Praćenje objekata u videozapisima

undergraduate thesis

• Title (english): Object Tracking in Video Sequences
• Author: Marko Damjanić
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Zorana Doždor (član povjerenstva)
• Committee member: Franjo Matković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Razmatramo problem praćenja višestrukih objekata u videu. Predlažemo metodu SORT koja problem praćenja objekta kroz slijed okvira razbija na dijelove. Za dio s detekcijom objekta prolazimo kroz dvije različite familije rješenja čijoj se dubljoj funkcionalnosti približimo. Izlazi dijela detekcije su granični okviri detektiranih objekata. Koristi se predviđanje teoretskog pomaka pomoću Kalman filtra te asocijacija objekta između okvira pomoću Mađarskog algoritma. U implementaciji je korišten programski jezik C++ i vanjska biblioteka OpenCV koja olakšava rad sa slikama. Metodu smo evaluirali nad videima prometnog pločnika i autoceste. Postignuti rezultati su prikazani i diskutirani.

Abstract (english)

We are considering the problem of tracking multiple objects in a video. We propose the SORT method, which breaks the problem of tracking an object through a sequence of frames into parts. For the part with object detection, we go through two different families of solutions whose deeper functionality we approach. The outputs of the detection part are the bounding boxes of the detected objects. Predicting the theoretical displacement using the Kalman filter and the association of objects between frames using the Hungarian algorithm are both explained in detail. The implementation used the C++ programming language and an external OpenCV library that facilitates working with images. We evaluated the method over a busy sidewalk and highway videos. The achieved results are presented and discussed.

• Keywords: Pračenje višestrukih objekata; SORT; YOLO; Kalman filter; Mađarski algoritam; OpenCV; C++; računalni vid
• Keywords (english): Multi-Object tracking; SORT; YOLO; Kalman filter; Hungarian algorithm; OpenCV; C++; Computer vision
• Language: croatian
• URN:NBN: urn:nbn:hr:168:293735
• 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 09:58:24