PDF version of this entire document

Groupwise MDL Algorithm

We have previously described a groupwise method which uses a Minimum Description Length (MDL)formulation [16]. The main idea is that the complete set of images can be encrypted as a coded message, and the description length  [22] in bits used as an objective function. Rather than encoding the raw images, the encoding uses an appearance model, built using the estimated correspondences, to approximate the data; the encoding needs also to include details of the model itself and of the discrepancy between each image and its model approximation. As the registration proceeds, the correspondences, and hence the appearance model, are continually updated so as to minimise the description length.

Figure: Overlap measures (with corresponding 95#95 one standard error errorbars) for the MGH dataset as a function of the degree of degradation of registration correspondence, 79#79. The various graphs correspond to the various tissue weightings as defined in Section .

96#96

Figure: Generalisation & Specificity for various definitions of image distance (varying shuffle radius) with corresponding 95#95 one standard error errorbars as a function of the degree of degradation of the registration correspondence 79#79 for the MGH dataset

[Generalisation]97#97 [Specificity]98#98