Data Entry: Please note that the research database will be replaced by UNIverse by the end of October 2023. Please enter your data into the system https://universe-intern.unibas.ch. Thanks
15th IEEE Conference on Automatic Face and Gesture Recognition (FG 2019)
Volume
1
Place of Conference
Lille, France
Publisher
IEEE
ISSN/ISBN
978-1-7281-0089-0
Abstract
We propose a photo-realistic method for artificially ageing facial photographs by combining learned shape deformations with skin detail transfer between a donor and a receiver face. Facial ageing is a complicated process that most existing face models, such as 3d Morphable Models fail to express, due to lacking correspondence between the wrinkles of different individuals. We propose an exemplar-based approach to face ageing, where we transfer high-frequency details from an older face texture to a younger receiver. By warping the resulting image according to a learned shape ageing deformation, we obtain photo-realistic aged photographs. We evaluate the simulator with human perception experiments showing that we can indeed create results that are perceived to be real.