A study of face recognition and its application

This is a study on Facial Recognition, identifying the common issues faced by the algorithms and how facial recognition is going to impact our future. The objective of this project is to study Viola Jones algorithm, Convolutional Neural Network and how they work and how they did they impact the f...

Full description

Saved in:
Bibliographic Details
Main Author: Koh, Derrick Jun Ming
Other Authors: Ma Kai Kuang
Format: Final Year Project
Language:English
Published: 2018
Subjects:
Online Access:http://hdl.handle.net/10356/76312
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:This is a study on Facial Recognition, identifying the common issues faced by the algorithms and how facial recognition is going to impact our future. The objective of this project is to study Viola Jones algorithm, Convolutional Neural Network and how they work and how they did they impact the field of Computer Vision. The common challenges that the computer vision faces are Viewpoint variation, Scale variation, Deformation, Occlusion, Illumination conditions, Background clutter and Intra-class variation. Viola-Jones algorithm is a breakthrough because it is so fast and simple, it could spot faces in an image in real time. By ignoring face recognition and focusing on detection. They can only detect faces from the front. By detecting general patterns associated with faces in a cascade which result that the algorithm can work quickly in real-time. Convolutional Neural Network is a class of artificial neural networks working together to analyze an imagery. The Convolutional Neural Network have three basic layers that makes up the architecture commonly known as Convolutional Layer, Pooling layer, Fully-Connected layer. The first layer extracts and process the raw information the it receives and then pass it onto the next layer, it continues to do so until it reaches the final layer of the Neural Network. The final layer called the output layer which will then classify the image.