cs.CV（2025-12-10）

📊 共 19 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱三：空间感知 (Perception & SLAM) (14 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱四：生成式动作 (Generative Motion) (1) 支柱七：动作重定向 (Motion Retargeting) (1 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱三：空间感知 (Perception & SLAM) (14 篇)

#题目一句话要点标签🔗⭐
1 VHOI: Controllable Video Generation of Human-Object Interactions from Sparse Trajectories via Motion Densification VHOI：通过运动稠密化，从稀疏轨迹控制人体-物体交互视频生成 optical flow navigation human-object interaction
2 Splatent: Splatting Diffusion Latents for Novel View Synthesis Splatent：通过Splatting扩散模型潜在空间提升新视角合成质量 3D gaussian splatting 3DGS gaussian splatting
3 Relightable and Dynamic Gaussian Avatar Reconstruction from Monocular Video 提出RnD-Avatar，基于3DGS重建可重光照和动态人体Avatar，提升几何细节。 3D gaussian splatting 3DGS gaussian splatting
4 Generative Point Cloud Registration 提出生成式点云配准方法，利用2D生成模型提升3D匹配性能 point cloud geometric consistency
5 mmWEAVER: Environment-Specific mmWave Signal Synthesis from a Photo and Activity Description mmWeaver：利用照片和活动描述合成环境特定的毫米波信号 point cloud pose estimation MotionGPT
6 FastPose-ViT: A Vision Transformer for Real-Time Spacecraft Pose Estimation 提出FastPose-ViT，用于资源受限平台上的航天器实时姿态估计 pose estimation
7 Detection and Localization of Subdural Hematoma Using Deep Learning on Computed Tomography 提出多模态深度学习框架，用于脑部CT影像中硬膜下血肿的精准检测与定位 localization
8 MoRel: Long-Range Flicker-Free 4D Motion Modeling via Anchor Relay-based Bidirectional Blending with Hierarchical Densification MoRel：基于锚点中继双向融合和分层稠密化的长程无闪烁4D运动建模 3D gaussian splatting 3DGS gaussian splatting
9 FUSER: Feed-Forward MUltiview 3D Registration Transformer and SE(3)$^N$ Diffusion Refinement 提出FUSER以解决多视角点云配准问题 point cloud geometric consistency
10 GAINS: Gaussian-based Inverse Rendering from Sparse Multi-View Captures GAINS：基于高斯的稀疏多视角逆渲染，提升几何与材质恢复质量 monocular depth gaussian splatting ✅
11 TraceFlow: Dynamic 3D Reconstruction of Specular Scenes Driven by Ray Tracing TraceFlow：光线追踪驱动的动态高光场景三维重建 gaussian splatting
12 From Detection to Anticipation: Online Understanding of Struggles across Various Tasks and Activities 提出在线挣扎检测与预测框架，助力实时辅助系统理解人类技能表现 localization
13 Privacy-Preserving Computer Vision for Industry: Three Case Studies in Human-Centric Manufacturing 提出一种面向工业的隐私保护计算机视觉框架，应用于人机协作制造场景 navigation
14 ASSIST-3D: Adapted Scene Synthesis for Class-Agnostic 3D Instance Segmentation ASSIST-3D：用于类别无关3D实例分割的自适应场景合成 point cloud

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
15 Log NeRF: Comparing Spaces for Learning Radiance Fields Log NeRF：通过比较不同色彩空间，提升神经辐射场的学习效果 representation learning NeRF neural radiance
16 CLARGA: Multimodal Graph Representation Learning over Arbitrary Sets of Modalities CLARGA：提出一种通用的多模态图表示学习框架，适用于任意模态组合。 representation learning

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
17 FunPhase: A Periodic Functional Autoencoder for Motion Generation via Phase Manifolds FunPhase：通过相位流形实现运动生成的周期性函数自编码器 motion generation

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

#题目一句话要点标签🔗⭐
18 StereoWorld: Geometry-Aware Monocular-to-Stereo Video Generation StereoWorld：提出几何感知单目视频转立体视频生成框架，提升视觉保真度和几何一致性。 geometric consistency ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
19 FROMAT: Multiview Material Appearance Transfer via Few-Shot Self-Attention Adaptation 提出FROMAT，通过少样本自注意力适配实现多视角材质外观迁移 manipulation

⬅️ 返回 cs.CV 首页 · 🏠 返回主页