Identity-Preserving Aging of Face Images via Latent Diffusion Models

TL;DR

This paper introduces a novel method using latent diffusion models to synthetically age and de-age face images, improving face recognition accuracy while maintaining realism and biometric fidelity.

Contribution

It presents a new controllable, few-shot training approach for aging face images with latent diffusion models, enhancing recognition performance on benchmark datasets.

Findings

44% reduction in False Non-Match Rate

High visual realism in generated images

Maintains biometric fidelity

Abstract

The performance of automated face recognition systems is inevitably impacted by the facial aging process. However, high quality datasets of individuals collected over several years are typically small in scale. In this work, we propose, train, and validate the use of latent text-to-image diffusion models for synthetically aging and de-aging face images. Our models succeed with few-shot training, and have the added benefit of being controllable via intuitive textual prompting. We observe high degrees of visual realism in the generated images while maintaining biometric fidelity measured by commonly used metrics. We evaluate our method on two benchmark datasets (CelebA and AgeDB) and observe significant reduction (~44%) in the False Non-Match Rate compared to existing state-of the-art baselines.