Auto tagging of facial images for digital photo albums

With the constantly improving technology, taking images of yourself or with a group of friends and storing the images can be done anywhere by just using personal mobile phones. With such convenience, hundreds or thousands of images can be taken and uploaded to popular social network websites such as...

Full description

Saved in:
Bibliographic Details
Main Author: Lim, Chu Xiang.
Other Authors: Teoh Eam Khwang
Format: Final Year Project
Language:English
Published: 2013
Subjects:
Online Access:http://hdl.handle.net/10356/52620
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:With the constantly improving technology, taking images of yourself or with a group of friends and storing the images can be done anywhere by just using personal mobile phones. With such convenience, hundreds or thousands of images can be taken and uploaded to popular social network websites such as Facebook and Twitter every day. But, tagging of names to every one in every single image becomes a hassle and requires a long time every time the user uploads images to his/her personal album. Through this project, the student aims to automatically tag the name(s) of any individual in his/her digital photo albums. For this project, it will use three methods or techniques in doing identification of person. The methods are Local Binary Patterns (LBP), Gabor Filter, and Principal Component Analysis (PCA) with LBP. The testing images will be tested for accuracy against two training database, one of which is the cropped database and the other is aligned database. Histograms will be computed and the distances calculated and compared with each other. The shortest distance among all should be the person of interest in each testing image. In this project, the student uses David Beckham as his model for the entire process.