One of the challenges MRF faces is the amount of data needed to be stored, loaded, and processed, especially when a high resolution dictionary is needed or large multi-dimensional analyses need to be taken into account. A low rank approximation to the high resolution MRF dictionary using a coarse dictionary is an effective remedy to this problem. Here we present one of many possible ways to implement low rank approximation to an arbitrary fine MRF dictionary by a coarse dictionary equipped with polynomial fitting, so as to avoid the need of pre-calculating, storing, and processing the large, finely-resolved MRF dictionary.
This abstract and the presentation materials are available to members only; a login is required.