# Custom Structure Preservation in Face Aging

Source: [Custom Structure Preservation in Face Aging](<https://www.sota2.com/research/paper/68fa6d121e209131cea37756>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T00:24:36.116Z

Authors: Guillermo Gomez-Trenado, St\\'ephane Lathuili\\\`ere, Pablo Mesejo, \\'Oscar Cord\\'on (1) \_\_INSTITUTION\_4\_\_ DaSCI research institute, DECSAI, University of Granada, Granada, Spain, (2) LTCI, T\\'el\\'ecom-Paris, Intitute Polytechnique de Paris, Palaiseau, France)

Published: 2022-07-22T11:58:33.000Z

## 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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Face Aging on FFHQ-AT (test)](<https://www.sota2.com/research/sota/face-aging-on-ffhq-at-test.md>): Face Aging; FFHQ-AT (test); Age MAE: 2.91.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2207.11025>)

[PDF](<https://arxiv.org/pdf/2207.11025.pdf>)