A parallel vision approach to scene-specific pedestrian detection

doi:10.1016/j.neucom.2019.03.095

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 先进控制与自动化团队

	A parallel vision approach to scene-specific pedestrian detection
	Zhang, Wenwen 1,3; Wang, Kunfeng 1; Liu, Yating 1,2; Lu, Yue 1,2; Wang, Fei-Yue 1
刊名	NEUROCOMPUTING
	2020-06-21
卷号	394 页码:114-126
关键词	Pedestrian detection Specific scene Synthetic data Video surveillance Parallel vision
ISSN号	0925-2312
DOI	10.1016/j.neucom.2019.03.095
通讯作者	Wang, Kunfeng(kunfeng.wang@ia.ac.cn)
英文摘要	In recent years, with the development of computing power and deep learning algorithms, pedestrian detection has made great progress. Nevertheless, once a detection model trained on generic datasets (such as PASCAL VOC and MS COCO) is applied to a specific scene, its precision is limited by the distribution gap between the generic data and the specific scene data. It is difficult to train the model for a specific scene, due to the lack of labeled data from that scene. Even though we manage to get some labeled data from a specific scene, the changing environmental conditions make the pre-trained model perform bad. In light of these issues, we propose a parallel vision approach to scene-specific pedestrian detection. Given an object detection model, it is trained via two sequential stages: (1) the model is pre-trained on augmented-reality data, to address the lack of scene-specific training data; (2) the pre-trained model is incrementally optimized with newly synthesized data as the specific scene evolves over time. On publicly available datasets, our approach leads to higher precision than the models trained on generic data. To tackle the dynamically changing scene, we further evaluate our approach on the webcam data collected from Church Street Market Place, and the results are also encouraging. (C) 2019 Elsevier B.V. All rights reserved.
资助项目	National Natural Science Foundation of China[61533019] ; National Natural Science Foundation of China[U1811463]
WOS关键词	CAMERA NETWORKS
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000531730600012
资助机构	National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/39436]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_先进控制与自动化团队
通讯作者	Wang, Kunfeng
作者单位	1.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 3.Xi An Jiao Tong Univ, Sch Software Engn, Xian 710049, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Wenwen,Wang, Kunfeng,Liu, Yating,et al. A parallel vision approach to scene-specific pedestrian detection[J]. NEUROCOMPUTING,2020,394:114-126.
APA	Zhang, Wenwen,Wang, Kunfeng,Liu, Yating,Lu, Yue,&Wang, Fei-Yue.(2020).A parallel vision approach to scene-specific pedestrian detection.NEUROCOMPUTING,394,114-126.
MLA	Zhang, Wenwen,et al."A parallel vision approach to scene-specific pedestrian detection".NEUROCOMPUTING 394(2020):114-126.