Duboki model namijenjen za učenje i prepoznavanje karakteristika lica

master's thesis

• Title (english): Deep Model for Learning and Recognizing Facial Features
• Author: Luka Martić
• Mentor: Tomislav Burić (mentor)
• Committee member: Tomislav Burić (predsjednik povjerenstva)
• Committee member: Mervan Pašić (član povjerenstva)
• Committee member: Mario Krnić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-06, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Neizraziti skup je skup za koji se ne može jasno odrediti pripada li neki element skupu ili ne. Duboko učenje porodica je strojnog učenja koje se primarno temelji na umjetnim neuronskim mrežama. U okviru ovog diplomskog rada izgrađen je i predstavljen duboki model namijenjen za kategorizaciju karakteristika lica koje se modeliraju kao neizrazite veličine na skupu za učenje CelebA. Rezultati modela zaostaju za state-of-the-art rješenjima, ali predstavljaju podlogu za daljnje istraživanje novih metoda učenja neizrazitih veličina.

Abstract (english)

Fuzzy set is a set for which it is not possible to clearly determine whether an element belongs to it or not. Deep learning is a family of machine learning that is primarily based on artificial neural networks. Within this diploma thesis, a deep model was built and presented with the intention for the categorization of facial features that are modeled as fuzzy quantities and learned on CelebA dataset. The results of the model lag behind state-of-the-art solutions, but provide a basis for further research into new methods of learning fuzzy quantities.

• Keywords: neizraziti skup; neuronska mreža; duboko učenje; model; funkcija gubitka; rezidualna mreža; učenje karakteristika lica
• Keywords (english): fuzzy set; neural netowrk; deep learning; model; loss function; residual network; learning facial features
• Language: croatian
• URN:NBN: urn:nbn:hr:168:745647
• 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 15:32:19