In some registration applications additional user knowledge is available, which can improve and accelerate the registration process, especially for non-rigid registration. This is particularly important in the transfer of pre-operative plans to the operating room, e.g. for navigation. In case of tubular structures, such as vessels, a geometric representation can be extracted via segmentation and skeletonization. We present a new class of distance measures based on global filter kernels to compare such models efficiently with image data. The approach is validated in a non-rigid registration application with Powerdoppler ultrasound data. The importance and clinical use of 3D planning systems in liver surgery is increasing. First navigation systems based on intraoperative 3D ultrasound have been developed and clinically applied. Until now the transfer of preoperative models and plans to the patient in the operating room (OR) is mentally performed by the surgeon. Robust and fast methods are needed for a precise multi-modal non-rigid registration of the preoperative data and the intraoperative 3D ultrasound image volume.