AnyTrans: Translate AnyText in the Image with Large Scale Models

TL;DR

AnyTrans is a versatile framework that uses large-scale models to translate and seamlessly fuse multilingual text within images, leveraging contextual cues and open-source tools without training.

Contribution

It introduces a comprehensive, training-free framework utilizing LLMs and diffusion models for multilingual image text translation and fusion, with a new dataset for evaluation.

Findings

Effective translation of fragmented texts using LLMs.

Seamless text fusion preserving image style and realism.

Open-source, training-free implementation.

Abstract

This paper introduces AnyTrans, an all-encompassing framework for the task-Translate AnyText in the Image (TATI), which includes multilingual text translation and text fusion within images. Our framework leverages the strengths of large-scale models, such as Large Language Models (LLMs) and text-guided diffusion models, to incorporate contextual cues from both textual and visual elements during translation. The few-shot learning capability of LLMs allows for the translation of fragmented texts by considering the overall context. Meanwhile, the advanced inpainting and editing abilities of diffusion models make it possible to fuse translated text seamlessly into the original image while preserving its style and realism. Additionally, our framework can be constructed entirely using open-source models and requires no training, making it highly accessible and easily expandable. To encourage advancement in the TATI task, we have meticulously compiled a test dataset called MTIT6, which consists of multilingual text image translation data from six language pairs.