Inverse Scene Text Removal

TL;DR

This paper explores inverse scene text removal (ISTR), focusing on detecting whether images have undergone text removal and localizing such regions, with high accuracy and attempts to recover the removed text content.

Contribution

It introduces ISTR tasks for detecting and localizing removed text regions, and investigates text recovery, enhancing misuse detection and understanding of STR processes.

Findings

High accuracy in detecting STR-processed images

Effective localization of removed text regions

Initial success in recovering removed text content

Abstract

Scene text removal (STR) aims to erase textual elements from images. It was originally intended for removing privacy-sensitiveor undesired texts from natural scene images, but is now also appliedto typographic images. STR typically detects text regions and theninpaints them. Although STR has advanced through neural networksand synthetic data, misuse risks have increased. This paper investi-gates Inverse STR (ISTR), which analyzes STR-processed images andfocuses on binary classification (detecting whether an image has un-dergone STR) and localizing removed text regions. We demonstrate inexperiments that these tasks are achievable with high accuracies, en-abling detection of potential misuse and improving STR. We also at-tempt to recover the removed text content by training a text recognizerto understand its difficulty.