Aligning non-exact copies of artworks with their original

Bachelor thesis (2022)

Authors

T. Quin Electrical Engineering, Mathematics and Computer Science

Contributors

R.T. Wiersma Computer Graphics and Visualisation - EEMCS (supervisor 1)
E. Eisemann Computer Graphics and Visualisation - EEMCS (supervisor 1)
Ricardo Marroquim Computer Graphics and Visualisation - EEMCS (supervisor 1)
C. Lofi Web Information Systems - EEMCS (supervisor 2)
Faculty
Electrical Engineering, Mathematics and Computer Science
Copyright: © 2022 Tristan Quin
Convolutional neural network Image registration Image alignment Artwork Geometric transformation Semantic alignment
More Info
expand_more

Published Date

23-06-2022

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Abstract

This research investigates the efficacy and reliability of geometric matching for the specific case of aligning non-exact copies of artistic works with the original from which they were derived. The purpose of which is to provide a foundation for comparison in any further analysis conducted by conservators and art historians. An overview of the image alignment field as a whole is provided as well as a discussion of the particulars of aligning artworks. A variety of demonstrative results indicate the strengths and weaknesses of the technique in this domain and are accompanied by a discussion of the headline conclusions.