University of Twente Student Theses

Login

Free patch-based age estimation from facial images by convolutional neutral network

Li, Y. (2018) Free patch-based age estimation from facial images by convolutional neutral network.

[img] PDF
793kB
Abstract:In biometric and computer vision studies, estimating age from facial images is an important task with practical value. However, due to the complexity of aging process and the lack of appropriate databases, it is hard to develop an automatic age estimation system based on facial images. Aiming to overcome these difficulties, this paper proposes a free patch-based method for age estimation. The method selects a set of skin patches extracted from facial images as the studied objects, instead of examining the holistic images. For investigating the biometric features behind those skin patches, the deep learning tool Convolutional Neural Network is deployed. Along with the implementation of several experiments on patches, the possibility of classifying people’s age group by this proposed method is evaluated in this paper.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:56 civil engineering
Programme:Electrical Engineering BSc (56953)
Link to this item:http://purl.utwente.nl/essays/75945
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page