No abstract

This paper describes a new functional generation method for human motion with emotion. In the method, the empirical data of the human joint angles in some typical motions with emotions are utilized. Based on the frequency analysis, the data are then classified into two groups, the basic factor and the additional factor. The former factor corresponds to a basic motion like "walking", whereas the Iatter means a descriptive part of the basic motion like "brisk" or "fast". Then, apart from the analysed emprical motion, more complex human motions with emotions are generated by interpolation andlor extrapolation of these factors in the frequency-phase domain. The sequences of the short animations obtained will illustrate the power of the method.

No abstract

A recognition of human walking motion by using an accelerometer.

No abstract