Run Run Shaw Library
 Run Run Shaw Library

Home >
Student Final Year Projects >
Electronic Engineering - Undergraduate Final Year Projects >

Please use this identifier to cite or link to this item:

Title: Deep Face Analysis with Convolution Neural Network
Authors: Wong, Pak Lok
Department: Department of Electronic Engineering
Issue Date: 2016
Supervisor: Supervisor: Mr. Ting, Van C W; Assessor: Dr. Yeung, Alan K H
Abstract: Gender-age analysis played a significant part in facial image analysis, it owns great potential in relevant amount of applications, especially since the rise of global social media platforms. However, despite advances in automatic gender-age estimation, it also encounters a low-accuracy problem since real-world images are captured by different poses, angles, lighting, environmental factors, etc. This project aims at experimenting on a gender-age classifier using unconstrained images that classifies a person's gender and age automatically. Deep Convolutional Neural Network (CNN) was hired to design for main gender-age classification model. Caffe which is a deep learning framework, was chosen to construct and implement for our deep learning network. The final gender and age classification rates achieved by our trained model were 95.46% and 84.21% respectively on Adience benchmark database. The result of this project illustrated that CNN could be used to improve gender and age classification with better performances even considering smaller size of unconstrained images.
Appears in Collections:Electronic Engineering - Undergraduate Final Year Projects

Files in This Item:

File SizeFormat
fulltext.html146 BHTMLView/Open

Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer