Detekcija karakterističnih točaka na licima

master's thesis

• Title (english): Facial Feature Point Detection
• Author: Juraj Pejnović
• Mentor: Tomislav Pribanić (mentor)
• Committee member: Tomislav Pribanić (predsjednik povjerenstva)
• Committee member: Zvonko Kostanjčar (član povjerenstva)
• Committee member: Tomislav Petković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-09-19, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad prolazi kroz teorijsku podlogu analize slike lica i pronalaska karakterističnih točaka. Prolazi kroz metode i alate potrebne da bi se trenirala umjetna neuronska mreža. Rad također daje teorijsku podlogu za područje dubokog učenja i umjetnih neuronskih mreža korištenih za pronalazak točaka. Navodi se nekoliko različitih baza podatak slika dostupnih javno ili uz pristanak vlasnika. Nudi se rješenje jedne od mogućih neuronskih mreža za pronalazak lica i karakterističnih točaka. Nakon treniranja, rad daje pregled rezultata uz korištenje različitih izvora podataka i ispituje mogućnosti i razloge različitih dobivenih rezultata.

Abstract (english)

The thesis covers the theoretical background of image analysis and facial feature point detection. It goes through the methods and tools necessarry for the training of an aritificial neural network. The thesis also covers the areas of deep learning and artificial neural networks used for feature point detection. A few different data bases are mentioned which are either available publicly or with the consent of the author. A solution for a possible structure of an artificial neural network that can find facial feature points is proposed. After the training, test images from different data bases are compared and the final results are discussed.

• Keywords: karakteristična; točka; lice; detekcija; strojno učenje; neuronska mreža; duboko učenje; konvolucija; SCface; analiza slike; keras; tensorflow
• Keywords (english): feature; point; face; detection; machine learning; neural network; deep learning; convolution; SCface; image analysis; keras; tensorflow
• Language: croatian
• URN:NBN: urn:nbn:hr:168:660492
• 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: 2020-01-07 21:11:54