Monocular contextual constraint for stereo matching with adaptive weights assignment

doi:10.1016/j.imavis.2022.104424

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 遥感图像处理团队

	Monocular contextual constraint for stereo matching with adaptive weights assignment
	Zhang, Chenghao 1,4; Meng, Gaofeng 1,3,4; Su, Bing 2; Xiang, Shiming 1,4; Pan, Chunhong 1
刊名	IMAGE AND VISION COMPUTING
	2022-05-01
卷号	121 页码:10
关键词	Deep learning Stereo matching Monocular contextual constraint Adaptive weights assignment
ISSN号	0262-8856
DOI	10.1016/j.imavis.2022.104424
通讯作者	Meng, Gaofeng(gfmeng@nlpr.ia.ac.cn)
英文摘要	Matching-based stereo disparity estimation has difficulty in dealing with occlusion, weak and repetitive textures in binocular vision. By contrast, monocular vision, estimating depth from a single image, is not subject to these challenges. Inspired by this, in this study, we propose an adaptive co-learning framework with monocular and stereo branches named CLStereo to improve stereo performance. This framework introduces a monocular branch as contextual constraints to transfer the prior knowledge learned from the monocular branch to the stereo branch. An adaptive weights assignment is further proposed to balance the co-learning of both branches without mutually tuning. CLStereo can be seamlessly embedded into many existing deep stereo models to boost their performance, especially in occluded, weak, and repetitive texture areas. Extensive experiments demonstrate that we achieve the state-of-the-art performance on the Scene Flow dataset and improve deep stereo models by at least 4% on KITTI 2012 and 2015 benchmarks. (c) 2022 Elsevier B.V. All rights reserved.
资助项目	National Key Research and De-velopment Program of China[2020AAA0109702] ; National Natural Science Foundation of China[61976208] ; National Natural Science Foundation of China[61802407] ; National Natural Science Foundation of China[62071466]
WOS研究方向	Computer Science ; Engineering ; Optics
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000783099100003
资助机构	National Key Research and De-velopment Program of China ; National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/48344]
专题	自动化研究所_模式识别国家重点实验室_遥感图像处理团队
通讯作者	Meng, Gaofeng
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Renmin Univ China, Gaoling Sch Artificial Intelligence, Beijing 100872, Peoples R China 3.Chinese Acad Sci, HK Inst Sci & Innovat, Ctr Artificial Intelligence & Robot, Hong Kong 999077, Peoples R China 4.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Chenghao,Meng, Gaofeng,Su, Bing,et al. Monocular contextual constraint for stereo matching with adaptive weights assignment[J]. IMAGE AND VISION COMPUTING,2022,121:10.
APA	Zhang, Chenghao,Meng, Gaofeng,Su, Bing,Xiang, Shiming,&Pan, Chunhong.(2022).Monocular contextual constraint for stereo matching with adaptive weights assignment.IMAGE AND VISION COMPUTING,121,10.
MLA	Zhang, Chenghao,et al."Monocular contextual constraint for stereo matching with adaptive weights assignment".IMAGE AND VISION COMPUTING 121(2022):10.