Please use this identifier to cite or link to this item:
Title: Conditional adversarial synthesis of 3D facial action unit
Authors: Liu, Zhilei
Song, Guoxian
Cai, Jianfei
Cham, Tat-Jen
Zhang, Juyong
Keywords: Engineering::Computer science and engineering
Issue Date: 2019
Source: Liu, Z., Song, G., Cai, J., Cham, T.-J., & Zhang, J. (2019). Conditional adversarial synthesis of 3D facial action unit. Neurocomputing, 355, 200-208. doi:10.1016/j.neucom.2019.05.003
Journal: Neurocomputing
Abstract: Employing deep learning-based approaches for fine-grained facial expression analysis, such as those involving the estimation of Action Unit (AU) intensities, is difficult due to the lack of a large-scale dataset of real faces with sufficiently diverse AU labels for training. In this paper, we consider how AU-level facial image synthesis can be used to substantially augment such a dataset. We propose an AU synthesis framework that combines the well-known 3D Morphable Model (3DMM), which intrinsically disentangles expression parameters from other face attributes, with models that adversarially generate 3DMM expression parameters conditioned on given target AU labels, in contrast to the more conventional approach of generating facial images directly. In this way, we are able to synthesize new combinations of expression parameters and facial images from desired AU labels. Extensive quantitative and qualitative results on the benchmark DISFA dataset and BP4D dataset demonstrate the effectiveness of our method on 3DMM facial expression parameter synthesis and data augmentation for deep learning-based AU intensity estimation.
ISSN: 0925-2312
DOI: 10.1016/j.neucom.2019.05.003
Rights: © 2019 Elsevier B.V. All rights reserved.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:IMI Journal Articles

Citations 20

Updated on Feb 5, 2023

Web of ScienceTM
Citations 20

Updated on Feb 6, 2023

Page view(s)

Updated on Feb 5, 2023

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.