CORC  > 自动化研究所  > 中国科学院自动化研究所
An End-to-end Video Text Detector with Online Tracking
Yu HY(俞宏远)2,3; Zhang, Chengquan4; Li, Xuan4; Han, Junyu4; Ding, Errui4; Wang L(王亮)1,2,3
2019
会议日期2019年
会议地点悉尼
英文摘要

Video text detection is considered as one of the most difficult tasks in document analysis due to the following two challenges: 1) the difficulties caused by video scenes, i.e., motion blur, illumination changes, and occlusion; 2) the properties of text including variants of fonts, languages, orientations, and shapes. Most existing methods attempt to enhance the
performance of video text detection by cooperating with video text tracking, but treat these two tasks separately. In this work, we propose an end-to-end video text detection model with online tracking to address these two challenges. Specifically, in the detection branch, we adopt ConvLSTM to capture spatial structure information and motion memory. In the tracking branch, we convert the tracking problem to text instance association, and an appearance-geometry descriptor with memory mechanism is proposed to generate robust representation of text instances. By integrating these two branches into one trainable framework, they can promote each other and the computational cost is significantly reduced. Experiments on existing video text benchmarks including ICDAR2013 Video, Minetto and YVT demonstrate that the proposed method significantly outperforms state-of-the-art methods. Our method improves F-score by about 2% on all datasets and it can run
realtime with 24.36 fps on TITAN Xp.
 

会议录出版者IEEE
会议录出版地IEEE
内容类型会议论文
源URL[http://ir.ia.ac.cn/handle/173211/48517]  
专题中国科学院自动化研究所
通讯作者Wang L(王亮)
作者单位1.Chinese Academy of Sciences Artificial Intelligence Research
2.中国科学院大学
3.中国科学院自动化研究所,NLPR,CASIA
4.Department of Computer Vision Technology(VIS), Baidu Inc.
推荐引用方式
GB/T 7714
Yu HY,Zhang, Chengquan,Li, Xuan,et al. An End-to-end Video Text Detector with Online Tracking[C]. 见:. 悉尼. 2019年.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace