Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR
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...
Main Authors: | , , , , , |
---|---|
Format: | Conference item |
Language: | English |
Published: |
IEEE
2023
|
_version_ | 1797112864468631552 |
---|---|
author | Shin, S Golodetz, S Vankadari, M Zhou, K Markham, A Trigoni, N |
author_facet | Shin, S Golodetz, S Vankadari, M Zhou, K Markham, A Trigoni, N |
author_sort | Shin, S |
collection | OXFORD |
description | 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. |
first_indexed | 2024-03-07T08:29:54Z |
format | Conference item |
id | oxford-uuid:eaaf225f-10bb-43e3-aded-c1beaedb6399 |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-07T08:29:54Z |
publishDate | 2023 |
publisher | IEEE |
record_format | dspace |
spelling | oxford-uuid:eaaf225f-10bb-43e3-aded-c1beaedb63992024-03-04T15:52:34ZSample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDARConference itemhttp://purl.org/coar/resource_type/c_5794uuid:eaaf225f-10bb-43e3-aded-c1beaedb6399EnglishSymplectic ElementsIEEE2023Shin, SGolodetz, SVankadari, MZhou, KMarkham, ATrigoni, NDeep 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. |
spellingShingle | Shin, S Golodetz, S Vankadari, M Zhou, K Markham, A Trigoni, N Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR |
title | Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR |
title_full | Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR |
title_fullStr | Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR |
title_full_unstemmed | Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR |
title_short | Sample, crop, track: self-supervised mobile 3D object detectionfor urban driving LiDAR |
title_sort | sample crop track self supervised mobile 3d object detectionfor urban driving lidar |
work_keys_str_mv | AT shins samplecroptrackselfsupervisedmobile3dobjectdetectionforurbandrivinglidar AT golodetzs samplecroptrackselfsupervisedmobile3dobjectdetectionforurbandrivinglidar AT vankadarim samplecroptrackselfsupervisedmobile3dobjectdetectionforurbandrivinglidar AT zhouk samplecroptrackselfsupervisedmobile3dobjectdetectionforurbandrivinglidar AT markhama samplecroptrackselfsupervisedmobile3dobjectdetectionforurbandrivinglidar AT trigonin samplecroptrackselfsupervisedmobile3dobjectdetectionforurbandrivinglidar |