Registration-based High Order Slice Interpolation

Given a 3D image, i.e. a stack of 2D images or the frames of a movie, this algorithm interpolates intermediate images.
Interpolation is achieved by registering the pixels of two subsequent images and globally (between all pairs of subsequent images) spline interpolating the registered displacement fields. Along the spline trajectory of each pixel the intensities are also spline interpolated.
The method works best if the subsequent images are not too different.
This algorithm can be used for example to resample 3D images with anisotropic resolution to isotropic resolution.
If you want to use this algorithm for a publication, please consider citing:
Horváth A., Pezold S., Weigel M., Parmar K., Cattin P. (2017) High Order Slice Interpolation for Medical Images. In: Tsaftaris S., Gooya A., Frangi A., Prince J. (eds) Simulation and Synthesis in Medical Imaging. SASHIMI 2017. Lecture Notes in Computer Science, vol 10557. Springer, Cham

https://link.springer.com/chapter/10.1007/978-3-319-68127-6_8