ReinforceNet: A reinforcement learning embedded object detection framework with region selection network

doi:10.1016/j.neucom.2021.02.073

	ReinforceNet: A reinforcement learning embedded object detection framework with region selection network
	Zhou, Man 1,2; Wang, Rujing 2; Xie, Chengjun 2; Liu, Liu 1,2; Li, Rui 1,2; Wang, Fangyuan 1,2; Li, Dengshan 1,2
刊名	NEUROCOMPUTING
	2021-07-05
卷号	443
关键词	Reinforcement learning Convolutional neural network Region selection network Object detection
ISSN号	0925-2312
DOI	10.1016/j.neucom.2021.02.073
通讯作者	Xie, Chengjun(cjxie@iim.ac.cn)
英文摘要	In recent years, researchers have explored reinforcement learning based object detection methods. However, existing methods always suffer from barely satisfactory performance. The main reasons are that current reinforcement learning based methods generate a sequence of inaccurate regions without a reasonable reward function, and regard the non-optimal one at the final step as the detection result for lack of an effective region selection and refinement strategy. To tackle the above problems, we propose a novel reinforcement learning based object detection framework, namely ReinforceNet, possessing the capability of the region selection and refinement by integrating reinforcement learning agents & rsquo; action space with Convolutional Neural Network based feature space. In ReinforceNet, we redevelop a reward function that enables the agent to be trained effectively and provide more accurate region proposals. In order to further optimize them, we design Convolutional Neural Network based region selection network (RS-net) and bounding box refinement network (BBR-net). Particularly, the former consists of two sub-networks: Intersection-of-Union network (IoU-net) and Completeness network (CPL-net) which are employed jointly for selecting the optimal region proposal. The latter aims to refine the selected one as the final result. Extensive experimental results on two standard datasets PASCAL VOC 2007 and VOC 2012 demonstrate that ReinforceNet is capable of improving the region selection and learning better agent action representations for reinforcement learning, resulting in the state-of-the-art performance. (c) 2021 Elsevier B.V. All rights reserved.
资助项目	National Natural Science Foundation of China[31401293] ; National Natural Science Foundation of China[31671586] ; National Natural Science Foundation of China[61773360]
WOS关键词	NEURAL-NETWORKS
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000647206400011
资助机构	National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.hfcas.ac.cn:8080/handle/334002/122182]
专题	中国科学院合肥物质科学研究院
通讯作者	Xie, Chengjun
作者单位	1.Univ Sci & Technol China, Hefei 230026, Peoples R China 2.Chinese Acad Sci, Hefei Inst Phys Sci, Inst Intelligent Machines, Hefei 230031, Peoples R China
推荐引用方式 GB/T 7714	Zhou, Man,Wang, Rujing,Xie, Chengjun,et al. ReinforceNet: A reinforcement learning embedded object detection framework with region selection network[J]. NEUROCOMPUTING,2021,443.
APA	Zhou, Man.,Wang, Rujing.,Xie, Chengjun.,Liu, Liu.,Li, Rui.,...&Li, Dengshan.(2021).ReinforceNet: A reinforcement learning embedded object detection framework with region selection network.NEUROCOMPUTING,443.
MLA	Zhou, Man,et al."ReinforceNet: A reinforcement learning embedded object detection framework with region selection network".NEUROCOMPUTING 443(2021).