Realtime multi-scale scene text detection with scale-based region proposal network

doi:10.1016/j.patcog.2019.107026

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 模式分析与学习团队

	Realtime multi-scale scene text detection with scale-based region proposal network
	He, Wenhao 1,2; Zhang, Xu-Yao 1,2; Yin, Fei 1,2; Luo, Zhenbo 5; Ogier, Jean-Marc 4; Liu, Cheng-Lin 1,2,3
刊名	PATTERN RECOGNITION
	2020-02-01
卷号	98 页码:14
关键词	Scene text detection Multi-scale Speedup Scale-based region proposal network
ISSN号	0031-3203
DOI	10.1016/j.patcog.2019.107026
通讯作者	Liu, Cheng-Lin(liucl@nlpr.ia.ac.cn)
英文摘要	Multi-scale approaches have been widely used for achieving high accuracy for scene text detection, but they usually slow down the speed of the whole system. In this paper, we propose a two-stage framework for realtime multi-scale scene text detection. The first stage employs a novel Scale-based Region Proposal Network (SRPN) which can localize text of wide scale range and estimate text scale efficiently. Based on SRPN, non-text regions are filtered out, and text region proposals are generated. Moreover, based on text scale estimation by SRPN, small or big texts in region proposals are resized into a unified normal scale range. The second stage then adopts a Fully Convolutional Network based scene text detector to localize text words from proposals of the first stage. Text detector in the second stage detects texts of narrow scale range but accurately. Since most non-text regions are eliminated through SRPN efficiently, and texts in proposals are properly scaled to avoid multi-scale pyramid processing, the whole system is quite fast. We evaluate both performance and speed of the proposed method on datasets ICDAR2015, ICDAR2013, and MSRA-TD500. On ICDAR2015, our system can reach the state-of-the-art F-measure score of 85.40% at 16.5 fps (frame per second), and competitive performance of 79.66% at 35.1 fps, either of which is more than 5 times faster than previous best methods. On ICDAR2013 and MSRA-TD500, we also achieve remarkable speedup by keeping competitive performance. Ablation experiments are also provided to demonstrate the reasonableness of our method. (C) 2019 Elsevier Ltd. All rights reserved.
资助项目	National Natural Science Foundation of China (NSFC)[61721004] ; National Natural Science Foundation of China (NSFC)[61411136002] ; National Natural Science Foundation of China (NSFC)[61733007] ; National Natural Science Foundation of China (NSFC)[61633021] ; NVIDIA NVAIL program
WOS关键词	VIDEO
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	ELSEVIER SCI LTD
WOS记录号	WOS:000497600300013
资助机构	National Natural Science Foundation of China (NSFC) ; NVIDIA NVAIL program
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/29381]
专题	自动化研究所_模式识别国家重点实验室_模式分析与学习团队
通讯作者	Liu, Cheng-Lin
作者单位	1.UCAS, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Inst Automat, NLPR, Beijing 100190, Peoples R China 3.CAS Ctr Excellence Brain Sci & Intelligence Techn, Beijing 100190, Peoples R China 4.Univ La Rochelle, Lab L3i, La Rochelle, France 5.Beijing Samsung Telecom R&D Ctr, Beijing, Peoples R China
推荐引用方式 GB/T 7714	He, Wenhao,Zhang, Xu-Yao,Yin, Fei,et al. Realtime multi-scale scene text detection with scale-based region proposal network[J]. PATTERN RECOGNITION,2020,98:14.
APA	He, Wenhao,Zhang, Xu-Yao,Yin, Fei,Luo, Zhenbo,Ogier, Jean-Marc,&Liu, Cheng-Lin.(2020).Realtime multi-scale scene text detection with scale-based region proposal network.PATTERN RECOGNITION,98,14.
MLA	He, Wenhao,et al."Realtime multi-scale scene text detection with scale-based region proposal network".PATTERN RECOGNITION 98(2020):14.