Face-MakeUp: Multimodal Facial Prompts for Text-to-Image Generation

TL;DR

This paper introduces Face-MakeUp, a multimodal approach that uses facial prompts to improve text-to-image diffusion models for generating specific facial images, leveraging a large dataset and feature extraction techniques.

Contribution

The paper presents a new dataset, FaceCaptionHQ-4M, and a method to incorporate multi-scale features into diffusion models for better facial image generation.

Findings

Face-MakeUp outperforms existing methods on face-related datasets.

The model effectively preserves facial identity features.

The approach demonstrates strong performance in generating desired facial images.

Abstract

Facial images have extensive practical applications. Although the current large-scale text-image diffusion models exhibit strong generation capabilities, it is challenging to generate the desired facial images using only text prompt. Image prompts are a logical choice. However, current methods of this type generally focus on general domain. In this paper, we aim to optimize image makeup techniques to generate the desired facial images. Specifically, (1) we built a dataset of 4 million high-quality face image-text pairs (FaceCaptionHQ-4M) based on LAION-Face to train our Face-MakeUp model; (2) to maintain consistency with the reference facial image, we extract/learn multi-scale content features and pose features for the facial image, integrating these into the diffusion model to enhance the preservation of facial identity features for diffusion models. Validation on two face-related test datasets demonstrates that our Face-MakeUp can achieve the best comprehensive performance.All codes are available at:https://github.com/ddw2AIGROUP2CQUPT/Face-MakeUp