Author :
Ha, Victor H S ; Moura, Josd M E
Author_Institution :
Digital Media R&D Center, Samsung Electron., Suwon, South Korea
Abstract :
Everybody has a common sense notion of the shape of an object. This notion is invariant to rigid motions of the object, such as translations, rotations, and reflections. In many applications, however, the object is viewed through a deforming transformation so that the object appears to have a different shape. The concept of intrinsic shape of an object as an invariant to a more general class of deforming transformations and then recovering this invariant from the actual data available is described briefly.
Keywords :
pattern recognition; transforms; deforming transformation; intrinsic object shape; pattern recognition; reflections; rigid object motions; rotations; translations; Discrete Fourier transforms; Discrete wavelet transforms; Geometry; Humans; Layout; Object detection; Pattern recognition; Reflection; Research and development; Shape;
Conference_Titel :
Signals, Systems and Computers, 2002. Conference Record of the Thirty-Sixth Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-7803-7576-9
DOI :
10.1109/ACSSC.2002.1197164
