arXivDaily arXiv每日学术速递 周一至周五更新

视觉与机器人

3D 视觉

三维重建、NeRF、Gaussian Splatting、点云和空间智能。

共收录 7719 信号源:cs.CV, cs.GR, cs.RO

1. 点云 7719 篇

2008.05309 2020-08-13 cs.CV cs.RO 76%

Factor Graph based 3D Multi-Object Tracking in Point Clouds

Johannes Pöschmann, Tim Pfeifer, Peter Protzel

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

Comments 8 pages, 4 figures, accepted by IEEE Intl. Conf. on Intelligent Robots and Systems (IROS) 2020, visualization of the results of our offline tracker available at https://www.youtube.com/watch?v=mvZmli4jrZQ

详情

展开后加载摘要…

URL PDF HTML 收藏
2001.07360 2020-04-27 cs.CV cs.RO 76%

From Planes to Corners: Multi-Purpose Primitive Detection in Unorganized 3D Point Clouds

Christiane Sommer, Yumin Sun, Leonidas Guibas, Daniel Cremers, Tolga Birdal

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

Comments Accepted to IEEE Robotics and Automation Letters 2020 | Video: https://youtu.be/nHWJrA6RcB0 | Code: https://github.com/c-sommer/orthogonal-planes

Journal ref IEEE Robotics and Automation Letters 5(2) 2020, 1764-1771

详情

展开后加载摘要…

URL PDF HTML 收藏
2002.09147 2020-02-24 cs.RO cs.CV eess.IV 76%

SemanticPOSS: A Point Cloud Dataset with Large Quantity of Dynamic Instances

Yancheng Pan, Biao Gao, Jilin Mei, Sibo Geng, Chengkun Li, Huijing Zhao

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

Comments submited to IEEE Intelligent Vehicles Symposium(2020)

详情

展开后加载摘要…

URL PDF HTML 收藏
1909.11811 2019-09-27 cs.RO cs.CV 76%

A fast, complete, point cloud based loop closure for LiDAR odometry and mapping

Jiarong Lin, Fu Zhang

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

详情

展开后加载摘要…

URL PDF HTML 收藏
1907.07160 2019-07-17 cs.CV cs.RO eess.IV 76%

EnforceNet: Monocular Camera Localization in Large Scale Indoor Sparse LiDAR Point Cloud

Yu Chen, Guan Wang

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

详情

展开后加载摘要…

URL PDF HTML 收藏
1808.00057 2019-04-02 cs.CV cs.RO 76%

Learning to See Forces: Surgical Force Prediction with RGB-Point Cloud Temporal Convolutional Networks

Cong Gao, Xingtong Liu, Michael Peven, Mathias Unberath, Austin Reiter

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

Comments MICCAI 2018 workshop, CARE(Computer Assisted and Robotic Endoscopy)

详情

展开后加载摘要…

URL PDF HTML 收藏
1312.5033 2013-12-19 cs.RO cs.CV 76%

Evaluation of Plane Detection with RANSAC According to Density of 3D Point Clouds

Tomofumi Fujiwara, Tetsushi Kamegawa, Akio Gofuku

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

Comments 3 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
2108.06230 2023-01-20 cs.CV 76%

Generative Zero-Shot Learning for Semantic Segmentation of 3D Point Clouds

Björn Michele, Alexandre Boulch, Gilles Puy, Maxime Bucher, Renaud Marlet

专题命中 点云 :point cloud(title);分类 cs.CV;3D vision(journal_ref)

Comments For the published code, see https://github.com/valeoai/3DGenZ

Journal ref Proceedings of the 2021 International Conference on 3D Vision (3DV 2021), pp. 992-1002

详情

展开后加载摘要…

URL PDF HTML 收藏
1710.07563 2017-10-23 cs.CV 76%

SEGCloud: Semantic Segmentation of 3D Point Clouds

Lyne P. Tchapmi, Christopher B. Choy, Iro Armeni, JunYoung Gwak, Silvio Savarese

专题命中 点云 :point cloud(title);分类 cs.CV;3D vision(comments)

