PERT: A Progressively Region-based Network for Scene Text Removal

TL;DR

PERT introduces a region-based, multi-stage approach for scene text removal that explicitly guides erasure, achieving state-of-the-art results with high speed and lower complexity by focusing on targeted text regions.

Contribution

The paper proposes PERT, a novel multi-stage, region-guided scene text removal network that improves accuracy and exhaustiveness without relying on adversarial training.

Findings

Outperforms previous methods significantly in accuracy.

Operates at 71 FPS, ensuring high efficiency.

Reduces parameter complexity by at least 25%.

Abstract

Scene text removal (STR) contains two processes: text localization and background reconstruction. Through integrating both processes into a single network, previous methods provide an implicit erasure guidance by modifying all pixels in the entire image. However, there exists two problems: 1) the implicit erasure guidance causes the excessive erasure to non-text areas; 2) the one-stage erasure lacks the exhaustive removal of text region. In this paper, we propose a ProgrEssively Region-based scene Text eraser (PERT), introducing an explicit erasure guidance and performing balanced multi-stage erasure for accurate and exhaustive text removal. Firstly, we introduce a new region-based modification strategy (RegionMS) to explicitly guide the erasure process. Different from previous implicitly guided methods, RegionMS performs targeted and regional erasure on only text region, and adaptively perceives stroke-level information to improve the integrity of non-text areas with only bounding box level annotations. Secondly, PERT performs balanced multi-stage erasure with several progressive erasing stages. Each erasing stage takes an equal step toward the text-erased image to ensure the exhaustive erasure of text regions. Compared with previous methods, PERT outperforms them by a large margin without the need of adversarial loss, obtaining SOTA results with high speed (71 FPS) and at least 25% lower parameter complexity. Code is available at https://github.com/wangyuxin87/PERT.