Once Detected, Never Lost: Surpassing Human Performance in Offline LiDAR based 3D Object Detection | |
Fan L(范略)4,5,6; Yang YX(杨禹雪)5,6; Mao YM(毛一鸣)3; Wang F(王峰)1; Chen YT(陈韫韬)2; Wang NY(王乃岩)1; Zhang ZX(张兆翔)2,5,6,7 | |
2023-10 | |
会议日期 | 2023/10/2-2023/10/6 |
会议地点 | 法国巴黎 |
关键词 | 点云目标检测 自动驾驶 |
DOI | 10.1109/ICCV51070.2023.01815 |
英文摘要 | This paper aims for high-performance offline LiDAR-based 3D object detection. We first observe that experienced human annotators annotate objects from a track-centric perspective. They first label objects in a track with clear shapes, and then leverage the temporal coherence to infer the annotations of obscure objects. Drawing inspiration from this, we propose a high-performance offline detector in a track-centric perspective instead of the conventional object-centric perspective. Our method features a bidirectional tracking module and a track-centric learning module. Such design allows our detector to infer and refine a complete track once the object is detected at a certain moment. We refer this characteristic to "onCe detecTed, neveR Lost" and name the proposed system CTRL. Extensive experiments demonstrate the remarkable performance of our method, surpassing the human-level annotating accuracy and previous state-of-the-art methods in the highly competitive Waymo Open Dataset leaderboard without model ensemble. The code is available at https://github.com/tusen-ai/SST. |
语种 | 英语 |
URL标识 | 查看原文 |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/57420] ![]() |
专题 | 自动化研究所_智能感知与计算研究中心 |
作者单位 | 1.图森未来 2.中国科学院香港创新研究院,人工智能与机器人研究中心 3.湖南大学 4.中国科学院大学未来技术学院 5.中国科学院大学 6.中国科学院自动化所 7.多模态人工智能系统全国重点实验室 |
推荐引用方式 GB/T 7714 | Fan L,Yang YX,Mao YM,et al. Once Detected, Never Lost: Surpassing Human Performance in Offline LiDAR based 3D Object Detection[C]. 见:. 法国巴黎. 2023/10/2-2023/10/6. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论