Comments Accepted as a spotlight at the International Conference of 3D Vision (3DV 2017)

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.12901 2026-03-31 cs.CV cs.GR cs.LG cs.RO 75%

SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms

SimULi: 基于无散射变换的实时激光雷达和摄像头模拟

Haithem Turki, Qi Wu, Xin Kang, Janick Martinez Esturo, Shengyu Huang, Ruilong Li, Zan Gojcic, Riccardo de Lutio

专题命中 点云 :NeRF(abstract);3DGS(abstract);分类 cs.CV、cs.GR、cs.RO

AI总结 SimULi 提出一种实时渲染任意相机模型和激光雷达数据的方法,通过扩展 3DGUT 并引入自动化拼接策略和射线剔除,提升了渲染速度和精度,适用于自动驾驶场景的高保真测试。

Comments ICLR 2026 - project page: https://research.nvidia.com/labs/sil/projects/simuli

详情

展开后加载摘要…

URL PDF HTML 收藏
2308.05478 2023-08-11 cs.RO cs.AI cs.CV 74%

Reviewing 3D Object Detectors in the Context of High-Resolution 3+1D Radar

Patrick Palmer, Martin Krueger, Richard Altendorfer, Ganesh Adam, Torsten Bertram

专题命中 点云 :3D vision(abstract,comments);point cloud(abstract);分类 cs.CV、cs.RO

Comments Published at CVPR 2023 Workshop on 3D Vision and Robotics (https://drive.google.com/file/d/1xj4R5ucH3PaR7QdRDJbbkjS-3iBUsruR/view)

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.15577 2026-08-14 cs.CV 版本更新 74%

Shortest-Path Decomposition for Foliage-Robust 3D Tree Modeling and Above-Ground Biomass Estimation from Point Clouds

适用于点云的抗 foliage 三维树木建模与地上生物量估算的最短路径分解方法

Di Wang, Shi Li

机构 * School of Software Engineering, Xi'an Jiaotong University, Xi'an 710049, China(软件工程学院,西安交通大学,西安710049,中国) Shaanxi Joint (Key) Laboratory for Artificial Intelligence (Xi’an Jiaotong University), Xi'an 710049, China(陕西省人工智能联合(重点)实验室(西安交通大学),西安710049,中国)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 该研究提出拓扑驱动的 foliage 抑制最短路径分解方法,从单一点云恢复树木分支层次,在有叶条件下的AGB估算精度优于传统QSM方法,为相关森林清查业务应用消除了障碍。

Comments 14 pages, 13 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.10708 2026-08-12 cs.CV 新提交 74%

Self-Geometry: GT-Free and Plug-and-Play Test-Time Adaptation for Geometrically Consistent 3D Vision Foundation Models

自几何:面向几何一致的3D视觉基础模型的无GT、即插即用测试时自适应方法

Seokhyun Youn, Dahyeon Kye, Sung-Ho Bae, Jihyong Oh

机构 * Chung-Ang University(中央大学) Kyung Hee University(庆熙大学)

专题命中 点云 :3D vision(title);分类 cs.CV

AI总结 本文针对3D视觉基础模型测试时几何一致性不足的问题,提出即插即用的Self-Geometry自适应流水线,结合多视图与对极一致性损失等,在6种模型和4个基准上实现位姿与几何估计的一致提升。

Comments Project page: https://cmlab-korea.github.io/Self-Geometry/

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.19171 2026-07-22 cs.CV 新提交 74%

Point Ladder Tuning: Parameter-Efficient Hierarchical Adaptation for 3D Point Cloud Understanding

点阶梯调优:用于3D点云理解的参数高效分层适配

Junlin Chang, Longhao Zou, Rui Li

机构 * Beihang University(北京航空航天大学) Pengcheng Laboratory(鹏城实验室)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 研究针对3D点云理解中微调预训练主干参数效率低的问题,提出点阶梯调优框架PLT,通过构建分层网络、融合局部与全局特征、生成动态提示等进行参数高效分层适配,实验表明其以极少参数达最优性能。

