ABSTRACT
There are several meanings of the word shape. Although the use of the words “shape” or “shape analysis” is very common, the definition is seldom made precise in a mathematical sense. According to the Oxford English Dictionary, it means “the external form or appearance of someone or something as produced by their outline”. Kendall [7, 17] described shape as a mathematical property that remains unchanged under certain (shape-preserving) transformations such as rotation, translation, and global scaling. Shape analysis seeks to represent shapes as mathematical quantities, such as vectors or functions, that can be manipulated using appropriate rules and metrics. Statistical shape analysis is concerned with quantifying shape as a random object and
in Bayesian
FIGURE 17.1 Statistical shape analysis of surfaces should be invariant to translation, scale, rotation and re-parameterization.
