DiffClean: Diffusion-based Makeup Removal for Accurate Age Estimation

2025-07-17

AI Summary

To address the challenge of facial makeup interfering with age estimation for minors, this paper proposes a high-fidelity makeup-removal method based on text-guided diffusion models. The approach jointly leverages semantic textual prompts and image reconstruction to effectively remove both digitally synthesized and real-world makeup artifacts, thereby enhancing the robustness of age estimation and face verification. Evaluated on cross-domain makeup datasets, the method improves binary classification accuracy (minor vs. adult) by 4.8% and boosts the true match rate (TMR) of face verification by 8.9% at a false match rate (FMR) of 0.01%. To the best of our knowledge, this is the first work to introduce text-guided diffusion models into makeup removal, achieving a balanced trade-off between identity preservation and makeup removal fidelity. The proposed framework establishes a novel paradigm for trustworthy face recognition tailored to age-gated applications.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Privacy

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Accurate age verification can protect underage users from unauthorized access to online platforms and e-commerce sites that provide age-restricted services. However, accurate age estimation can be confounded by several factors, including facial makeup that can induce changes to alter perceived identity and age to fool both humans and machines. In this work, we propose DiffClean which erases makeup traces using a text-guided diffusion model to defend against makeup attacks. DiffClean improves age estimation (minor vs. adult accuracy by 4.8%) and face verification (TMR by 8.9% at FMR=0.01%) over competing baselines on digitally simulated and real makeup images.

Problem

• Remove facial makeup to improve age estimation accuracy
• Defend against makeup attacks on facial verification systems
• Enhance minor vs adult classification in age-restricted services

Innovation

• Uses text-guided diffusion model
• Erases makeup traces effectively
• Improves age and face verification