Foregroundness-Aware Task Disentanglement and Self-Paced Curriculum Learning for Domain Adaptive Object Detection

doi:10.1109/TNNLS.2023.3331778

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

	Foregroundness-Aware Task Disentanglement and Self-Paced Curriculum Learning for Domain Adaptive Object Detection
	Liu, Yabo 1,2; Wang, Jinghua 3; Xiao, Linhui 2,4; Liu, Chengliang 1; Wu, Zhihao 1; Xu, Yong 1,2
刊名	IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
	2023-11-28
页码	12
关键词	Curriculum learning domain adaptive object detection location metric task disentanglement
ISSN号	2162-237X
DOI	10.1109/TNNLS.2023.3331778
通讯作者	Wang, Jinghua(wangjh2012@foxmail.com) ; Xu, Yong(yongxu@ymail.com)
英文摘要	Unsupervised domain adaptive object detection (UDA-OD) is a challenging problem since it needs to locate and recognize objects while maintaining the generalization ability across domains. Most existing UDA-OD methods directly integrate the adaptive modules into the detectors. This integration procedure can significantly sacrifice the detection performances, though it enhances the generalization ability. To solve this problem, we propose an effective framework, named foregroundness-aware task disentanglement and self-paced curriculum adaptation (FA-TDCA), to disentangle the UDA-OD task into four independent subtasks of source detector pretraining, classification adaptation, location adaptation, and target detector training. The disentanglement can transfer the knowledge effectively while maintaining the detection performance of our model. In addition, we propose a new metric, i.e., foregroundness, and use it to evaluate the confidence of the location result. We use both foregroundness and classification confidence to assess the label quality of the proposals. For effective knowledge transfer across domains, we utilize a self-paced curriculum learning paradigm to train adaptors and gradually improve the quality of the pseudolabels associated with the target samples. Experiment results indicate that our method achieves state-of-the-art results on four cross-domain object detection tasks.
资助项目	Natural Science Foundation China (NSFC)
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:001120244700001
资助机构	Natural Science Foundation China (NSFC)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/55063]
专题	多模态人工智能系统全国重点实验室
通讯作者	Wang, Jinghua; Xu, Yong
作者单位	1.Harbin Inst Technol, Shenzhen Key Lab Visual Object Detect & Recognit, Shenzhen 518055, Peoples R China 2.Peng Cheng Lab, Shenzhen 518055, Peoples R China 3.Harbin Inst Technol, Sch Comp Sci & Technol, Shenzhen 518055, Peoples R China 4.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Liu, Yabo,Wang, Jinghua,Xiao, Linhui,et al. Foregroundness-Aware Task Disentanglement and Self-Paced Curriculum Learning for Domain Adaptive Object Detection[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2023:12.
APA	Liu, Yabo,Wang, Jinghua,Xiao, Linhui,Liu, Chengliang,Wu, Zhihao,&Xu, Yong.(2023).Foregroundness-Aware Task Disentanglement and Self-Paced Curriculum Learning for Domain Adaptive Object Detection.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,12.
MLA	Liu, Yabo,et al."Foregroundness-Aware Task Disentanglement and Self-Paced Curriculum Learning for Domain Adaptive Object Detection".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2023):12.