 1.0以解决长视频生成控制问题 reinforcement learning world model world models
19 FusionRS: A Large-Scale RGB-Infrared Remote Sensing Dataset for Dual-Modal Vision-Language Foundation Models 提出FusionRS以解决RGB-红外遥感数据缺乏问题 representation learning foundation model
20 BadWorld: Adversarial Attacks on World Models 提出BadWorld框架以解决视觉世界模型的对抗攻击问题 world model world models
21 GraphWorld: Long-Horizon Planning with World Models for End-to-End Autonomous Driving 提出GraphWorld以解决长时间规划问题 world model world models
22 Structure-aware Knowledge-guided Heterogeneous Mamba for Zygomaticomaxillary Suture Assessment 提出SKMamba以解决颧上颌缝成熟度评估问题 Mamba large language model ✅
23 Contrastive Learning for Seismic Horizon Tracking with Domain-Specific Priors 提出自监督融合方法以解决无监督地震地平线追踪问题 MAE contrastive learning
24 Beer-Lambert Guided Representation Learning for Unsupervised Anomaly Detection in Sub-THz Food Inspection Images 提出Beer-Lambert引导表示学习以解决Sub-THz食品检测中的异常检测问题 representation learning
25 RGFVR: Reference-Guided Face Video Restoration with Flow Matching 提出参考引导的面部视频修复方法以解决身份保留问题 flow matching ✅
26 LUCID: Learned Undersampling-Adaptive Consistency-Guided Inference with Deterministic Flow Matching for Sparse-View CT Reconstruction 提出LUCID框架以解决稀疏视图CT重建中的伪影问题 flow matching
27 When the Past Matters: FlashBack Memory for Precipitation Nowcasting 提出FlashBack Memory以解决降水短期预报中的长依赖问题 MAE spatiotemporal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (11 篇)

#题目一句话要点标签🔗⭐
28 Training-Free Open-Vocabulary Visual Grounding for Remote Sensing Images and Videos 提出RSVG-ZeroOV以解决遥感图像和视频的开放词汇视觉定位问题 open-vocabulary open vocabulary spatial relationship
29 Local-GS: Accelerating 3D Gaussian Splatting via Tile-Local Warp Coherence 提出Local-GS以解决3D高斯点云渲染效率问题 3D gaussian splatting 3DGS gaussian splatting
30 ActiveSAM: Image-Conditional Class Pruning for Fast and Accurate Open-Vocabulary Segmentation 提出ActiveSAM以解决开放词汇语义分割效率低下问题 open-vocabulary open vocabulary embodied AI ✅
31 Fi-Gaussian: Frequency-Aware Implicit Gaussian Splatting for Single Image Dehazing 提出Fi-Gaussian以解决单幅图像去雾中的高频细节损失问题 gaussian splatting splatting
32 Dehaze-GaussianImage: Zero-Shot Dehazing via Efficient 2D Gaussian Splatting Representation 提出Dehaze-GaussianImage以解决图像去雾问题 gaussian splatting splatting
33 Continuous Splatting meets Retinex: Continuous Gaussian Splatting and Implicit Reflectance Modeling for Low-Light Image Enhancement 提出CGS-Retinex以解决低光照图像增强中的色彩失真问题 gaussian splatting splatting
34 PROSE: Training-Free Egocentric Scene Registration with Vision-Language Models 提出PROSE以解决无训练的自我中心场景注册问题 scene understanding egocentric foundation model
35 MVM-IOD: An Industrial Object-Centric Benchmark Dataset for the Evaluation of 3D Reconstruction Methods 提出MVM-IOD数据集以解决工业3D重建与相机姿态估计问题 3D reconstruction gaussian splatting splatting
36 Uncertainty Quality of VGGT: An Analysis on the DTU Benchmark Dataset 提出VGGT以提升3D重建的不确定性质量 3D reconstruction VGGT feature matching
37 MVOFormer: Flow-Semantic Transformer for Robust Monocular Visual Odometry 提出MVOFormer以解决单目视觉里程计的鲁棒性问题 visual odometry multimodal
38 MMDiff: Extending Diffusion Transformers for Multi-Modal Generation 提出MMDiff以实现多模态生成的高效框架 depth estimation

🔬 支柱一：机器人控制 (Robot Control) (8 篇)

#题目一句话要点标签🔗⭐
39 RealityBridge: Bridging Editable 3D Gaussian Splatting Driving Simulations and Real-World Videos 提出RealityBridge以解决3D高斯点云视频的Sim-to-Real问题 sim-to-real 3D gaussian splatting 3DGS
40 Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation 提出Qwen-RobotWorld以统一语言条件下的机器人世界建模 manipulation world model worldmodel
41 Focus When Necessary: Adaptive Routing and Collaborative Grounding for Training-Free Visual Grounding 提出LazyMCoT以解决高分辨率图像细节感知不足问题 manipulation large language model multimodal ✅
42 EgoPhys: Learning Generalizable Physics Models of Deformable Objects from Egocentric Video 提出EgoPhys以解决可变形物体物理建模问题 manipulation egocentric
43 Latent Space Reinforcement Learning for Inverse Material Estimation in Food Fracture Simulation 提出潜在空间强化学习以解决食品破裂模拟中的逆材料估计问题 manipulation reinforcement learning PPO
44 Decoupled Object-Centric Video Understanding for Generating Robotic Manipulation Commands 提出一种对象中心的视频理解框架以生成机器人操作命令 manipulation
45 Text-Vision Co-Instructed Image Editing 提出文本-视觉共同指导的图像编辑方法以解决编辑精度不足问题 manipulation
46 ResEdit: Residual embeddings for precise generative image editing 提出残差嵌入以解决生成图像编辑中的一致性问题 manipulation

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
47 HAFMat: Hybrid Priors Guided Adaptive Fusion for Single-Image Human Material Estimation 提出HAFMat以解决单图像人类材料估计问题 physically plausible

⬅️ 返回 cs.CV 首页 · 🏠 返回主页