M. Ruhnke, B. Steder, G. Grisetti, and W Burgard.
Unsupervised Learning of Compact 3D Models Based on the Detection of Recurrent Structures.
In Proc. of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). October 2010.

Abstract

In this paper we describe a novel algorithm for constructing a compact representation of 3D laser range data. Our approach extracts an alphabet of local scans from the scene. The words of this alphabet are used to replace recurrent local 3D structures and which leads to a substantial compression of the entire point cloud. We optimize our model in terms of complexity and accuracy by minimizing the Bayesian information criterion (BIC). Experimental evaluations on large real-world data show that our method allows robots to accurately reconstruct environments with as few as 70 words.

BibTeX entry:

@inproceedings{ruhnke10iros,
  author = {Ruhnke, M. and Steder, B. and Grisetti, G. and Burgard, W},
  title = {Unsupervised Learning of Compact 3D Models Based on the Detection             
       of Recurrent Structures},
  booktitle = {Proc.~of the IEEE/RSJ International Conference on Intelligent Robots and
     Systems (IROS)},
  month = {October},
  year = {2010}
}