Multi‐view facial action unit detection via deep feature enhancement

Multi‐view facial action unit (AU) analysis has been a challenging research topic due to multiple disturbing variables, including subject identity biases, variational facial action unit intensities, facial occlusions and non‐frontal head‐poses. A deep feature enhancement (DFE) framework is presented...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Electronics letters Jg. 57; H. 25; S. 970 - 972
Hauptverfasser: Tang, Chuangao, Lu, Cheng, Zheng, Wenming, Zong, Yuan, Li, Sunan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Stevenage John Wiley & Sons, Inc 01.12.2021
Wiley
Schlagworte:
Bias
Complex variables
Computer vision and image processing techniques
Datasets
Image recognition
Mapping
ISSN:0013-5194, 1350-911X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Multi‐view facial action unit (AU) analysis has been a challenging research topic due to multiple disturbing variables, including subject identity biases, variational facial action unit intensities, facial occlusions and non‐frontal head‐poses. A deep feature enhancement (DFE) framework is presented to tackle some of these coupled complex disturbing variables for multi‐view facial action unit detection. The authors' DFE framework is a novel end‐to‐end three‐stage feature learning model with taking subject identity biases, dynamic facial changes and head‐pose into consideration. It contains three feature enhancement modules, including coarse‐grained local and holistic spatial feature learning (LHSF), spatio‐temporal feature learning (STF) and head‐pose feature disentanglement (FD). Experimental results show that the proposed method achieved state‐of‐the‐art recognition performance on the FERA2017 dataset. The code is released at http://aip.seu.edu.cn/cgtang/. A deep feature enhancement (DFE) framework is proposed, in which a novel end‐to‐end three‐stage feature learning model is presented taking subjects identity biases, dynamic facial changes and head‐pose into consideration. Experimental results show that the proposed method achieved state‐of‐the‐art recognition performance on the FERA2017 dataset.
Bibliographie:Chuangao Tang and Cheng Lu has equally contributed for the article.
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0013-5194
1350-911X
DOI:10.1049/ell2.12322