Dear Helmi, PCA is quite commonly used on various biomechanical data [e.g. Moro, F. L., Spröwitz, A., Tuleu, A., Vespignani, M., Tsagarakis, N. G., Ijspeert, A. J. & Caldwell, D. G. 2013 Horse-like walking, trotting, and galloping derived from kinematic Motion Primitives (kMPs) and their application to walk/trot transitions in a compliant quadruped robot. Biol Cybern (doi:10.1007/s00422-013-0551-9)] although it would probably be better to use some derived property such as the joint centre locations rather than the raw marker positions. That way you could compare your data with other experiments where different methodologies have been used for the motion capture, and use some of the slightly more robust marker cluster approaches. Error is always going to be a problem simply because motion capture isn't terribly accurate.
Cheers Bill -- Bill Sellers, T: 0161 275 1719, M: 0785 7655786, [email protected] University of Manchester, D1239 Michael Smith Building, Manchester, M13 9PT. On 24 Aug 2014, at 03:34, Helmi Hadi <[email protected]> wrote: > Hi, > > I was wondering if GMM analysis could be done on motion tracking subjects? > Subjects in motion tracking usually wear metallic studs which appear on like > stick figures on the computer. As GMM analyses points, will it not be easier > to just use the points generated by the motion tracking studs? > > I can see a problem where the placement of the studs on the subject may not > be accurate and registering the whole body may prove problematic. > > Thank you > > Regards > Helmi Hadi > Universiti Sains Malaysia > > -- > MORPHMET may be accessed via its webpage at http://www.morphometrics.org > > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
