Realistic Video Post-Processing of Rendered 3D Scene

Student: Raskevich Aleksandr

Supervisor: Ilya Makarov

Educational Programme: Data Science (Master)

Year of Graduation: 2020

In this work, we tried to use GAN's to improve the graphics in old video games to a model of its more modern version. As an old game, I took The Witcher 1 (2007) and its more modern version of The Witcher 3 (2015). We used different GAN's architectures: DiscoGan, CoGAN, CycleGAN and U-GAT-IT. We will also try to develop our own architecture, which will solve the task better than current solutions. The following assessment methods were used to control the quality of the obtained images: Inception score, Frechet-Inception distance and Kernel-Inception distance. To train GAN's, we used our dataset, which consists of 100k images of The Witcher 1 and about 500k images of The Witcher 3.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses