|M.Sc Student||Lavva Irina|
|Subject||Robust Methods for Geometric Primitive Recovery and|
Estimation from Range Images
|Department||Department of Industrial Engineering and Management||Supervisor||Professor Ilan Shimshoni|
|Full Thesis text|
We present a method for the recovery of partially occluded 3D geometric primitives from range images which might also include non-primitive objects. The method uses a technique for estimating the principal curvatures and Darboux frame from range images. After estimating the principal curvatures and the Darboux frames from the entire scene, a search for the known patterns of these features in geometric primitives, is performed. If a specific pattern is identified then the presence of the corresponding primitive is confirmed using these local features. The features are also used to recover the primitive's characteristics. The suggested application is very efficient since it combines the segmentation, classification and fitting processes, which are part of any recovery process, in a single process, which advances monotonously through the recovery procedure. We view the problem as a robust statistics problem and we therefore use techniques from that field. A Mean-Shift based algorithm is used for robust estimation of shape parameters, such as recognizing which types of shapes in the scene exist and after that full recovery of planes, spheres and cylinders. A RANSAC based algorithm is used for robust whole model estimation for the more complex primitives, like cones and tori. In the final step of our algorithm a new method to remove false detections is also presented. The methods have been tested on series of real complex cluttered scenes, yielding accurate and robust recoveries of primitives.