Internet Archive Scholar

Face Recognition based on Frontalization of Multiple Poses and Expressions using GGAN and PCA

M Shanmugam, Department of ECE, Government Engineering College, Krishnarajapete, Mandya, Visvesvaraya Technological University, Belagavi, India,, V M Viswanatha, Narendra Kumar D N, Department of ECE, HKE'S SLN College of Engineering, Raichur, Visvesvaraya Technological University, Belagavi, India, University Visvesvaraya College of Engineering, Bangalore University, Bengaluru, India
2022 Ymer  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (912.7 kB)
https://web.archive.org/web/20220609145359/http://ymerdigital.com/uploads/YMER210617.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL. The file type is application/pdf.

Face identification is an essential application in recent times as it has several applications like surveillance, criminal identification, etc. Many face recognition methods are still facing various challenges by researchers in real-time applications because of multiple rotational angles of faces and different expressions. To overcome this issue, we come up with the novel approach of Global Generative Adversarial Networks (G-GAN) to convert various expressions and side images into frontal
more » ... and Principal Component Analysis (PCA) to recognize the face in quick time with low cost, high accuracy and with use of less memory. The unique error loss algorithm is introduced in G-GAN during training which preserves identical features of profile images and photo-realistic good resolution front image. The single ground truth frontal image of an individual and G-GAN converted frontal pictures of individuals are considered, and PCA is applied to obtain features. The features are compared using Euclidean Distance (ED) to recognize an individual. The module used publicly available datasets like Indian male and female, ORL, and Multi PIE and achieved good results with a high recognition rate. Keywords: Biometrics, Face Recognition, GAN, Face Profile Images, PCA
doi:10.37896/ymer21.06/11 fatcat:dwtku6m5ajh7flspiclk4ryly4
Citation

M Shanmugam, Department of ECE, Government Engineering College, Krishnarajapete, Mandya, Visvesvaraya Technological University, Belagavi, India,, V M Viswanatha, Narendra Kumar D N, Department of ECE, HKE'S SLN College of Engineering, Raichur, Visvesvaraya Technological University, Belagavi, India, University Visvesvaraya College of Engineering, Bangalore University, Bengaluru, India. "Face Recognition based on Frontalization of Multiple Poses and Expressions using GGAN and PCA." Ymer 21.06 (2022) 92-109