Data-Driven Evaluation of Non-Rigid Registration via Appearance Modelling

This paper presents a generic method for assessing the quality of non-rigid registration (NRR) algorithms, that does not depend on the existence of any ground truth, but depends solely on the data itself. The data is taken to be a set of images. The output of any non-rigid registration of such a set of images is a dense correspondence across the whole set. Given such a dense correspondence, it is possible to build a generative statistical model of appearance variation across the set. Evaluating the quality of the registration algorithm is hence mapped to the problem of evaluating the quality of the resultant statistical model; that is, when the model is compared to the image data from which it was generated. It should be noted that this approach does not depend on the specifics of the registration algorithm used or on the specifics of the modelling approach used. We derive indices of model specificity and generalisation that can be used to assess the quality of such models. This approach is validated by comparing our assessment of registration quality with that derived from ground-truth anatomical labelling. We demonstrate that not only is our approach capable of reliably assessing NRR without ground truth, but it is also more sensitive than the ground-truth-dependent approach. Finally, to demonstrate the practicality of our method, different NRR algorithms – both pairwise and groupwise– are compared in terms of their performance on MR brain data.