Praćenje objekata u videozapisima

undergraduate thesis

• Title (english): Object Tracking in Video Sequences
• Author: Luka Mijić
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Darijan Marčetić (član povjerenstva)
• Committee member: Marin Oršić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Tema rada je problematika praćenja objekata u videozapisima. Rad sadrži dijelove
posvećene potrebnoj teorijskoj podlozi za shvaćanje temeljnih koncepata. Tako su ukratko opisani histogrami, integralne slike, prostori boja. Glavni fokus rada je na Fragtrack algoritmu za praćenje objekata. Algoritam se bazira na uspoređivanju histograma fragmenata predloška objekta. U radu su opisani svi koraci implementacije Fragtrack algoritma. Nakon toga su izvedeni testovi zbog provjere uspješnosti izvedbe algoritma. Testirani su neki složeniji slučajevi, primjerice slučajevi gdje objekt koji se prati biva prekriven ili mijenja svoj oblik ili veličinu. Implementacija je pokazala da se uspješno može nositi s tim problemima, ali ta se otpornost mora platiti vremenom izvođenja koje nažalost nije u stvarnom vremenu.

Abstract (english)

Subject of this work is tracking objects in image sequences. Work contains section
dedicated to theoretical groundwork required for fully understanding concepts that are discussed. Main focus of this work is Fragtrack object tracking algorithm. FragTrack is algorithm that is based on comparing histograms of fragments created from template image. All steps in implementation of algorithm are mentioned. After the implementation comes testing. Some more complex cases were tested. For example when partial occlusion occurs, when object that is being tracked changes it’s shape or size. Implementation was able to handle those difficult cases, but because of that algorithm is not viable for running in real time.

• Keywords: Fragtrack; prećenje objekata; praćenje bazirano na fragmentima; računalni vid; histogram
• Keywords (english): Fragtrack; object tracking; tracking based on fragments; computer vision; histogram
• Language: croatian
• URN:NBN: urn:nbn:hr:168:216299
• 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: 2019-03-13 23:03:53