MF-MOS: A Motion-Focused Model for Moving Object Segmentation

被引:7
作者
Cheng, Jintao [1 ]
Zeng, Kang [1 ]
Huang, Zhuoxu [2 ]
Tang, Xiaoyu [1 ]
Wu, Jin [3 ]
Zhang, Chengxi [4 ]
Chen, Xieyuanli [5 ]
Fan, Rui [6 ,7 ,8 ]
机构
[1] South China Normal Univ, Sch Elect & Informat Engn, Foshan 528225, Peoples R China
[2] Aberystwyth Univ, Dept Comp Sci, Aberystwyth SY23 3DB, Dyfed, Wales
[3] Hong Kong Univ Sci & Technol, Dept Elect & Comp Engn, Hong Kong, Peoples R China
[4] Jiangnan Univ, Sch Internet Things Engn, Wuxi, Jiangsu, Peoples R China
[5] Natl Univ Def Technol, Coll Intelligence Sci & Technol, Changsha, Peoples R China
[6] Tongji Univ, Coll Elect & Informat Engn, Shanghai Res Inst Intelligent Autonomous Syst, Shanghai 201804, Peoples R China
[7] Tongji Univ, State Key Lab Intelligent Autonomous Syst, Shanghai 201804, Peoples R China
[8] Tongji Univ, Frontiers Sci Ctr Intelligent Autonomous Syst, Shanghai 201804, Peoples R China
来源
2024 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2024) | 2024年
基金
中国国家自然科学基金;
关键词
D O I
10.1109/ICRA57147.2024.10611400
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Moving object segmentation (MOS) provides a reliable solution for detecting traffic participants and thus is of great interest in the autonomous driving field. Dynamic capture is always critical in the MOS problem. Previous methods capture motion features from the range images directly. Differently, we argue that the residual maps provide greater potential for motion information, while range images contain rich semantic guidance. Based on this intuition, we propose MF-MOS, a novel motion-focused model with a dual-branch structure for LiDAR moving object segmentation. Novelly, we decouple the spatial-temporal information by capturing the motion from residual maps and generating semantic features from range images, which are used as movable object guidance for the motion branch. Our straightforward yet distinctive solution can make the most use of both range images and residual maps, thus greatly improving the performance of the LiDAR-based MOS task. Remarkably, our MF-MOS achieved a leading IoU of 76.7% on the MOS leaderboard of the SemanticKITTI dataset upon submission, demonstrating the current state-of-the-art performance. The implementation of our MF-MOS has been released at https://github.com/SCNU-RISLAB/MF-MOS.
引用
收藏
页码:12499 / 12505
页数:7
相关论文
共 23 条
[1]   Automatic Labeling to Generate Training Data for Online LiDAR-Based Moving Object Segmentation [J].
Chen, Xieyuanli ;
Mersch, Benedikt ;
Nunes, Lucas ;
Marcuzzi, Rodrigo ;
Vizzo, Ignacio ;
Behley, Jens ;
Stachniss, Cyrill .
IEEE ROBOTICS AND AUTOMATION LETTERS, 2022, 7 (03) :6107-6114
[2]   Moving Object Segmentation in 3D LiDAR Data: A Learning-Based Approach Exploiting Sequential Data [J].
Chen, Xieyuanli ;
Li, Shijie ;
Mersch, Benedikt ;
Wiesmann, Louis ;
Gall, Jurgen ;
Behley, Jens ;
Stachniss, Cyrill .
IEEE ROBOTICS AND AUTOMATION LETTERS, 2021, 6 (04) :6529-6536
[3]   Online static point cloud map construction based on 3D point clouds and 2D images [J].
Chi, Peng ;
Liao, Haipeng ;
Zhang, Qin ;
Wu, Xiangmiao ;
Tian, Jiyu ;
Wang, Zhenmin .
VISUAL COMPUTER, 2024, 40 (04) :2889-2904
[4]  
Cortinhal Tiago, 2020, Advances in Visual Computing. 15th International Symposium, ISVC 2020. Proceedings. Lecture Notes in Computer Science (LNCS 12510), P207, DOI 10.1007/978-3-030-64559-5_16
[5]   The Pascal Visual Object Classes (VOC) Challenge [J].
Everingham, Mark ;
Van Gool, Luc ;
Williams, Christopher K. I. ;
Winn, John ;
Zisserman, Andrew .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2010, 88 (02) :303-338
[6]   RangeDet: In Defense of Range View for LiDAR-based 3D Object Detection [J].
Fan, Lue ;
Xiong, Xuan ;
Wang, Feng ;
Wang, Naiyan ;
Zhang, Zhaoxiang .
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, :2898-2907
[7]  
Fan R., 2023, Autonomous driving perception
[8]   OctoMap: an efficient probabilistic 3D mapping framework based on octrees [J].
Hornung, Armin ;
Wurm, Kai M. ;
Bennewitz, Maren ;
Stachniss, Cyrill ;
Burgard, Wolfram .
AUTONOMOUS ROBOTS, 2013, 34 (03) :189-206
[9]   Remove, then Revert: Static Point cloud Map Construction using Multiresolution Range Images [J].
Kim, Giseop ;
Kim, Ayoung .
2020 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2020, :10758-10765
[10]   RVMOS: Range-View Moving Object Segmentation Leveraged by Semantic and Motion Features [J].
Kim, Jaeyeul ;
Woo, Jungwan ;
Im, Sunghoon .
IEEE ROBOTICS AND AUTOMATION LETTERS, 2022, 7 (03) :8044-8051