Doukas_HeadGAN_One-Shot_Neural_Head_Synthesis_and_Editing_ICCV_2021_paper.pdf (9.28 MB)
HeadGAN: one-shot neural head synthesis and editing
conference contribution
posted on 2023-06-10, 01:29 authored by Michail Christos Doukas, Stefanos Zafeiriou, Viktoriia SharmanskaViktoriia SharmanskaRecent 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.
History
Publication status
- Published
File Version
- Published version
Journal
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021Publisher
Computer Vision FoundationPage range
14398-14407Event name
International Conference on Computer Vision (ICCV)Event location
VirtualEvent type
conferenceEvent date
October 11 - 17 2021Department affiliated with
- Informatics Publications
Full text available
- Yes
Peer reviewed?
- Yes
Legacy Posted Date
2021-10-22First Open Access (FOA) Date
2021-10-22First Compliant Deposit (FCD) Date
2021-10-22Usage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC