Symmetry-Constrained Rectification Network for Scene Text Recognition

By Mingkun Yang, Yushuo Guan, Minghui Liao, Xin He, Kaigui Bian, Song Bai, Cong Yao, Xiang Bai. 2019

137 citations recorded by the source corpus.

Reading text in the wild is a very challenging task due to the diversity of text instances and the complexity of natural scenes. Recently, the community has paid increasing attention to the problem of recognizing text instances with irregular shapes. One intuitive and effective way to handle this problem is to rectify irregular text to a canonical form before recognition. However, these methods might struggle when dealing with highly curved or distorted text instances. To tackle this issue, we propose in this paper a Symmetry-constrained Rectification Network (ScRN) based on local attributes of text instances, such as center line, scale and orientation. Such constraints with an accurate description of text shape enable ScRN to generate better rectification results than existing methods and thus lead to higher recognition accuracy. Our method achieves state-of-the-art performance on text with both regular and irregular shapes. Specifically, the system outperforms existing algorithms by a large margin on datasets that contain quite a proportion of irregular text instances, e.g., ICDAR 2015, SVT-Perspective and CUTE80.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Errui Ding
2. Xiang Bai
3. Junyu Han
4. Chengquan Zhang
5. Xuan Li
6. Wenyu Liu
7. Deli Yu
8. Mingkun Yang

In Computer Vision and Pattern Recognition

1. Mingkun Yang
2. Chengquan Zhang
3. Errui Ding
4. Yongchao Xu
5. Junyu Han
6. Xiang Bai
7. Wenyu Liu
8. Jingtuo Liu

Citations in the corpus

• All You Need Is Boundary: Toward Arbitrary-Shaped Text Spotting (2019, background): “In some latest works (Zhan and Lu 2019; Yang et al. 2019; Luo, Jin, and Sun 2019), recognition network is a common attentional sequence-to-sequence network.”
• Towards Accurate Scene Text Recognition With Semantic Reasoning Networks (2020, background): “Most of the recent works have attempted to improve the performance of scene text recognition from the perspective of extracting more robust and effective visual features, such as upgrading the backbone networks [5, 21, 31], adding rectification modules [30, 31, 40, 44] and improving attention mechan”
• Scene Text Detection and Recognition: The Deep Learning Era (2018, unknown): “Yang et al. (2019) propose to predict lo-”