Image comparison and scaling via nonlinear elasticity
Analysis of PDEs
2023-03-21 v2 Computer Vision and Pattern Recognition
Abstract
A nonlinear elasticity model for comparing images is formulated and analyzed, in which optimal transformations between images are sought as minimizers of an integral functional. The existence of minimizers in a suitable class of homeomorphisms between image domains is established under natural hypotheses. We investigate whether for linearly related images the minimization algorithm delivers the linear transformation as the unique minimizer.
Cite
@article{arxiv.2303.10103,
title = {Image comparison and scaling via nonlinear elasticity},
author = {John M. Ball and Christopher L. Horner},
journal= {arXiv preprint arXiv:2303.10103},
year = {2023}
}
Comments
SSVM2023 Proceedings to appear. New references added plus related minor changes