Bidirectional Regression for Arbitrary-Shaped Text Detection

TL;DR

This paper introduces a novel bidirectional regression method for arbitrary-shaped text detection that effectively utilizes context information and boundary pixels, leading to improved accuracy on challenging scene text benchmarks.

Contribution

It proposes a new text instance expression integrating foreground and background info, with a post-processing algorithm for precise text instance reconstruction.

Findings

Achieves 83.4% F-score on Total-Text

Achieves 82.4% F-score on MSRA-TD500

Outperforms or matches state-of-the-art methods

Abstract

Arbitrary-shaped text detection has recently attracted increasing interests and witnessed rapid development with the popularity of deep learning algorithms. Nevertheless, existing approaches often obtain inaccurate detection results, mainly due to the relatively weak ability to utilize context information and the inappropriate choice of offset references. This paper presents a novel text instance expression which integrates both foreground and background information into the pipeline, and naturally uses the pixels near text boundaries as the offset starts. Besides, a corresponding post-processing algorithm is also designed to sequentially combine the four prediction results and reconstruct the text instance accurately. We evaluate our method on several challenging scene text benchmarks, including both curved and multi-oriented text datasets. Experimental results demonstrate that the proposed approach obtains superior or competitive performance compared to other state-of-the-art methods, e.g., 83.4% F-score for Total-Text, 82.4% F-score for MSRA-TD500, etc.