Praćenje više objekata pomoću metode višestrukih hipoteza

undergraduate thesis

• Title (english): Multiple object tracking via multiple hypothesis method
• Author: Matko Bošnjak
• Mentor: Ivan Marković (mentor)
• Committee member: Ivan Marković (član povjerenstva)
• Committee member: Luka Petrović (član povjerenstva)
• Committee member: Jelena Gregorić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-09-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad prikazuje probleme i izazove praćenja više objekata te predstavlja tri glavne skupine metoda za praćenje više objekata. Prikazan je i rad Kalmanovog filtra koji se koristi u tim metodama za predstavljanje stanja objekata koji se prate i nji- hovo propagiranje u buduće vremenske korake. Evaluirane su sposobnosti jedne od tih metoda, praćenje više objekata metodom višestrukih hipoteza. Evaluirana je njena izvedenica, TOMHT, odnosno implemntacija TOMHT sustava u programskom jeziku Pythonu na generiranim i vanjskim podacima. Implementacija i postavke Kalmano- vog filtra su prilagod̄ene obradi podataka sa javnog izazova za praćenje više objekata MOTChallenge te su rezultati prikazani odgovarajućim metrikama.

Abstract (english)

This thesis presents the problems and challenges of tracking multiple objects and introduces three main groups of methods for tracking multiple objects. The operation of the Kalman filter used in these methods for representing the states of the objects be- ing tracked and their propagation into future time steps is also presented. The capabili- ties of one of these methods, tracking multiple objects using the Multiple Hypotheses Tracking method, were evaluated. Its derivative, TOMHT, the implementation of the TOMHT system in the Python programming language on generated and external data was evaluated. The implementation and settings of the Kalman filter have been adap- ted for processing data from the public MOTChallenge, and the results are presented with the corresponding metrics.

• Keywords: praćenje više objekata; metoda višestrukih hipoteza; Kalmanov filter; RFS; JPDAF; MHT; TOMHT; HOMHT
• Keywords (english): multiple object tracking; multiple hypothesis tracking; Kalman filter; RFS; JPDAF; MHT; TOMHT; HOMHT
• Language: croatian
• URN:NBN: urn:nbn:hr:168:250376
• 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: 2023-10-11 08:30:15