FFHQ-Makeup: Paired Synthetic Makeup Dataset with Facial Consistency Across Multiple Styles

2025-08-05

AI Summary

Existing makeup datasets suffer from limited scale, low photorealism, and misalignment in identity or facial expression between unmade-up and made-up images. To address these issues, this paper introduces MakeupPairs—the first high-quality, paired dataset for multi-style makeup transfer. Built upon FFHQ, it employs a disentangled framework that separates identity representation from makeup features and integrates realistic makeup style modeling. The dataset comprises 90,000 high-resolution (1024×1024) unmade-up/made-up image pairs, covering 18,000 distinct identities and five finely defined makeup styles. All pairs strictly preserve identity, pose, and facial expression consistency, significantly improving cross-style fidelity and facial structural coherence. MakeupPairs enables supervised learning for applications such as virtual makeup try-on and facial privacy protection. Both the dataset and code are publicly released.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Application Domains: Humanities & Computational Social Science

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Paired bare-makeup facial images are essential for a wide range of beauty-related tasks, such as virtual try-on, facial privacy protection, and facial aesthetics analysis. However, collecting high-quality paired makeup datasets remains a significant challenge. Real-world data acquisition is constrained by the difficulty of collecting large-scale paired images, while existing synthetic approaches often suffer from limited realism or inconsistencies between bare and makeup images. Current synthetic methods typically fall into two categories: warping-based transformations, which often distort facial geometry and compromise the precision of makeup; and text-to-image generation, which tends to alter facial identity and expression, undermining consistency. In this work, we present FFHQ-Makeup, a high-quality synthetic makeup dataset that pairs each identity with multiple makeup styles while preserving facial consistency in both identity and expression. Built upon the diverse FFHQ dataset, our pipeline transfers real-world makeup styles from existing datasets onto 18K identities by introducing an improved makeup transfer method that disentangles identity and makeup. Each identity is paired with 5 different makeup styles, resulting in a total of 90K high-quality bare-makeup image pairs. To the best of our knowledge, this is the first work that focuses specifically on constructing a makeup dataset. We hope that FFHQ-Makeup fills the gap of lacking high-quality bare-makeup paired datasets and serves as a valuable resource for future research in beauty-related tasks.

Problem

• Lack of high-quality paired bare-makeup facial datasets
• Inconsistencies in facial identity and expression in synthetic makeup
• Limited realism in existing synthetic makeup approaches

Innovation

• Improved makeup transfer method disentangles identity and makeup
• Generates 90K high-quality bare-makeup image pairs
• Preserves facial consistency across multiple makeup styles