Poster (Scientific congresses, symposiums and conference proceedings)
Disentangled Face Identity Representationsfor Joint 3D Face Recognition and Neutralisation
Kacem, Anis; cherenkova, kseniya; Aouada, Djamila
2022International Conference on Virtual Reality
Peer reviewed
 

Files


Full Text
V3090 (1).pdf
Author postprint (756.29 kB)
Download

All documents in ORBilu are protected by a user license.

Send to



Details



Abstract :
[en] In this paper, we propose a new deep learning based approach for disentangling face identity representations from expressive 3D faces. Given a 3D face, our approach not only extracts a disentangled identity representation, but also generates a realistic 3D face with a neutral expression while predicting its identity. The proposed network consists of three components; (1) a Graph Convolutional Autoencoder (GCA) to encode the 3D faces into latent representations, (2) a Generative Adversarial Network (GAN) that translates the latent representations of expressive faces into those of neutral faces, (3) and an identity recognition sub-network taking advantage of the neutralized latent representations for 3D face recognition. The whole network is trained in an end-to-end manner. Experiments are conducted on three publicly available datasets showing the effectiveness of the proposed approach.
Research center :
Interdisciplinary Centre for Security, Reliability and Trust (SnT) > Computer Vision Imaging & Machine Intelligence (CVI²)
Disciplines :
Computer science
Author, co-author :
Kacem, Anis ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2
cherenkova, kseniya;  Artec3D
Aouada, Djamila  ;  University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > CVI2
External co-authors :
no
Language :
English
Title :
Disentangled Face Identity Representationsfor Joint 3D Face Recognition and Neutralisation
Publication date :
28 May 2022
Event name :
International Conference on Virtual Reality
Event date :
26-28 May 2022
By request :
Yes
Audience :
International
Peer reviewed :
Peer reviewed
Focus Area :
Computational Sciences
FnR Project :
FNR11643091 - Face Identification Under Deformations, 2017 (01/05/2018-31/10/2021) - Djamila Aouada
Available on ORBilu :
since 05 November 2021

Statistics


Number of views
133 (30 by Unilu)
Number of downloads
30 (5 by Unilu)

Scopus citations®
 
3
Scopus citations®
without self-citations
3

Bibliography


Similar publications



Contact ORBilu