Selective Scene Text Removal

TL;DR

This paper introduces a new task called selective scene text removal (SSTR), allowing users to specify and remove only target words in scene images, advancing beyond traditional methods that remove all scene text.

Contribution

The paper proposes a novel multi-module framework for efficient training and implementation of selective scene text removal, enabling targeted text removal in images.

Findings

Successfully removes specified target words in scene images

Efficient training achieved with the proposed multi-module structure

Method outperforms conventional all-text removal approaches

Abstract

Scene text removal (STR) is the image transformation task to remove text regions in scene images. The conventional STR methods remove all scene text. This means that the existing methods cannot select text to be removed. In this paper, we propose a novel task setting named selective scene text removal (SSTR) that removes only target words specified by the user. Although SSTR is a more complex task than STR, the proposed multi-module structure enables efficient training for SSTR. Experimental results show that the proposed method can remove target words as expected.