Vizualno praćenje gibajućih objekata pomoću dubokog učenja

undergraduate thesis

• Title (english): Visual Moving Object Tracking via Deep Learning
• Author: Lovro Bučar
• Mentor: Ivan Petrović (mentor)
• Committee member: Ivan Petrović (član povjerenstva)
• Committee member: Borna Bićanić (član povjerenstva)
• Committee member: Karlo Koledić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Cilj ovog rada bio je upoznati se s neuronskom mrežom dubokih afiniteta (eng.
Deep Affinity Network, DAN), istražiti kako funkcionira i, konačno, testirati ju lokalno
sa zadanim podatcima.
Istraživanje smo započeli čitanjem rada u kojem se detaljno opisuje funkcioniranje
mreže DAN i uobičajene pristupe praćenju objekata u videu. Nakon toga smo preuzeli
program s priloženog GitHub repozitorija i pokrenuli testiranje algoritma s videima iz
izazova za praćenje objekata.
Testiranjem i uspored̄ivanjem modela DAN s ostalim algoritmima koji rade isti
posao saznali smo da se naš model ističe s najboljim performansama u najbitnijim
mjerilima.

Abstract (english)

The aim of this thesis was to get acquainted with a neural network called Deep
Affinity Network (DAN), understand how it works and, finally, test it locally with
given datasets.
We started the research with reading a paper which describes the functionality of
DAN and the conventions of multiple object tracking. Thereafter, we downloaded the
program from the attached GitHub repository and tested the algorithm with multiple
object tracking challenges.
Through testing and comparing DAN to other algorithms that do the same task we
found that our model achieves the best performance in most important metrics.

• Keywords: Neuronska mreža; Praćenje objekata; Mreža dubinskih afiniteta
• Keywords (english): Neural network; Multiple object tracking; Deep Affinity Network
• Language: croatian
• URN:NBN: urn:nbn:hr:168:396133
• 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: 2023-01-15 09:51:07