Call For Paper

Volume 10 Issue 3

March 2024

Submit Paper Here
Download Paper Format
Copyright Form
NEWS & UPDATES
News for Authors:

We have started accepting articles by online means directly through website. Its our humble request to all the researchers to go and check the new method of article submission on below link: Submit Manuscript

Follow us on Social Media:

Dear Researchers, to get in touch with the recent developments in the technology and research and to gain free knowledge like , share and follow us on various social media. Facebook

title

FACE RECOGNITION USING DEEP LEARNING

Author(s):

J. Ashok Kumar

Keywords:

CNN, Face detection, Face recognition, Tensor flow

Abstract

Face detection and recognition in unconstrained environment are challenging due to various poses, illuminations and occlusions.However, conventional methods could no longer satisfy the demand at present, due to its low recognition accuracy and restrictions of many occasions. In this paper, we presented the deep learning method to achieve facial detection and face recognition. To solve the face recognition problem, we propose a deep cascaded Multi-Task Framework which exploits the inherent correlation between detection and alignment to boost up their performance. In particular, our Framework leverages a cascaded architecture with three stages of carefully designed deep convolutional networks to predict face in a coarse-to-fine manner.we present a system, called FaceNet, that directly learns a mapping from face images to a compact Euclidean space where distances directly correspond to a measure of face similarity. Once this space has been produced,tasks such as face recognition, verification and clustering can be easily implemented using standard techniques with FaceNet embeddings as feature vectors. Our method uses a deep convolutional network trained to directly optimize the embedding itself, rather than an intermediate bottleneck layer as in previous deep learning approaches. To train, we use triplets of roughly aligned matching / non-matching face patches generated using a novel online triplet loss method and soft max is also used for face verification. The benefit of our approach is much greater representational efficiency: we achieve state of the art face recognition performance using only 128-bytes preface. With the implementation of both Multi-task Framework, FaceNet and Tensor Flow we are creating a model for a better Face Recognition.

Other Details

Paper ID: IJSARTV
Published in: Volume : 6, Issue : 5
Publication Date: 5/1/2020

Article Preview




Download Article