HeadGAN: one-shot neural head synthesis and editing

Doukas, Michail Christos, Zafeiriou, Stefanos and Sharmanska, Viktoriia (2021) HeadGAN: one-shot neural head synthesis and editing. International Conference on Computer Vision (ICCV), Virtual, October 11 - 17 2021. Published in: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021. 14398-14407. Computer Vision Foundation

[img] PDF - Published Version
Download (9MB)

Abstract

Recent attempts to solve the problem of head reenactment using a single reference image have shown promising results. However, most of them either perform poorly in terms of photo-realism, or fail to meet the identity preservation problem, or do not fully transfer the driving pose and expression. We propose HeadGAN, a novel system that conditions synthesis on 3D face representations, which can be extracted from any driving video and adapted to the facial geometry of any reference image, disentangling identity from expression. We further improve mouth movements, by utilising audio features as a complementary input. The 3D face representation enables HeadGAN to be further used as an efficient method for compression and reconstruction and a tool for expression and pose editing.

Item Type: Conference Proceedings
Schools and Departments: School of Engineering and Informatics > Informatics
SWORD Depositor: Mx Elements Account
Depositing User: Mx Elements Account
Date Deposited: 22 Oct 2021 10:39
Last Modified: 22 Oct 2021 10:39
URI: http://sro.sussex.ac.uk/id/eprint/102428

View download statistics for this item

📧 Request an update