Feature Detection Using Curvature Maps and the Min-Cut/Max-Flow Algorithm

Feature detection. A longer version can be found <a href=\"http://www.cse.seas.wustl.edu/techreportfiles/getreport.asp?510\">here</a>. Automatic detection of features in three-dimensional objects is a criticalpart of shape matching tasks such as object registration and recognition. Previousapproaches often required some type of user interaction to select features.Manual selection of corresponding features and subjective determination of thedifference between objects are time consuming processes requiring a high levelof expertise. The Curvature Map represents shape information for a point and itssurrounding region and is robust with respect to grid resolution and mesh regularity.It can be used as a measure of local surface similarity.We use these curvaturemap properties to extract feature regions of an object. To make the selection of thefeature region less subjective, we employ a min-cut/max-flow graph cut algorithmwith vertex weights derived from the curvature map property. A multi-scale approachis used to minimize the dependence on user defined parameters. We showthat by combining curvature maps and graph cuts in a multi-scale framework, wecan extract meaningful features in a robust way. Feature detection, min-cut max-flow

Download paper here

authors: Timothy Gatzke and Cindy Grimm

Authors: Timothy Gatzke and Cindy Grimm
Download Paper