Institutional Repository of Vilnius University
Mažos skiriamosios gebos taškinės grafikos paveikslų gavimas pasitelkiant neuroninius tinklus /
Abstract
dc:descriptionBy doing this work attempt was made to review how GAN and transformer works, a few of their modifications, to create a model capable of generating from high quality images pixelart. While the experiments was is progress the model was being made, by using generator and discriminator principles, to create a model to generate out images, by using two related datasets (one consisting pixel-art, and the other was transformtion of pixel art to high quality images). In this work was researched some information about generative adversial networks modifications and their working principles - StyleGAN and its few modeifications, CycleGAN, Gansformer and Deep Unsuppervised Pixelization models. In this work the review was made on mentiones GAN and transformer networks development, various new discoveries with this models. Also the analysis was made on already existing GAN models and their upgrades. Therefore there is some research made on this exact tast (transforrming high quality images to pixel-art) this work was ntrying to modificate existing models, to reach higher effiziency, by researching with new upgraded GANs. For this work images was taken from Johanes Koph work example generated images from article ”Depixelizing pixel-art”.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Puplinskas, Eligijus,
- Contributors dc:contributor
-
- Petkevičius, Linas
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD107115746&prefLang=en_US