Analysis Of Facial Representation For Human Age Estimation


Chang, Quan Yan (2019) Analysis Of Facial Representation For Human Age Estimation. Masters thesis, Multimedia University.

Full text not available from this repository.


Age estimation has gained enormous interest among researchers due to the excessive demand in day-to-day application. There are several types of age estimation methods such as estimation on gait and facial images. In this thesis, two approaches, namely Two-Step implementation and Score-Level implementation are proposed to take the advantage of the Active Appearance Model in extracting age-related features. The Appearance Feature Model (AFM) is utilised to extract the useful information for facial age estimation. Specifically, Histogram of Oriented Gradient (HOG) and Opponent-Color Local Binary Pattern (CLBP) are integrated with AAM for these proposed implementations. In Two-Step implementation, Histogram based Active Appearance Model (HAAM) and Opponent-Color LBP based Active Appearance Model (CAAM) are presented. HAAM is a process of extracting the facial feature with AAM and pass to HOG to enhance the features. On the other hand, CAAM is the process of extracting the facial feature with AAM and pass to CLBP to get the color information of facial feature. In Score-Level implementation, H+AAM and C+AAM are introduced.

Item Type: Thesis (Masters)
Additional Information: Call No.: TA1653 .C43 2019
Uncontrolled Keywords: Human face recognition (Computer science)
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1001-1280 Transportation engineering
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 21 Sep 2020 21:20
Last Modified: 21 Sep 2020 21:20


Downloads per month over past year

View ItemEdit (login required)