Improving Text Generation on Images with Synthetic Captions

TL;DR

This paper explores a cost-effective fine-tuning method for latent diffusion models to enhance the accuracy of text generation within images, using synthetic captions and minimal training.

Contribution

It introduces a low-cost fine-tuning approach leveraging SDXL and synthetic captions to improve text legibility in images without extensive dataset training.

Findings

Fine-tuning with synthetic captions improves text accuracy.

Adding random letters enhances text generation quality.

No additional multimodal encoders are needed.

Abstract

The recent emergence of latent diffusion models such as SDXL and SD 1.5 has shown significant capability in generating highly detailed and realistic images. Despite their remarkable ability to produce images, generating accurate text within images still remains a challenging task. In this paper, we examine the validity of fine-tuning approaches in generating legible text within the image. We propose a low-cost approach by leveraging SDXL without any time-consuming training on large-scale datasets. The proposed strategy employs a fine-tuning technique that examines the effects of data refinement levels and synthetic captions. Moreover, our results demonstrate how our small scale fine-tuning approach can improve the accuracy of text generation in different scenarios without the need of additional multimodal encoders. Our experiments show that with the addition of random letters to our raw dataset, our model's performance improves in producing well-formed visual text.