cs.CV（2026-10-26）

📊 共 33 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (12 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (9 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (9 🔗1) 支柱一：机器人控制 (Robot Control) (2) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱三：空间感知与语义 (Perception & Semantics) (12 篇)

#题目一句话要点标签🔗⭐
1 SG-Splat: Stable Geometry Learning under Pose Uncertainty for Pose-Free 3D Gaussian Splatting 提出SG-Splat以解决姿态不确定性下的3D高斯表示问题 3D gaussian splatting 3DGS gaussian splatting
2 Zero-Shot Monocular Depth Estimation for Visual Perception in Lower-Limb Prostheses 提出单目深度估计方法以解决下肢假肢的视觉感知问题 depth estimation monocular depth foundation model ✅
3 Complexity-guided Regularization for Generalizable Human Gaussian Splatting 提出复杂度引导正则化以解决人类高斯点云细节重建问题 gaussian splatting splatting ✅
4 VidEdit3D: Flexible 3D Gaussian Editing with Video Diffusion 提出VidEdit3D以解决3D场景编辑灵活性不足的问题 3D gaussian splatting 3DGS gaussian splatting
5 Does VGGT Know What It Does Not Know? 提出线性探测器以识别VGGT模型的相机姿态失败 3D reconstruction VGGT
6 Synthetic Deformable Tubular Benchmark: A mechanically annotated dataset for evaluating deformable 3D reconstruction 提出合成可变管道基准以评估可变形3D重建问题 3D reconstruction optical flow
7 DUET4D: Self-Corrective Coupled Guidance for Sparse-View 4D Reconstruction 提出DUET4D以解决稀疏视图4D重建中的不确定性问题 gaussian splatting splatting
8 Metric Scale for Insta360 Structure-from-Motion from the Camera's Embedded IMU 提出基于IMU的度量尺度恢复方法以解决SfM中的尺度不确定性问题 metric depth
9 EARS: Evidence-Aware Retrieval via Sound and Motion for Budgeted Video Reasoning 提出EARS以解决预算视频推理中的证据检索问题 optical flow ✅
10 Generative Structure from Motion with Native 3D Diffusion 提出GenSfM以解决3D结构与相机运动恢复问题 3D reconstruction ✅
11 Learning Scene Coordinate Reconstruction from Unposed Images via Pose Graph Optimization 提出混合框架以解决无姿态图像的场景坐标重建问题 optical flow
12 DimFlow: Bidirectional Grid-Expert Learning for Dynamic Gaussian Reconstruction 提出DimFlow以解决动态高斯重建中的特征兼容性问题 NeRF

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

#题目一句话要点标签🔗⭐
13 OVM-Occ: Open-Vocabulary Monocular Occupancy Prediction via Geometry-Aware Vision-Language Distillation 提出OVM-Occ以解决室内场景开放词汇单目占用预测问题 distillation open-vocabulary open vocabulary
14 DriveUniTok: A Unified Multimodal Tokenizer for Driving World-Action Models 提出DriveUniTok以解决多模态世界状态建模问题 reinforcement learning policy learning world model
15 OSWAM: An Effective World Action Model with One-Step Action Generation 提出OSWAM以解决实时动作生成延迟问题 world action model world action models distillation ✅
16 S$^2$-RL: Sample-Set Dual Reinforcement Learning for Generative Semantic Segmentation Dataset Distillation 提出S$^2$-RL框架以解决语义分割数据集蒸馏中的计算瓶颈问题 reinforcement learning diffusion policy distillation
17 WorldPager: Zero-Stall Tiered Memory for Long-Horizon World Models on a Single Consumer GPU 提出WorldPager以解决长视频生成中的内存瓶颈问题 world model world models
18 RSKD: Grounding Semantic Consistency in Driving VLMs via Distillation 提出RSKD以解决自动驾驶VLMs中的语义一致性问题 distillation multimodal
19 PosePlaner: Denoising Any Feed-Forward Pose Predictor from Pairwise Planar Geometry 提出PosePlaner以解决多视角姿态预测中的几何信息利用不足问题 flow matching 3D reconstruction VGGT
20 LayoutMamba: Efficient Document Layout Analysis with Hybrid Mamba-Transformer Networks 提出LayoutMamba以解决文档布局分析中的效率与精度问题 Mamba
21 Efficient Multi-modal Video Fusion Network with Dynamic Region-Guided Temporal Interaction 提出动态区域引导的多模态视频融合网络以提升视频融合质量 MAE optical flow

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
22 VibeVideoBench:Evaluating Multimodal Agents on Code-Driven Video Generation 提出VibeVideoBench以评估代码驱动的视频生成多模态智能体 multimodal
23 Multimodal Concert Video Editing: System Design and Exploratory Evaluation 提出多模态框架以优化多摄像头音乐会视频编辑 multimodal
24 LOCUS: Locate Once, Check Until Settled. Draft-and-Verify Visual Grounding with a Box Head and Coordinate-Token Verification 提出LOCUS以解决视觉语言模型中的定位精度与效率问题 visual grounding
25 Accuracy and efficiency of compact foundation models for surface-water mapping during floods 评估紧凑型基础模型在洪水期间的水面映射效率与准确性 foundation model
26 FL4N: Feed-forward Latent Model for 4D Novel View Synthesis 提出FL4N以解决动态场景4D新视角合成问题 TAMP
27 SeeRF: Learning to See Beyond Fields of View via Generative Image Registration and Fusion 提出SeeRF以解决多模态传感器视场限制问题 foundation model ✅
28 MemVLN: Streaming Episodic Memory for Vision-Language Navigation 提出MemVLN以解决长时间导航中的记忆保持问题 VLN
29 Reason or Recite: Complementary Specialists with an LLM Judge for Traffic Anomaly Reasoning 提出Reason-or-Recite框架以解决交通异常推理问题 chain-of-thought
30 Fed-MAPTrack: Federated Modality-aware Prompt Tuning for RGB-X Tracking 提出Fed-MAPTrack以解决RGB-X跟踪中的传感器异构问题 multimodal

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
31 GIFT: Representation Geometry Matters for Single-Domain Generalized Object Detection 提出GIFT以解决单域泛化目标检测中的几何结构问题 MPC foundation model
32 Copying As Defense: Protecting Images from Editing By Eliciting Model Copying 提出Copying-as-Defense以防止图像编辑攻击 manipulation

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
33 Beyond Vision-Language Models: Visual-Semantic Affect Graphs for Emotion Recognition in Artworks 提出视觉语义情感图以解决艺术作品情感识别问题 motion prediction multimodal

⬅️ 返回 cs.CV 首页 · 🏠 返回主页