Face Aging With Conditional Generative Adversarial Networks

Abstract

It has been recently shown that Generative Adversarial Networks (GANs) can produce synthetic images of exceptional visual fidelity. In this work, we propose the GAN-based method for automatic face aging. Contrary to previous works employing GANs for altering of facial attributes, we make a particular emphasize on preserving the original person's identity in the aged version of his/her face. To this end, we introduce a novel approach for "Identity-Preserving" optimization of GAN's latent vectors. The objective evaluation of the resulting aged and rejuvenated face images by the state-of-the-art face recognition and age estimation solutions demonstrate the high potential of the proposed method.

5 pages, 3 figures, accepted at ICIP 2017. With respect to v1: (1) changed the abbreviation of the main model from "acGAN" to "Age-cGAN" in order to avoid confusion with "Auxiliary Classifier Generative Adversarial Networks" introduced by Odena et al.; (2) corrected a typo in Formula 1

Cited by in corpus (10)

• How Generative Adversarial Networks and Their Variants Work: An Overview
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Style Aggregated Network for Facial Landmark Detection
• Image-based model parameter optimization using Model-Assisted Generative Adversarial Networks
• IterGANs: Iterative GANs to Learn and Control 3D Object Transformation
• Generative adversarial networks for generation and classification of physical rehabilitation movement episodes
• Towards Learning Fine-Grained Disentangled Representations from Speech
• Style and Content Disentanglement in Generative Adversarial Networks
• Fast Face Image Synthesis with Minimal Training
• A Study of Cross-domain Generative Models applied to Cartoon Series