Comments Accepted to ECCV 2026. Code: https://github.com/JunLinChang/ECCV2026-PLT

详情

展开后加载摘要…

URL PDF HTML 收藏
2501.07399 2026-07-16 cs.RO 74%

Efficiently Closing Loops in LiDAR-Based SLAM Using Point Cloud Density Maps

利用点云密度地图高效关闭LiDAR基于SLAM中的环路

Saurabh Gupta, Tiziano Guadagnino, Benedikt Mersch, Niklas Trekel, Meher V. R. Malladi, Cyrill Stachniss

机构 * University of Bonn, Center for Robotics(波恩大学机器人中心) Lamarr Institute for Machine Learning and Artificial Intelligence(拉马尔机器学习与人工智能研究所)

专题命中 点云 :point cloud(title);分类 cs.RO

AI总结 本文提出了一种鲁棒的环路关闭检测方法,通过生成局部地图并使用地面对齐模块处理平面和非平面运动,实现高效的SLAM系统。

Comments Accepted for publication at the International Journal of Robotics Research on 14 April, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.06516 2026-07-08 cs.CV 新提交 74%

Point as Skeleton: Accumulated Point Cloud Enhanced Autoregressive Generation for Closed-Loop Autonomous Driving Simulation

点作为骨架:累积点云增强自回归生成用于闭环自动驾驶模拟

Songbur Wong, Xiaosong Jia, Junqi You, Bo Zhang, Pei Xu, Renqiu Xia, Yuping Qiu, Shaofeng Zhang, Zelin Zhao, Xuechao Yan, Yuchen Zhou, Yurui Chen, Wen Guo, Hang Xu, Junchi Yan

机构 * Shanghai Jiao Tong University(上海交通大学) Yinwang Intelligent Technology Co., Ltd.(银望智能科技有限公司) Fudan University(复旦大学) Shanghai Artificial Intelligence Laboratory(上海人工智能实验室) University of Science and Technology of China(中国科学技术大学) Georgia Institute of Technology(佐治亚理工学院)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 针对自动驾驶模拟难题,提出“点作为骨架”框架,通过自回归生成器结合多种条件合成视频,引入Reset-and-Roll支持闭环,用点云骨架稳定误差,实现基于nuPlan的闭环生成接口,实验证明其提升了闭环自回归生成质量。

详情

展开后加载摘要…

URL PDF HTML 收藏
2312.08230 2026-07-08 cs.CV cs.LG 版本更新 74%

Partial Symmetry Detection for 3D Geometry using Contrastive Learning with Geodesic Point Cloud Patches

使用测地线点云补丁对比学习进行3D几何的部分对称检测

Gregor Kobsik, Isaak Lim, Leif Kobbelt

机构 * Visual Computing Institute, RWTH Aachen University(视觉计算研究所,亚琛工业大学)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 研究3D几何中部分对称检测难题,提出自监督对比学习框架SymCL,通过映射局部测地线补丁到欧几里得群不变潜在空间,将对称检测转化为基于密度的聚类问题,能同时发现多实例对称关系,在新基准上定量评估并展示泛化能力。

Comments 8 pages, 8 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.26700 2026-06-26 cs.RO cs.AI 新提交 74%

Learning Motion Feasibility from Point Clouds in Cluttered Environments

从杂乱环境中的点云学习运动可行性

Sajid Ansari, Arthi, Girish Varma, Antony Thomas

机构 * International Institute of Information Technology Hyderabad(国际信息技术学院海得拉巴)

专题命中 点云 :point cloud(title);分类 cs.RO

AI总结 提出直接从RGB-D观测学习7自由度机械臂运动可行性预测的方法,构建包含270万抓取可行性标签的大规模基准,并评估三种分类器架构,最佳模型GRASPFC-PTX在新型物体上AUROC达0.996且预测速度远超基于采样的运动规划器。

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.04891 2026-06-04 cs.CV cs.CG 74%

Hierarchical Space Partition for Surface Reconstruction

表面重建的层次空间划分

Minjie Tang, Xiangfei Li

机构 * Independent Researcher(独立研究员) Huazhong University of Science and Technology(华中科技大学)

