Generative Adversarial Networks for Image Transformations

Student: Kaglinskaya Mariya

Supervisor: Alexey Umnov

Educational Programme: Applied Mathematics and Information Science (Bachelor)

Year of Graduation: 2018

Recent success in neural networks application allows to solve a wide range of challenging tasks. However, training deep neural network architectures usually requires large labelled datasets. Not for all tasks datasets exist, while collecting and labelling new data is expensive and time-consuming. Due to these reasons it is has become more popular to train models on synthetic images. However, learning from synthetic images may give poor results on real data due to the gap between real and synthetic data distributions. To bridge this gap, this work is aimed to consider and compare two models based on generative adversarial network architecture to provide a method for improving the quality of generated images: Cycle GAN and Simulated and Unsupervised learning method.

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