Multi-Step Procedures for the Localization of 2D and 3D Point Landmarks and Automatic ROI Size Selection

In this contribution, we are concerned with the detection and re ned localization of 3D point landmarks. We propose multi-step di erential procedures for subvoxel localization of 3D point landmarks. Moreover, we address the problem of choosing an optimal size for a region-ofinterest ROI around point landmarks. That is, to reliably localize the landmark position, on the one hand, as much as possible image information about the landmark should be incorporated. On the other hand, the ROI should be restricted such that other structures do not interfere. The multi-step procedures are generalizations of an existing two-step procedure for subpixel localization of 2D point landmarks. This two-step procedure combines landmark detection by applying a di erential operator with re ned localization through a di erential edge intersection approach. In this paper, we investigate the localization performance of this two-step procedure for an analytical model of a Gaussian blurred L-corner. The results...

• Type: Conference
• Year: 1998
• Where: ECCV
• Authors: Sönke Frantz, Karl Rohr, H. Siegfried Stiehl
• Keywords: 2D Point Landmarks, Computer Vision, ECCV 1998, Edge Intersection Approach, Multi-step Di, Multi-step Procedures, Point Landmark Localization