专题命中 点云 :3D vision(abstract,comments);point cloud(abstract);分类 cs.CV

AI总结 针对点云重建中因LiDAR扫描局限导致细节缺失的问题,提出基于平面分类与优先级生长的层次空间划分方法,并通过最小割优化生成水密多边形网格。

Comments Published in 2026 International Conference on 3D Vision (3DV)

Journal ref in 2026 International Conference on 3D Vision (3DV), Vancouver, BC, Canada, 2026, pp. 207-216

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.29663 2026-06-03 cs.RO 74%

EXACT-MPPI: Exact Signed-Distance Navigation for Arbitrary-Footprint Robots from Point Clouds via Path Integral Control

EXACT-MPPI:通过路径积分控制实现点云中任意足迹机器人的精确有符号距离导航

Chen Peng, Zhikang Ge, Wenwu Lu, Haiming Gao, Stavros Vougioukas, Peng Wei

机构 * ZJU-Hangzhou Global Scientific and Technological Innovation Center, Zhejiang University, Hangzhou, China(浙江大学杭州全球科技创新中心) College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, China(浙江大学生物系统工程与食品科学学院) Department of Biological and Agricultural Engineering, University of California, Davis, Davis, California, USA(加州大学戴维斯分校生物与农业工程系)

专题命中 点云 :point cloud(title);分类 cs.RO

AI总结 提出EXACT-MPPI框架,将解析精确有符号距离评估器嵌入模型预测路径积分控制器,无需中间地图表示,直接处理点云实现任意形状足迹机器人的安全导航。

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.25279 2026-05-26 cs.RO 74%

GreenSeg: Ground Segmentation Algorithm for Agricultural Robots in Mediterranean Greenhouses using RGB-D Point Clouds

GreenSeg: 基于RGB-D点云的地中海温室农业机器人地面分割算法

Fernando Cañadas-Aránega, José C. Moreno, José L. Blanco-Claraco

机构 * Department of Informatics, CIESOL, ceiA3, Universidad de Almería(信息学院,CIESOL,ceiA3,阿尔梅里亚大学)

专题命中 点云 :point cloud(title);分类 cs.RO

AI总结 针对地中海温室狭窄通道、异构地形和光学干扰等挑战,提出一种基于RGB-D感知的双层验证地面分割框架GreenSeg,通过全局平面拟合、曲率滤波和种子点区域生长实现稳定导航,在AGRICOBIOT I平台上验证了其在动态光照下优于基准方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.26759 2026-03-31 cs.CV 74%

Physics-Aware Diffusion for LiDAR Point Cloud Densification

面向激光雷达点云密集化的物理感知扩散模型

Zeping Zhang, Robert Laganière

机构 * University of Ottawa(渥太华大学)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 本文提出Scanline-Consistent Range-Aware Diffusion模型,通过概率细化提升激光雷达点云密集度,以156ms实现高保真结果,采用Ray-Consistency损失和Negative Ray增强技术抑制物理幻觉,提升3D检测性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.23162 2026-03-25 cs.RO 74%

LiZIP: An Auto-Regressive Compression Framework for LiDAR Point Clouds

LiZIP:一种用于激光雷达点云的自回归压缩框架

Aditya Shibu, Kayvan Karim, Claudio Zito

机构 * MACS Heriot Watt University Dubai, United Arab Emirates(MACS 哈罗德·瓦特大学迪拜,阿联酋)

专题命中 点云 :point cloud(title);分类 cs.RO

AI总结 本文提出LiZIP框架,通过神经预测编码实现轻量级近无损压缩,优于现有方法,在不同数据集上实现更高的压缩比和效率,适用于V2X和大规模云存储。

Comments 8 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.22229 2026-03-24 cs.CV 74%

Benchmarking Deep Learning Models for Aerial LiDAR Point Cloud Semantic Segmentation under Real Acquisition Conditions: A Case Study in Navarre

