An Application of Markov Random Fields to Range Sensing James Diebel and Sebastian Thrun, Stanford University, Palo Alto, CA

Please see the research page for details on other projects.

Abstract— This paper describes a highly successful application of MRFs to the problem of generating high-resolution range images. A new generation of range sensors combines the capture of low-resolution range images with the acquisition of registered high-resolution camera images. The MRF in this paper exploits the fact that discontinuities in range and coloring tend to co-align. This enables it to generate high-resolution, low-noise range images by integrating regular camera images into the range data. We show that by using such an MRF, we can substantially improve over existing range imaging technology.

This paper was presented as a poster at the 19th Annual Conference on Neural Information Processing Systems (NIPS05) in December, 2005, in Vancouver, British Columbia.

	Paper: NIPS 2005 paper [PDF] [Bibtex].	8/27/2005
	Figure: Diagram showing the graph structure of the multi-resolution MRF used in this work.	8/27/2005
	Figure: Before and after shots of range data of a bookshelf in our robotics lab, with and without texture mapped on.	8/27/2005
	Figure: Before and after shots of range data of door frPublisith and without texture mapped on.	8/27/2005