arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~

视觉与机器人

3D 视觉

三维重建、NeRF、Gaussian Splatting、点云和空间智能。

2026-02-12 至 2026-02-12 共收录 8 信号源:cs.CV, cs.GR, cs.RO

1. 点云 8 篇

2602.10492 2026-02-12 cs.CV cs.RO 81%

End-to-End LiDAR optimization for 3D point cloud registration

端到端激光雷达优化用于3D点云配准

Siddhant Katyan, Marc-André Gardner, Jean-François Lalonde

机构 * Université Laval(拉瓦尔大学) Bentley Systems(贝恩特系统)

专题命中 点云 :point cloud(title,abstract);分类 cs.CV、cs.RO

AI总结 本文提出端到端自适应激光雷达框架,通过动态调整参数优化点云配准,提升精度与效率,并在CARLA模拟中验证其优于固定参数方法的性能。

Comments 36th British Machine Vision Conference 2025, {BMVC} 2025, Sheffield, UK, November 24-27, 2025. Project page: https://lvsn.github.io/e2e-lidar-registration/

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.10871 2026-02-12 cs.HC cs.CV 79%

Viewpoint Recommendation for Point Cloud Labeling through Interaction Cost Modeling

通过交互成本建模实现点云标注的视角推荐

Yu Zhang, Xinyi Zhao, Chongke Bi, Siming Chen

机构 * School of Data Science, Fudan University(复旦大学数据科学学院) Department of Computer Science, University of Oxford(牛津大学计算机科学系) College of Intelligence and Computing, Tianjin University(天津大学智能与计算学院) Shanghai Key Laboratory of Data Science, Shanghai(上海数据科学重点实验室)

专题命中 点云 :point cloud(title,abstract);分类 cs.CV

AI总结 本文提出通过交互成本建模减少点云标注时间的方法,利用Fitts定律优化视角推荐,提升标注效率。

Comments Accepted to IEEE TVCG

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.18301 2026-02-12 cs.CV 79%

Contextual Range-View Projection for 3D LiDAR Point Clouds

基于上下文的视距投影用于3D激光雷达点云

Seyedali Mousavi, Seyedhamidreza Mousavi, Masoud Daneshtalab

机构 * School of Innovation, Design and Engineering(创新、设计与工程学院) Division of Intelligent Future Technologies(智能未来技术系) Mälardalen University(马尔默大学)

专题命中 点云 :point cloud(title,abstract);分类 cs.CV

AI总结 本文提出CAP和CWAP两种机制,通过结合实例中心和类别标签的上下文信息,改进3D激光雷达点云的视距投影,提升实例点保留率和目标类别性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2403.02818 2026-02-12 cs.CV 74%

Are Dense Labels Always Necessary for 3D Object Detection from Point Cloud?

密集标签是否总是必要于点云中的3D物体检测?

Chenqiang Gao, Chuandong Liu, Jun Shu, Fangcen Liu, Jiang Liu, Luyu Yang, Xinbo Gao, Deyu Meng

机构 * School of Intelligent Systems Engineering, the Shenzhen Campus of Sun Yatsen University(南方科技大学深圳校区智能系统工程学院) School of Computer Science, Wuhan University(武汉大学计算机学院) School of Mathematics and Statistics and Ministry of Education Key Lab of Intelligent Networks and Network Security, Xi’an Jiaotong University(西安交通大学数学与统计学系和教育部智能网络与网络安全重点实验室) School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications(重庆邮电大学通信与信息工程学院) Meta University of Maryland(美国马里兰大学)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 本文提出SS3D++方法,通过稀疏注释减少标注成本,在KITTI和Waymo数据集上实现性能媲美或超越全监督方法。

Comments update

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.11007 2026-02-12 cs.CV 57%

LaSSM: Efficient Semantic-Spatial Query Decoding via Local Aggregation and State Space Models for 3D Instance Segmentation

LaSSM:通过局部聚合和状态空间模型实现高效的语义-空间查询解码用于3D实例分割

Lei Yao, Yi Wang, Yawen Cui, Moyun Liu, Lap-Pui Chau

机构 * Department of Electrical and Electronic Engineering, The Hong Kong Polytechnic University(电子工程系,香港理工大学) School of Mechanical Science and Engineering, Huazhong University of Science and Technology(机械科学与工程学院,华中科技大学)

专题命中 点云 :point cloud(abstract);分类 cs.CV

AI总结 LaSSM通过局部聚合和状态空间模型实现高效语义-空间查询解码,提升3D实例分割性能。

Comments Accepted at IEEE-TCSVT

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.10806 2026-02-12 cs.CV 57%

DMP-3DAD: Cross-Category 3D Anomaly Detection via Realistic Depth Map Projection with Few Normal Samples

DMP-3DAD:通过真实深度图投影实现跨类3D异常检测

Zi Wang, Katsuya Hotta, Koichiro Kamide, Yawen Zou, Jianjian Qin, Chao Zhang, Jun Yu

机构 * Niigata University(Niigata大学) University of Toyama(Toyama大学) Iwate University(Iwate大学) NingboTech University(NingboTech大学)

专题命中 点云 :point cloud(abstract);分类 cs.CV

AI总结 DMP-3DAD通过真实深度图投影实现跨类3D异常检测,无需训练即可高效检测少样本场景下的异常点云。

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.02186 2026-02-12 cs.CV cs.LG 57%

GeoPurify: A Data-Efficient Geometric Distillation Framework for Open-Vocabulary 3D Segmentation

GeoPurify: 一种数据高效的几何蒸馏框架用于开放词汇3D分割

Weijia Dou, Xu Zhang, Yi Bin, Jian Liu, Bo Peng, Guoqing Wang, Yang Yang, Heng Tao Shen

机构 * Tongji University(同济大学) Tianjin University(天津大学) University of Electronic Science and Technology of China(电子科学与技术大学)

专题命中 点云 :point cloud(abstract);分类 cs.CV

AI总结 GeoPurify通过几何蒸馏和亲和网络提升3D分割的数据效率,实现高效且准确的开放词汇3D分割

Comments Accepted at ICLR 2026. Code available at: https://github.com/tj12323/GeoPurify

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.23599 2026-02-12 cs.LG math.RT math.ST stat.ML stat.TH 50%

On Transferring Transferability: Towards a Theory for Size Generalization

关于转移的转移:迈向大小泛化理论

Eitan Levin, Yuxin Ma, Mateo Díaz, Soledad Villar

机构 * Department of Computing and Mathematical Sciences, Caltech(加州理工学院计算与数学科学系) Department of Applied Mathematics and Statistics and the Mathematical Institute for Data Science, Johns Hopkins University(约翰霍普金斯大学应用数学与统计学系及数据科学数学研究所)

专题命中 点云 :point cloud(abstract)

AI总结 本文提出了一种通用框架,用于研究不同维度间的转移性,并通过理论分析和实验验证了大小泛化的能力。

Comments 75 pages, 10 figures, closest to version to be published in NeurIPS

详情

展开后加载摘要…

URL PDF HTML 收藏