The Robotics Institute

Carnegie Mellon Robotics Institute

Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Abstract

Most computer vision systems require accurate three-dimensional models. The problem of building such models from observations consists in taking multiple range image of the object from different viewing positions and orientations, referred to as "viewing poses", to match the data in the different images in order to recover the relative poses, and to merge the data into a single model using the estimated poses. The approaches proposed so far suffer from two major limitations. First, they require accurate knowledge of the relative viewing poses. Second, they either require a complicated feature extraction algorithm to be applied to the range image or they restrict the class of shapes that can be modelled. Our goal in this paper is to eliminate these two restrictions in order to allow modelling of natural, free-form objects from arbitrary unknown viewpoints.

Notes

Associated Center(s) / Consortia: Vision and Autonomous Systems Center

Text Reference

K. Higuchi, Martial Hebert, and Katsushi Ikeuchi, "Building 3-D Models from Unregistered Range Images," tech. report CMU-CS-93-214, Computer Science Department, Carnegie Mellon University, November, 1993

BibTeX Reference

@techreport{Hebert_1993_1371,
   author = "K. Higuchi and Martial Hebert and Katsushi Ikeuchi",
   title = "Building 3-D Models from Unregistered Range Images",
   booktitle = "",
   institution = "Computer Science Department",
   month = "November",
   year = "1993",
   number= "CMU-CS-93-214",
   address= "Pittsburgh, PA",
}

The Robotics Institute

Menu