cs.CV（2026-06-17）

📊 共 31 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (8 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (8) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗2) 支柱一：机器人控制 (Robot Control) (6) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱五：交互与反应 (Interaction & Reaction) (1 🔗1)

🔬 支柱三：空间感知与语义 (Perception & Semantics) (8 篇)

#题目一句话要点标签🔗⭐
1 Hand-4DGS: Feed-Forward 3D Gaussian Splatting for 4D Hand Reconstruction from Egocentric Videos 提出Hand-4DGS以解决动态4D手重建问题 3D gaussian splatting gaussian splatting splatting
2 Splaxel: Efficient Distributed Training of 3D Gaussian Splatting for Large-scale Scene Reconstruction via Pixel-level Communication 提出Splaxel以解决大规模场景重建中的通信效率问题 3D gaussian splatting 3DGS gaussian splatting
3 EDoF-NeRF: extended depth-of-field neural radiance fields using a coded aperture camera 提出EDoF-NeRF以解决深度范围限制问题 NeRF neural radiance field
4 OneCanvas: 3D Scene Understanding via Panoramic Reprojection 提出OneCanvas以解决3D场景理解中的空间推理问题 scene understanding embodied AI
5 CABLE: Cloud-Assisted Bandwidth-efficient LMM-based Encoding for V2X Systems 提出CABLE以解决V2X系统中的带宽效率问题 open-vocabulary open vocabulary multimodal
6 Technical Report for ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Leveraging DINOv3 for Robust Outdoor Scene Understanding in Field Robotics 提出基于DINOv3的解决方案以提升户外场景理解能力 scene understanding
7 NeuMesh++: Towards Versatile and Efficient Volumetric Editing with Disentangled Neural Mesh-based Implicit Field 提出NeuMesh++以解决神经隐式场编辑功能有限的问题 neural radiance field scene reconstruction ✅
8 From Bounding Boxes to Visual Reasoning: An On-Policy Data Annotation Tool for Vision-Language Models 提出ScreenAnnotator以解决视觉语言模型数据标注问题 open-vocabulary open vocabulary ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#题目一句话要点标签🔗⭐
9 SAMA: Semantic Anchor-aligned Augmentation for Unified Low-Resource Multimodal Information Extraction 提出SAMA以解决低资源多模态信息提取问题 large language model multimodal
10 Where Will They Go? Modelling Multimodal Pedestrian Manoeuvres from Ego-centric Videos 提出MMPM框架以解决行人轨迹预测中的多模态问题 multimodal
11 Toward Training-Free Zero-Shot Anomaly Detection in 3D Medical Images: A Batch-Based Approach Using 2D Foundation Models 提出CS3F框架以解决3D医学图像中的零-shot异常检测问题 foundation model
12 PEFT-MedSAM: Efficient Fine-Tuning of Medical Foundation Models for Explainable Skin Lesion Segmentation 提出PEFT-MedSAM以解决医学图像皮肤病变分割问题 foundation model
13 LandslideAgent with Multimodal LandslideBench: A Domain-Rule-Augmented Agent for Autonomous Landslide Identification and Analysis 提出LandslideAgent以解决滑坡自动识别与分析问题 multimodal
14 A Unified Framework for Efficient Remote Sensing Visual Question Answering: Adapting Dual, Hybrid, and Encoder-Decoder Architectures 提出统一框架以提高遥感视觉问答效率 foundation model multimodal
15 AMALIA-VL: A Native European Portuguese Open-Source Vision and Language Model 提出AMALIA-VL以解决欧洲葡萄牙语多模态模型不足问题 multimodal
16 Bridging Creative Intent and Visual Quality: Creator-Driven Recurrent Video Generation with Agentic Feedback Loops 提出CHIEF框架以解决视频生成中的创意指导不足问题 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#题目一句话要点标签🔗⭐
17 Visual-OPSD: Cross-Modal On-Policy Self-Distillation for Efficient Unified Multimodal Reasoning 提出Visual-OPSD以解决多模态推理中的效率问题 distillation multimodal
18 FlowObject: Flow Steering for Bridging Generative Priors and Reconstruction Fidelity 提出FlowObject以解决3D重建中的合成偏差问题 flow matching 3D gaussian splatting 3DGS
19 DreamReg: Belief-Driven World Model for 2D-3D Ultrasound Registration 提出DreamReg以解决2D-3D超声图像配准问题 world model world models
20 SpectralDiT: Timestep-Conditioned Spectral Residual Correction for Flow-Matching DiTs 提出SpectralDiT以改善流匹配扩散变换器的性能 flow matching classifier-free guidance
21 Spatially Stratified Distillation for Heterogeneous Radar Place Recognition 提出空间分层蒸馏以解决异构雷达位置识别问题 distillation
22 UniTemp: Unlocking Video Generation in Any Temporal Order via Bidirectional Distillation 提出UniTemp以解决视频生成的任意时间顺序问题 distillation ✅
23 Physics-IQ Verified 提出Physics-IQ Verified以提升视频生成模型的物理理解评估 world model world models ✅

🔬 支柱一：机器人控制 (Robot Control) (6 篇)

#题目一句话要点标签🔗⭐
24 Motion-Focused Latent Action Enables Cross-Embodiment VLA Training from Human EgoVideos 提出基于潜在动作的框架以解决无标签人类视频训练问题 manipulation VQ-VAE egocentric
25 Mem-World: Memory-Augmented Action-Conditioned World Models for Persistent Robot Manipulation 提出Mem-World以解决机器人操作中的持久世界建模问题 manipulation world model world models
26 Intrinsic 4D Gaussian Segmentation from Scene Cues 提出Intrinsic-GS以解决动态场景分割问题 manipulation gaussian splatting splatting
27 MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction 提出MolmoMotion以解决3D点轨迹预测问题 manipulation flow matching motion prediction
28 BindEdit: Taming Attention Leakage for Precise Multi-Object Image Editing 提出BindEdit以解决多对象图像编辑中的注意力泄漏问题 manipulation
29 URDF Synthesis from RGB-D Sequences via Differentiable Joint Inference and Energy-Consistent Verification 提出KinemaForge以解决关节物体URDF重建问题 manipulation

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

#题目一句话要点标签🔗⭐
30 Beyond the Current Observation: Evaluating Multimodal Large Language Models in Controllable Non-Markov Games 提出RNG-Bench以评估多模态大语言模型在可控非马尔可夫游戏中的表现 egocentric large language model foundation model

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
31 Taming I2V models for Image HOI Editing: A Cognitive Benchmark and Agentic Self-Correcting Framework 提出HOI-Edit基准与SCPE框架以解决复杂人机交互编辑问题 human-object interaction HOI ✅

⬅️ 返回 cs.CV 首页 · 🏠 返回主页