GlyphMastero: A Glyph Encoder for High-Fidelity Scene Text Editing

TL;DR

GlyphMastero introduces a glyph encoder with cross-level and multi-scale modeling to enhance high-fidelity scene text editing, especially for complex characters, by guiding diffusion models with stroke-level precision.

Contribution

It presents a novel glyph attention module and feature pyramid network to better capture hierarchical text structures for improved scene text editing.

Findings

Achieves 18.02% higher sentence accuracy than previous methods.

Reduces text-region Fréchet inception distance by 53.28%.

Enhances stroke-level precision in multi-lingual scene text editing.

Abstract

Scene text editing, a subfield of image editing, requires modifying texts in images while preserving style consistency and visual coherence with the surrounding environment. While diffusion-based methods have shown promise in text generation, they still struggle to produce high-quality results. These methods often generate distorted or unrecognizable characters, particularly when dealing with complex characters like Chinese. In such systems, characters are composed of intricate stroke patterns and spatial relationships that must be precisely maintained. We present GlyphMastero, a specialized glyph encoder designed to guide the latent diffusion model for generating texts with stroke-level precision. Our key insight is that existing methods, despite using pretrained OCR models for feature extraction, fail to capture the hierarchical nature of text structures - from individual strokes to stroke-level interactions to overall character-level structure. To address this, our glyph encoder explicitly models and captures the cross-level interactions between local-level individual characters and global-level text lines through our novel glyph attention module. Meanwhile, our model implements a feature pyramid network to fuse the multi-scale OCR backbone features at the global-level. Through these cross-level and multi-scale fusions, we obtain more detailed glyph-aware guidance, enabling precise control over the scene text generation process. Our method achieves an 18.02\% improvement in sentence accuracy over the state-of-the-art multi-lingual scene text editing baseline, while simultaneously reducing the text-region Fr\'echet inception distance by 53.28\%.