Learning Domain Invariant Features for Unsupervised Indoor Depth Estimation Adaptation

doi:10.1145/3672397

	Learning Domain Invariant Features for Unsupervised Indoor Depth Estimation Adaptation
	Zhang, Jiehua 1; Li, Liang 2; Yan, Chenggang 3; Wang, Zhan 4; Xu, Changliang 5,6; Zhang, Jiyong 3; Chen, Chuqiao 7
	2024-09-01
发表期刊	ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
ISSN	1551-6857
卷号	20 期号:9 页码:23
摘要	Predicting depth maps from monocular images has made an impressive performance in the past years. However, most depth estimation methods are trained with paired image-depth map data or multi-view images (e.g., stereo pair and monocular sequence), which suffer from expensive annotation costs and poor transferability. Although unsupervised domain adaptation methods are introduced to mitigate the reliance on annotated data, rare works focus on the unsupervised cross-scenario indoor monocular depth estimation. In this article, we propose to study the generalization of depth estimation models across different indoor scenarios in an adversarial-based domain adaptation paradigm. Concretely, a domain discriminator is designed for discriminating the representation from source and target domains, while the feature extractor aims to confuse the domain discriminator by capturing domain-invariant features. Further, we reconstruct depth maps from latent representations with the supervision of labeled source data. As a result, the feature extractor learned features possess the merit of both domain-invariant and low source risk, and the depth estimator can deal with the domain shift between source and target domains. We conduct the cross-scenario and cross-dataset experiments on the ScanNet and NYU-Depth-v2 datasets to verify the effectiveness of our method and achieve impressive performance.
关键词	Indoor depth estimation unsupervised learning transfer learning domain adaptation
DOI	10.1145/3672397
收录类别	SCI
语种	英语
WOS研究方向	Computer Science
WOS类目	Computer Science, Information Systems ; Computer Science, Software Engineering ; Computer Science, Theory & Methods
WOS记录号	WOS:001325876400003
出版者	ASSOC COMPUTING MACHINERY
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/39556
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Li, Liang; Yan, Chenggang
作者单位	1.Xi An Jiao Tong Univ, Xian, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 3.Hangzhou Dianzi Univ, Hangzhou, Peoples R China 4.Moreal Pte Ltd, Singapore, Singapore 5.State Key Lab Media Convergence Prod Technol Syst, Beijing, Peoples R China 6.Xinhua Zhiyun Technol Co Ltd, Beijing, Peoples R China 7.Hangzhou Dianzi Univ, Lishui Inst, Hangzhou, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Jiehua,Li, Liang,Yan, Chenggang,et al. Learning Domain Invariant Features for Unsupervised Indoor Depth Estimation Adaptation[J]. ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS,2024,20(9):23.
APA	Zhang, Jiehua.,Li, Liang.,Yan, Chenggang.,Wang, Zhan.,Xu, Changliang.,...&Chen, Chuqiao.(2024).Learning Domain Invariant Features for Unsupervised Indoor Depth Estimation Adaptation.ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS,20(9),23.
MLA	Zhang, Jiehua,et al."Learning Domain Invariant Features for Unsupervised Indoor Depth Estimation Adaptation".ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS 20.9(2024):23.