文章基本信息

标题：FACE RECOGNITION: LITERATURE REVIEW WITH EMPHASIS ON DEEP LEARNING
本地全文：下载
作者：RAJESHWAR MOGHEKAR ; SACHIN AHUJA
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2019
卷号：97
期号：12
页码：3332-3342
出版社：Journal of Theoretical and Applied
摘要：Under the broad umbrella of object recognition, Face recognition is one area with active research for last few decades mainly due to its applications and the challenges in the environment where they are used. The face recognition as a biometric authentication can work without much cooperation of human. The current face recognition techniques perform well in constrained environment but performance degrades in unconstrained environment as the images captured may vary in resolution, illumination, pose, occlusion and expressions. This sometimes makes intra class variance more than the inter class variance and leads to misclassification. In this article, an overview of some of the strategies adopted by the researchers to overcome the challenges like pose variation, low resolution and occlusion in unconstrained environment has been discussed. Moreover, this paper reviews the use of deep learning in face recognition to achieve accuracy at par with humans in image classification tasks. We also discuss the challenges with Deep learning and the strategies adopted to overcome them. This paper provide an up-to-date review of face recognition techniques.
关键词：Face recognition; Deep Learning; Low resolution; Occlusion; Pose invariant