How Control Information Influences Multilingual Text Image Generation and Editing?

TL;DR

This paper investigates how control information affects multilingual text image generation and editing, proposing a new framework that enhances quality by optimizing control features and employing a two-stage process.

Contribution

It introduces TextGen, a novel framework that improves multilingual text image generation by analyzing control information's role and optimizing input/output features with Fourier analysis.

Findings

Control information has unique characteristics compared to traditional inputs.

Control information plays different roles at various stages of denoising.

Output control features differ significantly in the frequency domain from base features.

Abstract

Visual text generation has significantly advanced through diffusion models aimed at producing images with readable and realistic text. Recent works primarily use a ControlNet-based framework, employing standard font text images to control diffusion models. Recognizing the critical role of control information in generating high-quality text, we investigate its influence from three perspectives: input encoding, role at different stages, and output features. Our findings reveal that: 1) Input control information has unique characteristics compared to conventional inputs like Canny edges and depth maps. 2) Control information plays distinct roles at different stages of the denoising process. 3) Output control features significantly differ from the base and skip features of the U-Net decoder in the frequency domain. Based on these insights, we propose TextGen, a novel framework designed to enhance generation quality by optimizing control information. We improve input and output features using Fourier analysis to emphasize relevant information and reduce noise. Additionally, we employ a two-stage generation framework to align the different roles of control information at different stages. Furthermore, we introduce an effective and lightweight dataset for training. Our method achieves state-of-the-art performance in both Chinese and English text generation. The code and dataset available at https://github.com/CyrilSterling/TextGen.