在真实采集条件下评估深度学习模型用于空中LiDAR点云语义分割:纳瓦拉案例研究

Alex Salvatierra, José Antonio Sanz, Christian Gutiérrez, Mikel Galar

机构 * Department of Statistics, Computer Science and Mathematics and Institute of Smart Cities (ISC), Public University of Navarre (UPNA)(统计、计算机科学与数学系和智能城市研究所(ISC),纳瓦拉公共大学(UPNA)) Tracasa Instrumental

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 本文评估了四种深度学习模型在真实飞行条件下大规模空中LiDAR数据集上的性能,揭示了空中数据中类别不平衡和几何变化的挑战,结果显示KPConv在多个类别上表现最佳。

Comments 6 pages, 2 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.14951 2026-03-17 cs.CV 74%

GT-PCQA: Geometry-Texture Decoupled Point Cloud Quality Assessment with MLLM

GT-PCQA: 一种基于多模态大语言模型的几何-纹理解耦点云质量评估方法

Guohua Zhang, Jian Jin, Meiqin Liu, Chao Yao, Weisi Lin, Yao Zhao

机构 * Beijing Jiaotong University(北京交通大学) Nanyang Technological University(南洋理工大学) University of Science and Technology Beijing(北京科技大学)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 本文提出GT-PCQA框架,通过2D-3D联合训练和几何-纹理解耦策略,解决点云质量评估中数据量少和模型对几何退化不敏感的问题,实现高效且泛化能力强的评估方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21667 2026-03-17 cs.CV 74%

Send Less, Perceive More: Masked Quantized Point Cloud Communication for Loss-Tolerant Collaborative Perception

少传多感知:基于掩码量化点云通信的容忍损失协同感知

Sheng Xu, Enshu Wang, Hongfei Xue, Jian Teng, Bingyi Liu, Yi Zhu, Pu Wang, Libing Wu, Chunming Qiao

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 本文提出QPoint2Comm框架,通过量化点云索引减少带宽消耗,提升3D信息精度,并采用掩码训练策略增强抗丢包能力,实验表明其在准确率、通信效率和抗丢包性能上均达新高。

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.11365 2026-03-13 cs.RO 74%

D-SLAMSpoof: An Environment-Agnostic LiDAR Spoofing Attack using Dynamic Point Cloud Injection

D-SLAMSpoof:一种基于动态点云注入的环境无关激光雷达欺骗攻击

Rokuto Nagata, Kenji Koide, Kazuma Ikeda, Ozora Sako, Kentaro Yoshioka

专题命中 点云 :point cloud(title);分类 cs.RO

AI总结 D-SLAMSpoof是一种通过动态点云注入实现的环境无关激光雷达欺骗攻击,利用扫描匹配原理提升攻击成功率,并提出ISD-SLAM作为仅依赖惯性信号的防御方法,有效缓解位置漂移问题。

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
2506.06944 2026-01-05 cs.CV cs.AI cs.LG 74%

Towards Streaming LiDAR Object Detection with Point Clouds as Egocentric Sequences

面向点云作为自车序列的流式LiDAR目标检测

Mellon M. Zhang, Glen Chou, Saibal Mukhopadhyay

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 PFCF提出了一种混合检测器,结合极坐标处理和笛卡尔推理,实现低延迟高精度的LiDAR目标检测。

Comments Accepted to WACV 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.07211 2025-12-09 cs.CV 74%

Object Pose Distribution Estimation for Determining Revolution and Reflection Uncertainty in Point Clouds

点云中确定旋转和反射不确定性时的物体姿态分布估计

Frederik Hagelskjær, Dimitrios Arapis, Steffen Madsen, Thorbjørn Mosekjær Iversen

机构 * SDU Robotics(SDU机器人研究所) The Mærsk Mc-Kinney Møller Institute(马士基麦金尼莫勒研究所) University of Southern Denmark(南部丹麦大学)

专题命中 点云 :point cloud(title);分类 cs.CV

AI总结 本文提出了一种基于深度学习的点云姿态分布估计方法,无需RGB输入,用于评估旋转和反射的不确定性。

Comments 8 pages, 8 figures, 5 tables, ICCR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