Computer recognition of facial expressions

Klemen Strušnik (2013) Computer recognition of facial expressions. EngD thesis.

Preview

Abstract

In this diploma I try one of the more popular methods for modelling facial images and the extracted data's dimension is later on reduced. I can manipulate the shape and texture with the images reduced data. From these changed images the LGBP algorithm creates histograms, that represents the input of one of the most popular methods for machine learning. In the first part I define emotions and the process of recognizing facial expressions. I mention the factors that can effect on the result of the recognition and some tools for recognizing facial features and facial expressions. In the second part the steps for preprocessing data are mentioned and relevant algorithms for understanding the process of recognition with LGBP and AAM are described - with the steps for implementation in Matlab. The last part consists of classification results of the processed data and the meaning behind them. In the end part I explain my findings and propose some optimization techniques for this application.

Item Type:

Thesis (EngD thesis)

Keywords:

Emotion recognition, Facial expression classification, Active appearance model, Active shape model, Local Gabor binary patterns

Number of Pages:

Language of Content:

Slovenian

Mentor / Comentors:

Name and Surname	ID	Function
doc. dr. Iztok Lebar Bajec	269	Mentor

Link to COBISS:

http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=10168148)

Institution:

University of Ljubljana

Department:

Faculty of Computer and Information Science

Item ID:

2175

Date Deposited:

19 Sep 2013 12:01

Last Modified:

03 Oct 2013 15:15

URI:

http://eprints.fri.uni-lj.si/id/eprint/2175

Actions (login required)

View Item