3D model feature extraction algorithm based on shape weighted

Feature extraction is the focus of the 3D model retrieval research. Recent years, some methods for feature extraction have been proposed. Although, the extraction algorithms are increasingly complex and computational, recall and precision rate is only improved slowly. This paper proposes a novel 3D model feature extraction algorithm based on shape weighted, taking into account the distance between center point and the sample point, as well as the area of the corresponding triangular patches, and building a histogram for feature descriptor. The presented algorithm takes full advantage of the model surface information and its calculation is simple and less time consuming. Experiments show that the algorithm improves recall and precision effectively.