Incremental few-shot object detection via knowledge transfer

doi:10.1016/j.patrec.2022.01.024

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 机器人理论与应用

	Incremental few-shot object detection via knowledge transfer
	Feng, Hangtao 1,2; Zhang, Lu1,2 ; Yang, Xu1,2 ; Liu, Zhiyong1,2,3
发表期刊	PATTERN RECOGNITION LETTERS
ISSN	0167-8655
	2022-04-01
卷号	156 页码:67-73
通讯作者	Liu, Zhiyong(zhiyong.liu@ia.ac.cn)
摘要	As a challenging problem in machine learning, incremental few-shot object detection (iFSD) [1] aims to incrementally detect novel classes with few examples, while keeping the previous knowledge without revisiting base classes. Here, we propose two models based on the observation that when new memories come, new connections will be created between memory cells in the brain [2] . The first one, which is called the multi-class head (MCH) model, simulates how humans add new memory-connections that every time novel classes come, a classification branch is added to predict novel classification. And the second model, called the bi-path multi-class head (BPMCH) model, adds a new backbone, which is initialized with the weight of the base class backbone, to transfer more knowledge of the base class to the novel class. Considering accuracy and speed, we choose the Fully Convolutional One-Stage Object Detection (FCOS) [3] + Adaptive Training Sample Selection (ATSS) [4] detector as our baseline. Our models are first trained on the base classes with abundant examples and then finetuned on novel classes with few examples, which not only maintain the knowledge learned from the base class but also transfer the knowledge to the novel class. Extensive experiments show that our models outperform the state-of-theart model ONCE [1] on the COCO [5] and PASCAL VOC [6] by a large margin. (c) 2022 Elsevier B.V. All rights reserved.
关键词	Machine learning Convolutional neural networks Transfer learning Incremental few-shot object detection
DOI	10.1016/j.patrec.2022.01.024
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Plan of China[2020AAA0108902] ; Strategic Priority Research Program of Chinese Academy of Science[XDB32050100] ; NSFC[61627808] ; Dongguan Core Technology Research Frontier Project, China[2019622101001]
项目资助者	National Key Research and Development Plan of China ; Strategic Priority Research Program of Chinese Academy of Science ; NSFC ; Dongguan Core Technology Research Frontier Project, China
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000789226600010
出版者	ELSEVIER
引用统计	被引频次：4[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/49378
专题	多模态人工智能系统全国重点实验室_机器人理论与应用
通讯作者	Liu, Zhiyong
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China 2.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing, Peoples R China 3.Chinese Acad Sci China, Ctr Excellence Brain Sci & Intelligence Technol, Beijing, Peoples R China
第一作者单位	中国科学院自动化研究所
通讯作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Feng, Hangtao,Zhang, Lu,Yang, Xu,et al. Incremental few-shot object detection via knowledge transfer[J]. PATTERN RECOGNITION LETTERS,2022,156:67-73.
APA	Feng, Hangtao,Zhang, Lu,Yang, Xu,&Liu, Zhiyong.(2022).Incremental few-shot object detection via knowledge transfer.PATTERN RECOGNITION LETTERS,156,67-73.
MLA	Feng, Hangtao,et al."Incremental few-shot object detection via knowledge transfer".PATTERN RECOGNITION LETTERS 156(2022):67-73.