prikaz prve stranice dokumenta Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone
  Download
PDF 2.12 MB
master's thesis
Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone
2023. urn:nbn:hr:166:238937
Farkaš, Luka
University of Split
Faculty of Science
Department of Informatics

Cite this document

Farkaš, L. (2023). Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone (Master's thesis). Split: University of Split, Faculty of Science. Retrieved from https://urn.nsk.hr/urn:nbn:hr:166:238937

Farkaš, Luka. "Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone." Master's thesis, University of Split, Faculty of Science, 2023. https://urn.nsk.hr/urn:nbn:hr:166:238937

Farkaš, Luka. "Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone." Master's thesis, University of Split, Faculty of Science, 2023. https://urn.nsk.hr/urn:nbn:hr:166:238937

Farkaš, L. (2023). 'Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone', Master's thesis, University of Split, Faculty of Science, accessed 22 November 2024, https://urn.nsk.hr/urn:nbn:hr:166:238937

Farkaš L. Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone [Master's thesis]. Split: University of Split, Faculty of Science; 2023 [cited 2024 November 22] Available at: https://urn.nsk.hr/urn:nbn:hr:166:238937

L. Farkaš, "Object tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone", Master's thesis, University of Split, Faculty of Science, Split, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:166:238937

Please login to the repository to save this object to your list.
Metadata
TitleObject tracking and detection with YOLOv8 and StrongSORT algorithms captured by drone
Title (croatian)Praćenje i detekcija objekata pomoću YOLOv8 i StrongSORT algoritama snimanih dronom
AuthorLuka Farkaš
MentorSaša Mladenović (mentor)
Committee memberSaša Mladenović (predsjednik povjerenstva)
Committee memberGoran Zaharija (član povjerenstva)
Committee memberDino Nejašmić (član povjerenstva)
GranterUniversity of Split
Faculty of Science
(Department of Informatics)
Split
Defense date and country2023-08-29, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
The popularity of computer vision tasks inspires the community to develop diverse object detection and multiple object tracking algorithms. This work is mainly based on the newest technologies in object detection (YOLOv8) and multiple object tracking (StrongSORT). The evolution of YOLOv8 and StrongSORT is reviewed, and algorithms are implemented into a unified model that detects and tracks objects in realtime. To enhance this work, we proposed EL-YOLOv8 to improve object detection on the VisDrone dataset which is a demanding object detection dataset for crowded and small object scenarios. The thesis also contains detailed dataset methodology, training approach, results, and model developed in Python.
Abstract (croatian)
Popularnost zadataka računalnog vida nadahnjuje zajednicu da razvijaju raznovrsne algoritme za detekciju i praćenje objekata. Ovaj rad je temeljen na najnovijim tehnologijama u području detekcije objekata (YOLOv8) i praćenju objekata (StrongSORT). Evolucija YOLOv8 i StrongSORT-a je istražena i algoritmi su implementirani u jedinstven model za koji detektira i prati objekte u realnom vremenu. Kako bi unaprijedili ovaj rad, predložili smo EL-YOLOv8 za poboljšanje detekcije na VisDrone skupu podataka koji je zahtjevan skup podataka za detekciju objekata u scenarijima gužve i detekcije malih objekata. Ovaj rad sadrži i detaljnu metodologiju na skupu podataka, pristup uvježbavanju, rezultate i model razvijen u Pythonu.
Keywords
Keywords (croatian)
Languageenglish
URN:NBNurn:nbn:hr:166:238937
Study programmeTitle: Data Science and Engineering
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra podatkovne znanosti i inženjerstva (magistar/magistra podatkovne znanosti i inženjerstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) icon
Created on2023-09-18 12:17:18