Stroke-Based Scene Text Erasing Using Synthetic Data for Training

TL;DR

This paper introduces a synthetic-data-trained scene text erasing method with a stroke mask prediction and background inpainting, achieving superior results on multiple datasets without relying on large real-world datasets.

Contribution

The authors propose a novel synthetic data generation approach and a specialized network architecture for scene text erasing, outperforming existing methods trained on real data.

Findings

Outperforms state-of-the-art methods on multiple datasets

Effective training with synthetic data alone

Partial text erasing with bounding box or detector input

Abstract

Scene text erasing, which replaces text regions with reasonable content in natural images, has drawn significant attention in the computer vision community in recent years. There are two potential subtasks in scene text erasing: text detection and image inpainting. Both subtasks require considerable data to achieve better performance; however, the lack of a large-scale real-world scene-text removal dataset does not allow existing methods to realize their potential. To compensate for the lack of pairwise real-world data, we made considerable use of synthetic text after additional enhancement and subsequently trained our model only on the dataset generated by the improved synthetic text engine. Our proposed network contains a stroke mask prediction module and background inpainting module that can extract the text stroke as a relatively small hole from the cropped text image to maintain more background content for better inpainting results. This model can partially erase text instances in a scene image with a bounding box or work with an existing scene-text detector for automatic scene text erasing. The experimental results from the qualitative and quantitative evaluation on the SCUT-Syn, ICDAR2013, and SCUT-EnsText datasets demonstrate that our method significantly outperforms existing state-of-the-art methods even when they are trained on real-world data.