Classification of 3D models for the 3D animation environments

Classification the 3D models in the graphical environment is a key problem with applications in computer graphics, virtual reality, especially the intelligent virtual human for humanoid animation. The challenging aspects of this problem are to find a suitable shapes´ feature that can be used to compare them quickly and a proper method to classify the models on the features. We propose a method of classifying shapes´ features for surface-based 3D shape models based on their shape similarity. The features of shape of 3D models are computed by first converting an input surface based model into an oriented point set model and then computing features histograms of distance and shape distributions. Then, support vector machines (SVM) are used to classify the features of the models. By the classification the models can be given semantic meanings in the 3D environment.