Generiranje slika generativnim suparničkim mrežama

master's thesis

• Title (english): Generating Images Using Generative Adversarial Networks
• Author: Ante Šarić
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Zoran Kalafatić (član povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-09-18, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Generativne suparničke mreže su generativan model čiji se proces učenja temelji na teoriji igara. Sastoje se od dva suprotstavljena modela: generatora i diskriminatora. Generator pokušava zavarati diskriminatora generirajući realistične lažne uzorke, dok diskriminator pokušava razlikovati lažne uzorke od pravih. U ovom radu opisano je i implementirano nekoliko odabranih generativnih suparničkih modela. Trenirali smo modele nad tri skupa podataka: LSUN Church, MNIST i CIFAR-10. Rezultati su prikazani slikama i tablicama.

Abstract (english)

Generative adversarial networks are a generative model with a game theory-based learning process. They consist of two opposed models: the generator and the discriminator. The generator attempts to fool the discriminator by generating realistic fake samples, while the discriminator attempts to tell the fake samples from the real ones. In this thesis we describe and implement a few selected generative adversarial models. We've trained the models on three datasets: LSUN Church, MNIST i CIFAR-10. The results are presented with images and tables.

• Keywords: Generativne suparničke mreže; duboke konvolucijske neuronske mreže; kvadratni gubitak; uvjetne generativne suparničke mreže
• Keywords (english): Generative adversarial networks; deep convolutional neural networks; least squares loss; conditional generative adversarial networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:756164
• 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 10:47:21