A Shape Distributions Retrieval Algorithm of 3D CAD Models Based on Normal Direction
This paper describes a new 3D model retrieval approach based on shape distributions, which represents a 3D shape as three D2 shape histograms. First, generate a sufficiently large number of random sample points on surface of 3D model and note the normal of each point. Second, we adopt the D2 shape function to measures the distance between two random points on the surface of the model, at the same times, compute the angles between the normals of the points and the line segment which is formed by the two random points. Last, according to the angles, we sort the line segments into three sets. For each set, we calculate shape distribution histograms associated with the D2 shape distribution function. Compare each pair of shape distribution histograms using well-known curve matching techniques and add the three weighted results. The sum is regarded as the similar coefficient of the two models. Experimentation shows this algorithm can efficiently give the similar degree of the two models and as well reflecting human perceptual similarity.
3D model retrieval shape distributions shape matching content-based retrieval.
Hongshen Wang Shusheng Zhang Kaixing Zhang Xiaoliang Bai
The Key Laboratory of Contemporary Designing and Integrated Manufacturing Technology,Ministry of Edu The Key Laboratory of Contemporary Designing and Integrated Manufacturing Technology,Ministry of Edu
国际会议
The 9th International Conference for Young Computer Scientists(第九届国际青年计算机大会)
安徽黄山
英文
891-896
2008-11-18（万方平台首次上网日期，不代表论文的发表时间）