cs.CV（2025-11-06）

📊 共 17 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱三：空间感知 (Perception & SLAM) (11 🔗2) 支柱一：机器人控制 (Robot Control) (5) 支柱二：RL算法与架构 (RL & Architecture) (1 🔗1)

🔬 支柱三：空间感知 (Perception & SLAM) (11 篇)

#题目一句话要点标签🔗⭐
1 FastGS: Training 3D Gaussian Splatting in 100 Seconds FastGS：基于多视角一致性的3D高斯溅射加速训练框架 3D gaussian splatting 3DGS gaussian splatting
2 CaRF: Enhancing Multi-View Consistency in Referring 3D Gaussian Splatting Segmentation CaRF：通过增强多视角一致性改进Referring 3D高斯溅射分割 3D gaussian splatting gaussian splatting scene understanding
3 BoRe-Depth: Self-supervised Monocular Depth Estimation with Boundary Refinement for Embedded Systems 提出BoRe-Depth模型，在嵌入式系统上实现高精度、高效率的单目深度估计，并提升边界质量。 depth estimation monocular depth ✅
4 Simple 3D Pose Features Support Human and Machine Social Scene Understanding 提出基于3D姿态特征的人机社交场景理解方法，超越现有AI模型。 depth estimation scene understanding social interaction
5 UniSplat: Unified Spatio-Temporal Fusion via 3D Latent Scaffolds for Dynamic Driving Scene Reconstruction UniSplat：通过3D潜在支架实现动态驾驶场景的统一时空融合重建 novel view synthesis scene reconstruction
6 Registration-Free Monitoring of Unstructured Point Cloud Data via Intrinsic Geometrical Properties 提出一种免配准的点云数据监控方法，用于检测3D物体几何精度。 point cloud
7 Self-Supervised Implicit Attention Priors for Point Cloud Reconstruction 提出自监督隐式注意力先验，用于点云重建，提升细节保持和鲁棒性。 point cloud
8 Temporal Zoom Networks: Distance Regression and Continuous Depth for Efficient Action Localization 提出边界距离回归与自适应时间细化以提升动作定位效率 localization
9 DMSORT: An efficient parallel maritime multi-object tracking architecture for unmanned vessel platforms DMSORT：一种高效的并行海事多目标跟踪架构，适用于无人船平台 navigation ✅
10 Room Envelopes: A Synthetic Dataset for Indoor Layout Reconstruction from Images 提出Room Envelopes数据集，用于图像室内布局重建，提升场景理解能力。 scene reconstruction
11 Improving Multi-View Reconstruction via Texture-Guided Gaussian-Mesh Joint Optimization 提出纹理引导的高斯-网格联合优化方法，提升多视角重建质量 novel view synthesis

🔬 支柱一：机器人控制 (Robot Control) (5 篇)

#题目一句话要点标签🔗⭐
12 DINOv2 Driven Gait Representation Learning for Video-Based Visible-Infrared Person Re-identification 提出DinoGRL框架，利用DINOv2驱动的步态特征学习，解决视频可见光-红外行人重识别问题。 gait representation learning
13 3D Gaussian Point Encoders 提出基于3D高斯点编码器的点云表示方法，加速3D识别任务。 running Mamba 3D gaussian splatting
14 Walking the Schrödinger Bridge: A Direct Trajectory for Text-to-3D Generation 提出基于Schrödinger桥的直接轨迹以解决文本到3D生成中的伪影问题 walking classifier-free guidance
15 EETnet: a CNN for Gaze Detection and Tracking for Smart-Eyewear EETnet：为智能眼镜设计的基于事件的低功耗注视检测与跟踪CNN running
16 Faithful Contouring: Near-Lossless 3D Voxel Representation Free from Iso-surface 提出 Faithful Contouring，实现近乎无损的3D体素表示，无需等值面提取。 manipulation

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗⭐
17 DORAEMON: A Unified Library for Visual Object Modeling and Representation Learning at Scale DORAEMON：一个用于大规模视觉对象建模和表征学习的统一库 representation learning ✅

⬅️ 返回 cs.CV 首页 · 🏠 返回主页