Curved scene text detection via transverse and longitudinal sequence connection
作者：
Highlights：
• We proposed a new curved dataset to facilitate curved detection method, whose annotation is based on relative objective method, which is very accurate.
• We proposed a novel CTD method that can effectively detect both curved and non-curved text.
• Seamless integration of a RNN method (TLOC) to significantly improve detection performance.
• Implementation of polygonal post-processing methods (NPS and PNMS) to further improve results.
• Our method achieves state-of-the-art performance on curved and non-curved datasets.
摘要
•We proposed a new curved dataset to facilitate curved detection method, whose annotation is based on relative objective method, which is very accurate.•We proposed a novel CTD method that can effectively detect both curved and non-curved text.•Seamless integration of a RNN method (TLOC) to significantly improve detection performance.•Implementation of polygonal post-processing methods (NPS and PNMS) to further improve results.•Our method achieves state-of-the-art performance on curved and non-curved datasets.
论文关键词：Scene text,Curved dataset,LSTM,CNN,Deep learning
论文评审过程：Received 17 March 2018, Revised 17 January 2019, Accepted 4 February 2019, Available online 5 February 2019, Version of Record 8 February 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.02.002
