Frequency-based pseudo-domain generation for domain generalizable object detection

doi:10.1016/j.neucom.2023.126265

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多模态人工智能系统全国重点实验室

	Frequency-based pseudo-domain generation for domain generalizable object detection
	Zhang, Siqi 1,3; Zhang, Lu 3; Liu, Zhi-Yong 1,2,3
刊名	NEUROCOMPUTING
	2023-07-14
卷号	542 页码:12
关键词	Domain generalization Object detection Transfer learning Self-Supervised learning
ISSN号	0925-2312
DOI	10.1016/j.neucom.2023.126265
通讯作者	Liu, Zhi-Yong(zhiyong.liu@ia.ac.cn)
英文摘要	Domain generalizable object detection (DGOD) aims to train a detector that performs well on multiple unseen target domains, which is crucial for deploying the detector in practice. Recent methods for DGOD typically inherit the idea from domain adaptation to align or disentangle features, but these meth-ods struggle to handle unknown target distributions. In this paper, we propose a unified framework to tackle the DGOD task from a novel pseudo-domain generation perspective. Our framework comprises two stages: distribution diversification and domain-invariant feature learning. In the distribution diver-sification stage, we design a Frequency-based Pseudo-domain Generator (FPG) to construct the pseudo domain via excavating latent style information and enhancing semantic information in frequency space. The generated pseudo domain can provide diverse training distributions, which enhances generalization performance. In the domain-invariant feature learning stage, we introduce Rotation Prediction and Semantic Consistency (RPSC) learning, including an auxiliary self-supervised task rotation prediction to encourage generalized feature learning and a semantic consistency loss to enforce the detector to be invariant of domain shifts. Extensive experiments are conducted on various object detection benchmarks, demonstrating the superiority of our approach over state-of-the-art methods in both single-source and multi-source settings.(c) 2023 Elsevier B.V. All rights reserved.
资助项目	National Key Research and Development Plan of China[2020AAA0108902] ; Strategic Priority Research Program of Chinese Acad- emy of Sciences[XDB32050100] ; NSFC[62206288]
WOS关键词	ADAPTATION
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:001003715300001
资助机构	National Key Research and Development Plan of China ; Strategic Priority Research Program of Chinese Acad- emy of Sciences ; NSFC
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/53420]
专题	多模态人工智能系统全国重点实验室
通讯作者	Liu, Zhi-Yong
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 2.Nanjing Artificial Intelligence Res IA, Nanjing 211100, Peoples R China 3.Chinese Acad Sci, Inst Automat, State Key Lab Multimodal Artificial Intelligence S, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Siqi,Zhang, Lu,Liu, Zhi-Yong. Frequency-based pseudo-domain generation for domain generalizable object detection[J]. NEUROCOMPUTING,2023,542:12.
APA	Zhang, Siqi,Zhang, Lu,&Liu, Zhi-Yong.(2023).Frequency-based pseudo-domain generation for domain generalizable object detection.NEUROCOMPUTING,542,12.
MLA	Zhang, Siqi,et al."Frequency-based pseudo-domain generation for domain generalizable object detection".NEUROCOMPUTING 542(2023):12.