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

The Program for Image Inpainting in Chosen Area

Student: Anastasiia Ponomareva

Supervisor: Rimma Akhmetsafina

Faculty: Faculty of Computer Science

Educational Programme: System and Software Engineering (Master)

Year of Graduation: 2019

Nowadays restoration of photos – both physical and digital – is very popular. It can be done manually by professionals; however, it requires special skills and equipment. This paper is dedicated to image inpainting – semi-automated way to fix damaged pictures and repair chosen areas there. In this work, basic groups of methods for image inpainting are studied, their advantages and drawbacks are mentioned. Also, several concrete algorithms - based on different approaches - have been implemented, so their features are described. Besides, results of experiments done on various images (with different resolution, different gap size) are presented and assessed. All tests were held on the same equipment without using GPU, so comparison gives an answer on what can and cannot be used in terms of lack of computational resources. Algorithms are implemented using Python and its frameworks PIL and PyTorch, which made work with image handling and machine learning easier. For common users, an application for demonstration purposes is implemented. It is made using Django framework for backend and Javascript, HTML and CSS for user interface. This paper contains 56 pages, 4 chapters, 23 illustrations, 28 bibliography items, 9 appendices. Keywords: computer vision, image inpainting, gap, textural inpainting, structural inpainting, deep prior image inpainting.

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