arrow
返回

Semantic-aware self-supervised depth estimation for stereo 3D detection

delete2023-03-01
delete2
PRE
AI
H
Hanqing Sun
J
Jiale Cao
Y
Yanwei Pang *
DOI:10.1016/j.patrec.2023.02.006delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Besides the 3D object supervision, the auxiliary disparity supervision is usually indispensable when train-ing a stereo-based 3D object detector. The disparity supervision is either transformed from LiDAR points or generated from pre-trained models. However, the former suffers from the high cost and over-sensitivity to airborne particles of LiDAR devices, and the latter from the limited cross-dataset transferability of con-temporary stereo matching models. To alleviate those problems, we propose a self-supervision frame-work for stereo-based 3D detection that relies on neither LiDARs nor external models. A Depth-based Self-supervision (DSelf) is proposed to unify the coordinate spaces of self-supervised losses and detection into a 3D space. However, the DSelf supervision is dense compared with the sparse LiDAR points, which introduces redundancy and irrelevancy into the stereo features. A Semantic-Aware Sampler (SASampler) is proposed to address the problems by an unbalanced sampling of foreground and background pixels. Combining our SASampler and DSelf supervision, the resultant detector (named S3D) achieves state-of-the-art detection results without explicit disparity supervisions. (c) 2023 Published by Elsevier B.V.
Keyword:
Stereo vision
3D object detection
Self-supervised learning
Depth estimation

期刊

Pattern Recognition Letters 封面图
Pattern Recognition Letters
IF:
3.3
论文数:
8.0K
被引数:
1.6W

机构

T
tianjin university
学者数:
8.0W
论文数: 5.8W
被引数: 88
引用论文

引用论文

Self-Monitoring of Blood Glucose in Type 2 Diabetes: Recent Studies
err2013-03-01
err0
errOAAI
errOliver Schnell; Hasan Alawi; Tadej Battelino; Antonio Ceriello; Peter Diem; Anne-Marie Felton; Wladyslaw Grzeszczak; Kari Harno; Peter Kempler; Ilhan Satman; Bruno Vergès
err分享
err收藏
err分享
err收藏
err分享
err收藏
Efficient binocular stereo correspondence matching with 1-D Max -Trees
err2020-07-01
err5
errOAAI
errBrandt, Rafael; Strisciuglio, Nicola; Petkov, Nicolai; Wilkinson, Michael H. F.
err分享
err收藏
没有更多内容