Custom Structure Preservation in Face Aging

TL;DR

This paper introduces a novel face aging architecture that allows adjustable structure preservation during age editing by disentangling style and content, using a masking mechanism without extra supervision, outperforming prior methods.

Contribution

The work presents a new style-content disentanglement approach with a masking mechanism for adjustable structure preservation in face aging, without requiring additional supervision.

Findings

Outperforms prior face aging methods in quality and structure preservation.

Allows user-controlled adjustment of structural details during inference.

Demonstrates effectiveness through quantitative, qualitative, and user study evaluations.

Abstract

In this work, we propose a novel architecture for face age editing that can produce structural modifications while maintaining relevant details present in the original image. We disentangle the style and content of the input image and propose a new decoder network that adopts a style-based strategy to combine the style and content representations of the input image while conditioning the output on the target age. We go beyond existing aging methods allowing users to adjust the degree of structure preservation in the input image during inference. To this purpose, we introduce a masking mechanism, the CUstom Structure Preservation module, that distinguishes relevant regions in the input image from those that should be discarded. CUSP requires no additional supervision. Finally, our quantitative and qualitative analysis which include a user study, show that our method outperforms prior art and demonstrates the effectiveness of our strategy regarding image editing and adjustable structure preservation. Code and pretrained models are available at https://github.com/guillermogotre/CUSP.