Symmetry-constrained Rectification Network for Scene Text Recognition

TL;DR

This paper introduces a symmetry-constrained rectification network that leverages local text attributes to effectively rectify irregular scene text, significantly improving recognition accuracy on challenging datasets.

Contribution

The proposed Symmetry-constrained Rectification Network (ScRN) utilizes local shape attributes to better rectify irregular text, outperforming existing methods especially on highly distorted text instances.

Findings

Achieves state-of-the-art performance on irregular text datasets

Outperforms existing algorithms on ICDAR 2015, SVT-Perspective, and CUTE80

Significantly improves recognition accuracy for curved and distorted text

Abstract

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.