Detekcija objekata na slikama

undergraduate thesis

• Title (english): Object Detection
• Author: Tomislav Kurtović
• Mentor: Tomislav Pribanić (mentor)
• Committee member: Tomislav Pribanić (predsjednik povjerenstva)
• Committee member: Kristijan Bartol (član povjerenstva)
• Committee member: David Bojanić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-09, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad predstavlja neke arhitekture modela koji se koriste prilikom detektiranja
objekata. Razvoj ove domene u zadnjih par godina bio je značajan, što potvrd̄uje složenost i preciznost dostupnih modela. RCNN-a bio je prvi značajan model u ovom
području. Modeli koji su kasnije razvijeni su YOLO i SSD modeli i oni su pomaknuli
granice i u brzini i u preciznosti detektiranja. Uključenost tvrtke Google u razvoj i distribuciju unaprijed treniranih modela takod̄er čini postupak treniranja... More jednostavnijim
za korisnika, eliminirajući brigu o konfiguraciji arhitekture. Nakon treniranja modela
slijedi implementacijski postupak u kojem se objašnjavaju svi važni koraci koji su
omogućili pokretanje modela unutar mobilne platforme. Less

Abstract (english)

This paper presents some model architectures used in the field of object detection.
The development of this domain in the last couple of years has been significant, which
the complexity and precision of the available models confirm. RCNN was the first
significant model in this area. After it other models such as the YOLO and SSD model
were developed and have shifted boundaries both in speed and in detection accuracy.
Also, Google’s involvement in the development and distribution of... More pretrained models
makes the training process easier for the user, eliminating the concern about architecture
configuration. After training the model follows the implementation process in
which all the important steps that enabled the launch of the model within the mobile
platform are explained. Less

• Keywords: Android; Tensorflow; Flutter; SSD MobileNet
• Keywords (english): Android; Tensorflow; Flutter; SSD MobileNet
• Language: croatian
• URN:NBN: urn:nbn:hr:168:952374
• 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: 2023-01-14 09:15:27