Suvremene metode detekcije objekata

undergraduate thesis

• Title (english): State-of-the-art Object Detection Methods
• Author: Anton Vivoda
• Mentor: Marin Šilić (mentor)
• Committee member: Marin Šilić (predsjednik povjerenstva)
• Committee member: Ivan Sabolić (član povjerenstva)
• Committee member: Iva Sović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-09-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad istražuje evoluciju detekcije objekata, s fokusom na jednoprolazne metode i dvoprolazne metode. Rad ističe razvoj ključnih algoritama kao što su R-CNN, Fast R-CNN, YOLO i SSD , te analizira njihove prednosti, ograničenja i primjene. Također se govori o važnosti javnih skupova podataka kao što su COCO, PASCAL VOC i ImageNet u poticanju napretka u ovom području. Pored toga, analizira se i uspoređuje rezultat YOLOv1 i R-CNN modela na malom skupu podataka.

Abstract (english)

This paper explores the evolution of object detection, focusing on single-pass and two-pass methods. It highlights the development of key algorithms such as R-CNN, Fast R-CNN, YOLO, and SSD, and analyzes their advantages, limitations, and applications. The importance of public datasets like COCO, PASCAL VOC, and ImageNet in driving progress in this field is also discussed. Additionally, the paper analyzes and compares the performance of the YOLOv1 and R-CNN models on a small dataset.

• Keywords: detekcija objekata; računalni vid; strojno učenje
• Keywords (english): object detection; computer vision; machine learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:983331
• 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: Open access
• Created on: 2024-10-12 22:26:26