Abstract
In this paper we present new insights in methods to solve the orientation representation problem in arbitrary dimensions. The gradient structure tensor is one of the most used descriptors of local structure in multi-dimensional images. We will relate its properties to the double angle method in 2D and the Knutsson mapping in three or higher dimensions. We present a general scheme to reduce the dimensionality of the mappings needed to solve the orientation representation problem and derive some properties of these reduced mappings. © Springer-Verlag Berlin Heidelberg 2003.
Cite
CITATION STYLE
Rieger, B., & Van Vliet, L. J. (2003). Representing orientation in n-dimensional spaces. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2756, 17–24. https://doi.org/10.1007/978-3-540-45179-2_3
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
