Most Likely Separation of Intensity and Warping Effects in Image Registration

TL;DR

This paper presents a nonlinear mixed-effects model for joint estimation of intensity and warping effects in 2D image registration, improving accuracy by simultaneous modeling rather than sequential preprocessing.

Contribution

It introduces a novel class of mixed-effects models for jointly estimating intensity and warp variations in images, with an efficient algorithm for fitting the model.

Findings

Successful application to facial and brain MRI datasets

Simultaneous estimation improves registration accuracy

Avoids bias from traditional sequential registration methods

Abstract

This paper introduces a class of mixed-effects models for joint modeling of spatially correlated intensity variation and warping variation in 2D images. Spatially correlated intensity variation and warp variation are modeled as random effects, resulting in a nonlinear mixed-effects model that enables simultaneous estimation of template and model parameters by optimization of the likelihood function. We propose an algorithm for fitting the model which alternates estimation of variance parameters and image registration. This approach avoids the potential estimation bias in the template estimate that arises when treating registration as a preprocessing step. We apply the model to datasets of facial images and 2D brain magnetic resonance images to illustrate the simultaneous estimation and prediction of intensity and warp effects.