Estimacija dobi u slikama lica

master's thesis

• Title (english): Age Estimation from Facial Images
• Author: Zorana Doždor
• Mentor: Zoran Kalafatić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Tomislav Hrkać (član povjerenstva)
• Committee member: Ante Đerek (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Estimacija dobi iz slika lica može se definirati kao automatski proces dodjeljivanja oznake točne dobi ili raspona dobi ulaznoj slici lica. Općeniti sustav za estimaciju dobi sastoji se od tri glavne komponente: predprocesiranje, ekstrakcija značajki i algoritam za estimaciju dobi. U ovom radu su za ekstrakciju značajki korištene konvolucijske neuronske mreže, pri čemu je isprobano nekoliko poznatih arhitektura. Implementiran je sustav za klasifikaciju dobi u dobne skupine kao i za estimaciju točne dobi te su oba pristupa testirana na poznatim skupovima podataka. Za oba pristupa su isprobani različiti algoritmi s ciljem kreiranja što robusnijeg sustava. Dobiveni rezultati su uspoređeni s trenutno najboljim rezultatima dostupnim u literaturi.

Abstract (english)

Age estimation from facial images is defined as to label the face image automatically with the exact age or age group. Every age estimation system consists of three main stages: preprocessing, feature extraction and age estimation algorithm. In this thesis, convolutional neural networks were used for feature extraction and several different known model architectures were tested. Both a system for classifying ages into age groups and a system for estimating the exact age have been implemented and have been tested on known data sets. Different algorithms were tested for both approaches in order to try and improve robustness of the system. The obtained results were compared with the current state of the art results.

• Keywords: duboko učenje; estimacija dobi; konvolucijske neuronske mreže; regresija; klasifikacija
• Keywords (english): deep learning; age estimation; convolutional neural network; regression; classification
• Language: croatian
• URN:NBN: urn:nbn:hr:168:606466
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 12:08:15