Text proposals with location-awareness-attention network for arbitrarily shaped scene text detection and recognition
作者：
Highlights：
• An end-to-end unified method is proposed for text detection and recognition.
• Predicting central mask of text instance for locating text region.
• Regression and mask prediction are performed for accurate position of text region.
• A local-awareness network-based model is proposed to recognize characters.
• Experimental results attest to the effectiveness of the proposed method.
摘要
•An end-to-end unified method is proposed for text detection and recognition.•Predicting central mask of text instance for locating text region.•Regression and mask prediction are performed for accurate position of text region.•A local-awareness network-based model is proposed to recognize characters.•Experimental results attest to the effectiveness of the proposed method.
论文关键词：Scene text detection,Scene text recognition,Text proposal,Attention model,Location-awareness-attention model
论文评审过程：Received 2 September 2021, Revised 3 May 2022, Accepted 8 May 2022, Available online 19 May 2022, Version of Record 2 June 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.117564
