The aim of this thesis is to introduce a new descriptor for the action recognition in video. Action recognition is considered as a combination of human action representation, and subsequent dir=ltr A new invariant action descriptor based upon spherical harmonics, is introduced to describe the STV. The generalizations of Fourier expansion of periodic functions on the line, and polar coordinate representation of function in the plane, to three dimensions lead to the theory of spherical harmonics. Spherical harmonic basis functions are constructed on a unit sphere with two parameters in the spherical coordinate system. To describe surfaces regardless of whether they are stellar or not, spherical harmonics in its parametric form is used. It is shown that the three sets of coefficient ( ) can completely define the shape. In this thesis, it is discussed how spherical harmonics are invariant with respect to transl Keywords: Computer vision, Action Recognition, Space-time volume, Spherical Harmonics
