研究了一种基于关键帧的动画编辑方法.首先,运动数据关键帧序列将通过PCA压缩处理,将高维的运动数据降到低维空间;然后,在运动数据的内插处理阶段,通过将数据归一化以减小RBF神经网络的误差,并采用RBF神经网络将经过压缩的关键帧序列与高层运动参数建立联系.研究表明,该算法能以更高的效率进行运动数据编辑,有利于提高动画作品的制作效率,并降低制作成本.%In this paper, an animation technique based on PCA and RBF neural network is proposed. Firstly, the technique of PCA is utilized to reduce animation data. Secondly, a RBF neural network is employed to establish the relationship between the point set and the animation attributes parameters. Research results show that the proposed scheme can be effective for character animation applications.
展开▼