The University of Edinburgh -
Division of Informatics
Forrest Hill & 80 South Bridge


Research Paper #601

Title:Computational Properties of Rotation Parametrizations
Authors:Borges,D; Fisher,RB
Date: 1992
Presented:Presented at the IX Brazilian Symposium on AI, Rio de Janeiro 1992
Keywords:
Abstract:There are various ways of representing a 3-D rotation in Euclidean space and some of them can be more sensitive to errors than others. We compare five well known parameterizations under circumstances encountered in image understanding systems by introducing perturbations on the parameters of each representation and measuring their influence on the output rotation. The exponential representation is found to be the least sensitive to these perturbations.
Download:NO ONLINE COPY


[Search These Pages] [DAI Home Page] [Comment]