Characterizing 2D Flow Fields with Zernike Polynomials

To identify the content of motion sequences such as gestures and video sequences, we propose a compact representation of flow fields using a basis of complex Zernike polynomials. Moments derived from projection of these polynomials have been used in recognition. Previously, local descriptions learned from examples, or specified a priori, have been used to identify local flow events. Affine expansions can describe local flows such as rotation or expansion, but cannot model extended flows such as those caused by motion of people or in gestures. The motion will be characterized by the trajectory of the center of motion, without tracking component parts, and the relative flow with respect to the center. A basis set composed of Zernike polynomials represents the components of flow. Tests show that the representation derived from real images agrees with our predicted values. The representation is sufficient for recognition of periodic motion. As well, individual gestures are well represented. We analyze how the representation varies with location and scale.