Interactive motion synthesis with optimal blending

Abstract

In this paper, we propose an interactive motion synthesis technique that synthesizes a continuous motion sequence from given elementary motions. A user can either specify the execution timings of motions or execute motions in a sequence with automatically determined execution timings. Our method is based on a previous approach that determined the appropriate synthesis method and blending range for each pair of sequential motions, while considering the constraints between the foot and ground in the elementary motions to prevent foot sliding. However, using only the foot‐ground constraints to determine the blending range may generate unnatural motions such as nonsmooth, too fast, or too slow transitions. Moreover, simple motion blending with a regular weight function can generate unnatural motions. To solve these problems, we have introduced an optimal blending range and a weight function, which are determined for each blending segment for the upper and lower body. We also introduced extensions for applying this method to interactive character control. We have successfully applied our method to both animation generation and interactive character control. Copyright © 2014 John Wiley & Sons, Ltd.

3D Dynamic Image Modeling Based on Machine Learning in Film and Television Animation

“…Motion capture data has high dimensionality, large amount of information, complex structure, spatiotemporal continuity and Riemannian manifold structure, all of which bring challenges to motion synthesis. Motion hybrid [12][13][14][15] is a simple and efficient motion synthesis model. Such methods first preprocess motion segments of the same type, including using the DTW algorithm to align them in time sequence, and then make each motion frame have similar spatial coordinates through linear transformation, that is, coordinate alignment.…”