Kolorizacija slika korištenjem dubokog učenja

undergraduate thesis

• Title (english): Image colorization using deep learning
• Author: Luka Ivčević
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Zorana Doždor (član povjerenstva)
• Committee member: Iva Sović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad se bavi problemom automatske kolorizacije sivih slika. Kao i mnogi ostali problemi računalnog vida, problem je rješen upotrebom dubokog učenja. Dan je pregled drugih radova koji se bave istom tematikom i koji problem rješavaju na sličan. Opisuju se prostori boja i na koji način format ulaznih slika utječe na proces učenja modela. Mreža je konvolucijska duboka neuronska mreža i slične je arhitekture kao mreže u literaturi. Konačno, pregledaju se rezultati i mreža se koristi za kolorizaciju sivih slika iz povijesti za koje ne poznaje stvarni izgled.

Abstract (english)

This paper deals with the problem of automatically colorizing grayscale images. As many other problems in the field of computer vision, the solution is based on deep learning methods. Other papers which present similar solutions to the same problem are referenced. Color spaces and the influence of image format on the learning proces are explored. The implemented network is a convolutional nerual network with a similar architecture to those referenced in the paper. Finally, an overview of the results is given and the network is used to colorize historical images that have no known ground truth.

• Keywords: kolorizacija slika; kolorizacija sivih slika; računalni vid; duboko učenje; neuronske mreže
• Keywords (english): image colorization; grayscale image colorization; computer vision; deep learning; neural networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:662038
• 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: 2024-10-12 20:52:33