I. The data characterization stage: a general picture of the situation This part I proposes two ways for the design of the data characterization stage evaluation : 1) a way based on the computation of distances between signals and distances between indicators yielded by the characterization stage, 2) a way based on individual opinions about the signal resemblance and dissemblance. These two investigation ways are considered through a didactic example of 10 signals with simulated values. Despite the first way seems "objective", we show that it is the contrary as it is the case is in each paper about data analysis. In fact, the subjectivity is always present but, more or less, the main problem being to be aware of this ! II. The design of the space-time windows and the subsequent analysis This part II proposes two examples where the space-time windowing is useful: gait and load lifting. Giving the main importance of the windowing and graphing result (for a very first analysis of numerous multidimensional signals), the discussion tackle subjective aspects with and without windowing.
© Copyright 2006 Ninth International Symposium on the 3-D Analysis of Human Movement, June 28-30, 2006, Valenciennes, France. Alle Rechte vorbehalten.