Vrednovanje tehnika rastresanja podataka za konvolucijske modele

master's thesis

• Title (english): Evaluation of jittering techniques for convolutional models
• Author: Natko Kraševac
• Mentor: Siniša Šegvić (mentor)
• Committee member: Siniša Šegvić (predsjednik povjerenstva)
• Committee member: Željka Mihajlović (član povjerenstva)
• Committee member: Tomislav Hrkać (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Duboki konvolucijski modeli su glavni sastojak mnogih praktičnih primjena računalnog vida. Međutim, takvi
modeli su ponekad nepraktični jer zahtijevaju velike skupove označenih podataka za učenje. Zbog toga
postupci za rastresanje podataka predstavljaju zanimljivo područje istraživanja. U okviru ovog rada proučene su i
implementirane metode za rastresanje podataka. Provedeni su eksperimenti i svaka od metoda je vrednovana i uspoređena s rezultatima
iz literature.

Abstract (english)

Deep convolutional models are the main instrument for many practical applications in the field of computer vision. However, these models are sometimes unpractical because they require great amounts of training data. Therefore, jittering and augmentation techniques impose themselves as interesting filed for reasearch. In this thesis, some of the recent techniques were researched and implemented. Experiments were conducted and each technique was evaluated and compared to results from the literature.,

• Keywords: rastresanje podataka; duboko učenje; konvolucijski modeli; klasifikacija; semantička segmentacija; cifar-10; camvid
• Keywords (english): data augmentation; deep learning; convolutional models; classification; semantic segmentation; cifar-10; camvid
• Language: croatian
• URN:NBN: urn:nbn:hr:168:839427
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 10:12:43