Show simple item record

AuthorElharrouss O.
AuthorAlmaadeed N.
AuthorAl-Maadeed, Somaya
Available date2022-05-19T10:23:10Z
Publication Date2020
Publication Name2020 IEEE International Conference on Informatics, IoT, and Enabling Technologies, ICIoT 2020
ResourceScopus
Identifierhttp://dx.doi.org/10.1109/ICIoT48696.2020.9089530
URIhttp://hdl.handle.net/10576/31117
AbstractIn this work, a new multitask convolutional neural network (CNN) is proposed aiming for the recognition of face under pose variations. Furthermore, the combination of pose estimation for each corresponding pose in a separate fashion allows robust face recognition in presence of various facial expressions as well as low illuminations. First, a CNN model for pose estimation is proposed. The pose estimation model is trained using a self-collected dataset built from three popular datasets including FLW, CEP, and CASIA-WebFace using three categories of face image capture such as Left side, Frontal and right side. Experimental evaluation has been conducted using two datasets: Pointing'04 and Schneiderman. Results reveal the robustness of the proposed pose estimation model. Moreover, the proposed face pose estimation is applied on three datasets to widen the dataset and make it bigger for training and testing deep learning models.
Sponsoridentity. Compared with the recent and related techniques, the proposed system has been shown to outperform related works and deliver state-of-the-art performance for pose estimation. The built large-scale dataset for training may improve the accuracy of the proposed system as increasing the number of images for each identity will cover a wider range of variations of the face. Also, the generated dataset can be improved in terms of the number of identities using the same methodology on other datasets like VGGFace2. Acknowledgment This publication was made by NPRP grant # NPRP8-140-2-065 from the Qatar National Research Fund (a member of the Qatar Foundation). The statements made herein are solely the responsibility of the authors.
Languageen
PublisherInstitute of Electrical and Electronics Engineers Inc.
SubjectConvolutional neural networks
Deep learning
Internet of things
Statistical tests
Experimental evaluation
Face pose estimation
Facial Expressions
Low illuminations
Pose estimation
Pose-invariant face recognition
Three categories
Training and testing
Face recognition
TitleLFR face dataset:Left-Front-Right dataset for pose-invariant face recognition in the wild
TypeConference Paper
Pagination124-130
dc.accessType Abstract Only


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record