A library for fast FFT-computed global mutual information-based rigid alignment using the GPU.
Requirements:
Python3.8.8, PyTorch 1.8.1, numpy, scipy, sklearn
(CUDA-compatible GPU for GPU acceleration.)
Related to the article (if you use this code, please cite it):
Johan Öfverstedt, Joakim Lindblad, and Nataša Sladoje. Fast computation of mutual information in the frequency domain with applications to global multimodal image alignment. Pattern Recognition Letters, Vol. 159, pp. 196-203, 2022. doi:10.1016/j.patrec.2022.05.022
Preprint: https://arxiv.org/abs/2106.14699
To use the library, please see the included example script "example.py".
Author of the code:
Johan Öfverstedt