Look Back Again
Authors:
Open Access
- 24 August 2021
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 638-644
Abstract
Nowadays, it is a trend that using a parallel-decoupled encoder-decoder (PDED) framework in scene text recognition for its flexibility and efficiency. However, due to the inconsistent information content between queries and keys in the parallel positional attention module (PPAM) used in this kind of framework(queries: position information, keys: context and position information), visual misalignment tends to appear when confronting hard samples(e.g., blurred texts, irregular texts, or low-quality images). To tackle this issue, in this paper, we propose a dual parallel attention network (DPAN), in which a newly designed parallel context attention module (PCAM) is cascaded with the original PPAM, using linguistic contextual information to compensate for the information inconsistency between queries and keys. Specifically, in PCAM, we take the visual features from PPAM as inputs and present a bidirectional language model to enhance them with linguistic contexts to produce queries. In this way, we make the information content of the queries and keys consistent in PCAM, which helps to generate more precise visual glimpses to improve the entire PDED framework's accuracy and robustness. Experimental results verify the effectiveness of the proposed PCAM, showing the necessity of keeping the information consistency between queries and keys in the attention mechanism. On six benchmarks, including regular text and irregular text, the performance of DPAN surpasses the existing leading methods by large margins, achieving new state-of-the-art performance. The code is available on \urlhttps://github.com/Jackandrome/DPAN.Keywords
Funding Information
- the Fundamental Research Funds for the Central Universities under Grant (WK3480000011)
- the national key Research and Development Program of China (2018YFB0804203)
- the Youth Innovation Promotion Association Chinese Academy of Sciences (2017209)
- the National Nature Science Foundation of China (62022076, U1936210, 62032006)
This publication has 24 references indexed in Scilit:
- Focusing Attention: Towards Accurate Text Recognition in Natural Images2017
- Fully Convolutional Instance-Aware Semantic Segmentation2017
- Deep Residual Learning for Image Recognition2016
- Robust Scene Text Recognition with Automatic Rectification2016
- Synthetic Data for Text Localisation in Natural Images2016
- Reading Text in the Wild with Convolutional Neural NetworksInternational Journal of Computer Vision, 2015
- A robust arbitrary text detection system for natural scene imagesExpert Systems with Applications, 2014
- Recognizing Text with Perspective Distortion in Natural Scenes2013
- Scene Text Recognition using Higher Order Language Priors2012
- Connectionist temporal classification2006