Evaluation of Stereosys vs. other Stereo Systems
SRI INTERNATIONAL MENLO PARK CA ARTIFICIAL INTELLIGENCE CENTER
Pagination or Media Count:
As previously reported Fischler, 1984, SRI International is implementing a complete, state-of-the-art stereo system that will produce dense three-dimensional 3-D data from stereo pairs of intensity images. This system forms a framework for much of our stereo research and will be a base component of our planned expert system for 3-D compilation. Ideally, we would assess the capabilities of our system by running it on a data set that has known ground truth against which to compare our results. Unfortunately, such data sets do not currently exist because of the extremely high cost of the ground work necessary to measure terrain elevations accurately for a close spacing and to assess the heights of all vegetation and buildings in the area. Lacking such a data set, we can only compare our results against those produced by other stereo systems or against the perceptions of a human looking at the same imagery in stereo on a CRT. To test our system, currently called STEREOSYS, we have run it on several data sets, including two for which we also have results produced by the DIMP system at the U.S. Army Engineer Topographic Laboratories ETL. Comparing our matching results to DIMP results or to human perception of what the correct match should be, we have begun to assess the strengths and weaknesses of STEREOSYSs matching techniques, as well as accumulating a catalog of examples of difficult areas for matching Hannah, 1985. A description of the experiments that we have conducted and our preliminary conclusions regarding the accuracy of the system are set forth in this report.