manual | faq | input formats | validation | credits | changelog | feedback

This page describes the input formats that LDDMM-Landmark can operate on.

The format for an input landmark file is as follows:

Where

Landmarks-1.0 N Label_1 Landmark 1 . . Label_i Landmark i . . Label_N Landmark N

is the number of landmarks in the file*N*is a label for the ith landmark with no whitespaces. Note that the label is not important for the matching; the order of the landmarks is what counts.*Label_i*is the ith landmark. The format should be:*Landmark i*

Consists of 3 whitespace-delimited double-precision floating point numbers x, y, z, a "valid" flag (0/1), and a dummy number (usually set to 1).*x y z valid dummy*

The format for an input landmark file is as follows:

Where

LandmarksWithVariability N Label_1 Landmark 1 . . Label_i Landmark i . . Label_N Landmark N

*LandmarksWithVariability*is the string "LandmarksWithVariability"is the number of landmarks in the file*N*is a label for the ith landmark with no whitespaces. Note that the label is not important for the matching; the order of the landmarks is what counts.*Label_i*is the ith landmark and its covariance data. The format should be:*Landmark i*

Consists of 3 whitespace-delimited double-precision floating point numbers x, y, z, a "valid" flag (0/1), a dummy value (usually 1), and the 3x3 covariance matrix.*x y z valid dummy cov1 cov2 ... cov9*

The format for an landmark file written out by LDDMM-Similitude, which can be used as input to LDDMM-Landmark is as follows:

N x 3[ x_1 y_1 z_1 x_2 y_2 z_2 . . x_N y_N z_N ]

The format for an landmark file written out by LDDMM-Similitude, to which covariance values can be added and used as input to LDDMM-Landmark is as follows:

N x 3[ x_1 y_1 z_1 x_2 y_2 z_2 . . x_N y_N z_N ] [ c_1_1 c_1_2 ... c_1_9 c_2_1 c_2_2 ... c_2_9 . . c_N_1 c_N_2 ... c_N_9 ]

*Last Modified*: Monday, 25th April, 2011 @ 11:20am