Combining Contour and Shape Primitives for Object Detection and Pose Estimation of Prefabricated Parts

Alexander Berner, Jun Li, Dirk Holz, Joerg Stückler, Sven Behnke, and Reinhard Klein
In proceedings of IEEE International Conference on Image Processing (ICIP), Melbourne, Australia, Sept. 2013
 

Abstract

Man-made objects such as mechanical construction parts can typically be described as a composition of shape primitives like cylinders, planes, cones and spheres. We propose a robust method for the detection and pose estimation of such objects in 3D point clouds. Our main contribution is to enhance a probabilistic graph-matching approach that detects objects using 3D shape primitives with distinct 2D primitives such as circular contours. With this extension, our method copes with difficult occlusion situations and can be applied for object manipulation in complex scenarios such as grasping from a pile or bin-picking. We demonstrate the performance of our approach in a comparison with a state-of-the-art feature-based method for objects of generic shape and a primitive-based approach using only 3D shapes and no contours.

Images

Download Paper

Download Paper

Bibtex

@INPROCEEDINGS{bernerli2013,
     author = {Berner, Alexander and Li, Jun and Holz, Dirk and St{\"u}ckler, Joerg and Behnke, Sven and Klein,
               Reinhard},
      title = {Combining Contour and Shape Primitives for Object Detection and Pose Estimation of Prefabricated
               Parts},
  booktitle = {IEEE International Conference on Image Processing (ICIP), Melbourne, Australia},
       year = {2013},
      month = sep,
   abstract = {Man-made objects such as mechanical construction parts can typically be described as a composition
               of shape primitives like cylinders, planes, cones and spheres. We propose a robust method for the
               detection and pose estimation of such objects in 3D point clouds. Our main contribution is to
               enhance a probabilistic graph-matching approach that detects objects using 3D shape primitives with
               distinct 2D primitives such as
               circular contours. With this extension, our method copes with difficult occlusion situations and can
               be applied for object manipulation in complex scenarios such as grasping from a pile or bin-picking.
               We demonstrate the performance of our approach in a comparison with a state-of-the-art feature-based
               method for objects of generic shape and a primitive-based approach using only 3D shapes and no
               contours.}
}