Detekcija lica u videu na temelju boje i informacije o dubini scene

undergraduate thesis

• Title (english): Face Detection in Video Based on Fusion of Colour and Scene Depth Information
• Author: Marko Samardžija
• Mentor: Slobodan Ribarić (mentor)
• Committee member: Slobodan Ribarić (predsjednik povjerenstva)
• Committee member: Darijan Marčetić (član povjerenstva)
• Committee member: Martin Soldić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-09-09, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Detekcija lica pomoću trodimenzionalne Kinect kamere uporabom Matlaba. U radu su opisani algoritmi pronalaska rubova, odvajanja prednjeg plana od pozadine i detekcije lica pomoću boje. Opis eksperimenta i prikaz rezultata pokazuju nam kako dodavanje informacije dubine utječe na poboljšanje postojećeg algoritma pronalaska pomoću boje. Teorija i način dolaska do rješenja opisani su također uz implementacijske detalje i objašnjenje dijelova koda. Programski kod proveden je na nizu videosekvenci iz vlastito oblikovane eksperimentalne baze. Program prikazuje lice pomoću žutih linija koje iscrtavaju piksele koji označavaju boju kože. Tablično je prikazana točnost i postotak detekcije programa te razlika između tih vrijednosti s programom bez dubine i sa dubinom.

Abstract (english)

Face detection via three dimensional Kinect camera using Matlab. Thesis contains edge finding, foreground extraction and color based face detection algorithms. The description of the experiment and results show us how adding the depth information affects the existing color based algorithm in a way that it improves it and gives better results. Theory and coming up with the solution are described as well as implementational details and explanation of parts of code. The program is run through a series of video sequences from own shaped experimental database. The program shows face through yellow lines that draw pixels which mark skin color. The detection rate and percentage of success are graphically shown with differences between those values with and without using depth.

• Keywords: boja; dubinska mapa; detekcija lica; YCbCr; heuristika; Matlab; Kinect
• Keywords (english): color; depth map; face detection; heuristic; kinect; matlab; YCbCr; Matlab; Kinect; Face recognition
• Language: croatian
• URN:NBN: urn:nbn:hr:168:629686
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 11:41:28