Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR

Conference item

Deep learning has led to great progress in the detection of mobile (i.e. movement-capable) objects in urban driving scenes in recent years. Supervised approaches typically require the annotation of large training sets; there has thus been great interest in leveraging weakly, semi- or selfsupervised methods to avoid this, with much success. Whilst weakly and semi-supervised methods require some annotation, self-supervised methods have used cues such as motion to relieve the need for annotation altogether. However, a complete absence of annotation typically degrades their performance, and ambiguities that arise during motion grouping can inhibit their ability to find accurate object boundaries. In this paper, we propose a new self-supervised mobile object detection approach called SCT. This uses both motion cues and expected object sizes to improve detection performance, and predicts a dense grid of 3D oriented bounding boxes to improve object discovery. We significantly outperform the state-of-the-art self-supervised mobile object detection method TCR on the KITTI tracking benchmark, and achieve performance that is within 30% of the fully supervised PV-RCNN++ method for IoUs ≤ 0.5. Our source code will be made available online.

Authors: Shin, S; Golodetz, S; Vankadari, M; Zhou, K; Markham, A

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: IEEE
• Host title: 2023 IEEE International Conference on Robotics and Automation (ICRA)
• Pages: 7090-7096
• Publication date: 2023-07-04
• Acceptance date: 2023-01-17
• Event title: 2023 IEEE International Conference on Robotics and Automation (ICRA 2023)
• Event location: London
• Event website: https://www.icra2023.org/
• Event start date: 2023-05-29
• Event end date: 2023-06-02
• DOI: 10.1109/icra48891.2023.10160980
• EISSN: 2577-087X
• ISSN: 1050-4729
• EISBN: 9798350323658
• ISBN: 9798350323665
• Language: English