3D Model Feature Extraction Algorithm Based on View Distance
Jing Zhou · Journal of Chinese Computer Systems · 2007
In 3D model retrieval,there is a major concern about how to extract features from 3D models.This paper introduces a new feature extraction algorithm based on view distance.This algorithm uses information about distance between normalized model face and the view point to generate six distance image,and then it exploits 2D Fourier transform to obtain 3D model feature vector.This method overcomes the drawbacks of the 3D projection-based silhouette algorithm concerning information lost in space and image interior.The experiment shows that the new algorithm is much better than the silhouette algorithm in retrieval precision.