AnyText2: Visual Text Generation and Editing With Customizable Attributes

TL;DR

AnyText2 advances text-to-image generation by enabling precise control over multilingual text attributes like font and color, improving realism, speed, and accuracy in scene image editing.

Contribution

It introduces a novel architecture and techniques for controlling text attributes in scene images, enhancing realism and accuracy over previous methods.

Findings

19.8% faster inference speed

3.3% and 9.3% improvements in text accuracy for Chinese and English

State-of-the-art performance demonstrated

Abstract

As the text-to-image (T2I) domain progresses, generating text that seamlessly integrates with visual content has garnered significant attention. However, even with accurate text generation, the inability to control font and color can greatly limit certain applications, and this issue remains insufficiently addressed. This paper introduces AnyText2, a novel method that enables precise control over multilingual text attributes in natural scene image generation and editing. Our approach consists of two main components. First, we propose a WriteNet+AttnX architecture that injects text rendering capabilities into a pre-trained T2I model. Compared to its predecessor, AnyText, our new approach not only enhances image realism but also achieves a 19.8% increase in inference speed. Second, we explore techniques for extracting fonts and colors from scene images and develop a Text Embedding Module that encodes these text attributes separately as conditions. As an extension of AnyText, this method allows for customization of attributes for each line of text, leading to improvements of 3.3% and 9.3% in text accuracy for Chinese and English, respectively. Through comprehensive experiments, we demonstrate the state-of-the-art performance of our method. The code and model will be made open-source in https://github.com/tyxsspa/AnyText2.