# MTRNet: A Generic Scene Text Eraser

## TL;DR

MTRNet is a versatile scene text eraser that uses a mask-based cGAN to effectively remove multilingual and curved text in real scenes, achieving state-of-the-art results without dataset-specific training.

## Contribution

The paper introduces MTRNet, a generic text removal network utilizing an auxiliary mask with a cGAN, enabling stable training and broad applicability across languages and text shapes.

## Key findings

- Achieves state-of-the-art results on multiple real-world datasets.
- Operates effectively without dataset-specific training.
- Handles multilingual and curved text in complex scenes.

## Abstract

Text removal algorithms have been proposed for uni-lingual scripts with regular shapes and layouts. However, to the best of our knowledge, a generic text removal method which is able to remove all or user-specified text regions regardless of font, script, language or shape is not available. Developing such a generic text eraser for real scenes is a challenging task, since it inherits all the challenges of multi-lingual and curved text detection and inpainting. To fill this gap, we propose a mask-based text removal network (MTRNet). MTRNet is a conditional adversarial generative network (cGAN) with an auxiliary mask. The introduced auxiliary mask not only makes the cGAN a generic text eraser, but also enables stable training and early convergence on a challenging large-scale synthetic dataset, initially proposed for text detection in real scenes. What's more, MTRNet achieves state-of-the-art results on several real-world datasets including ICDAR 2013, ICDAR 2017 MLT, and CTW1500, without being explicitly trained on this data, outperforming previous state-of-the-art methods trained directly on these datasets.