Skip to main content
. Author manuscript; available in PMC: 2014 Feb 26.
Published in final edited form as: Med Image Comput Comput Assist Interv. 2013;16(0 3):211–218. doi: 10.1007/978-3-642-40760-4_27

Fig. 2.

Fig. 2

Gaussian process segmentation of parotid gland. The initial label from the atlas-based segmentation only partially agrees with the manual segmentation. We extract contours from the image and use them in the kernel function k that allows us to sample label maps L~GP(0,k), supported by the image. Conditioning these on the atlas labels results in an improved segmentation.