• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Image Attribute Changing Using Deep Machine Learning

Student: Denis Tsyupa

Supervisor: Denis Derkach

Faculty: Faculty of Computer Science

Educational Programme: Applied Mathematics and Information Science (Bachelor)

Year of Graduation: 2021

The rapid development of social networks based on photo and video sharing, such as Instagram and TikTok, has spurred progress in the development of computer vision algorithms for various kinds of complex cinematic effects. This can be achieved by replacing attributes in images, also referred to as the modality transformation task. Recent advances in generative models allow generating photorealistic images, which opens up many application possibilities, in this paper we consider their application in the task of replacing attributes in images. We briefly describe the main approaches and their ideas invented recently, highlighting their advantages and disadvantages. In this project, we have developed an algorithm for fast real-time attribute swapping, capable of running on low-cost hardware, based on the mentioned approaches

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