Parameters Learning for Cloth Simulation

In virtual reality applications,it's difficult to tune the parameters of the cloth model to present vivid features of different fabric materials.In this paper,a method of learning parameters from real data is proposed.First,real data of the fabric motion are captured by motion capture devices.Then,parameters of the motion model are optimized through a genetic algorithm aiming to minimize the differences between simulated data and real data.Using this method,we obtained parameters regarding 6 different kinds of fabrics.Experiments showed that,with parameters learnt from real data,the relative errors per 5 seconds between simulation data and real data are below 6%.Furthermore,under the same external force,the simulated data of different fabrics could diversely feature the corresponding real fabrics well.