ERNIE-ViLG 2.0: Improving Text-to-Image Diffusion Model with Knowledge-Enhanced Mixture-of-Denoising-Experts

TL;DR

ERNIE-ViLG 2.0 advances Chinese text-to-image diffusion by integrating detailed knowledge and specialized denoising experts, achieving state-of-the-art image quality and alignment.

Contribution

It introduces knowledge-enhanced mechanisms and stage-specific denoising experts to improve image fidelity and text relevance in diffusion models.

Findings

Achieves zero-shot FID of 6.75 on MS-COCO

Outperforms recent models in image fidelity

Excels in image-text alignment with human evaluation

Abstract

Recent progress in diffusion models has revolutionized the popular technology of text-to-image generation. While existing approaches could produce photorealistic high-resolution images with text conditions, there are still several open problems to be solved, which limits the further improvement of image fidelity and text relevancy. In this paper, we propose ERNIE-ViLG 2.0, a large-scale Chinese text-to-image diffusion model, to progressively upgrade the quality of generated images by: (1) incorporating fine-grained textual and visual knowledge of key elements in the scene, and (2) utilizing different denoising experts at different denoising stages. With the proposed mechanisms, ERNIE-ViLG 2.0 not only achieves a new state-of-the-art on MS-COCO with zero-shot FID score of 6.75, but also significantly outperforms recent models in terms of image fidelity and image-text alignment, with side-by-side human evaluation on the bilingual prompt set ViLG-300.