A Multi-Task Learning Approach for Stereo Depth Estimation
Jin C(金晨)1,2; Luan DJ(栾德杰)3,4; Lei Z(雷峥)1; Yang GD(杨国栋)1; Li E(李恩)1
2024
Conference Name第36届中国控制与决策会议
Conference Date2024年5月25日-5月27日
Conference Place西安
PublisherIEEE
Abstract

Due to the uneven depth scale in the linear disparity space, calculating stereo disparity and subsequently converting it into depth, although widely used, leads to nonlinear error amplification. Moreover, the limited availability of depth annotations in stereo datasets has impeded the progress of end-to-end depth estimation techniques. This paper introduces a semi-supervised method for depth estimation using a multi-task network. The multi-task network comprises two branches for disparity estimation and depth estimation. During training, it leverages a pre-trained stereo disparity network to provide dense depth self-supervision, expediting the training of the depth branch. This network offers an effective solution to the scarcity of stereo depth datasets and the sparsity of depth information in point cloud annotations. The efficacy of the algorithm is validated on the KITTI 3D object dataset using sparse point clouds as depth annotations, showcasing remarkable depth estimation capabilities. Additionally, the paper transforms obtained depth maps into pseudo-lidar for 3D object detection, achieving promising results on the KITTI dataset.

Indexed ByEI
Language英语
Sub direction classification智能控制
planning direction of the national heavy laboratory视觉信息处理
Paper associated data
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/57494
Collection中科院工业视觉智能装备工程实验室_精密感知与控制
Corresponding AuthorYang GD(杨国栋)
Affiliation1.中国科学院工业视觉智能装备技术工程实验室,中国科学院自动化研究所
2.中国科学院大学人工智能学院
3.北京交通大学
4.中国铁道科学研究院
First Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Corresponding Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Jin C,Luan DJ,Lei Z,et al. A Multi-Task Learning Approach for Stereo Depth Estimation[C]:IEEE,2024.
Files in This Item: Download All
File Name/Size DocType Version Access License
CCDC.pdf(5065KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Jin C(金晨)]'s Articles
[Luan DJ(栾德杰)]'s Articles
[Lei Z(雷峥)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Jin C(金晨)]'s Articles
[Luan DJ(栾德杰)]'s Articles
[Lei Z(雷峥)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Jin C(金晨)]'s Articles
[Luan DJ(栾德杰)]'s Articles
[Lei Z(雷峥)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: CCDC.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.