Multi-scale Salient Feature Extraction on Mesh Models
We present a new method of extracting multi-scale salient features on meshes. It is based on robust estimation of curvature on multiple scales. The coincidence between salient feature and the scale of interest can be established straightforwardly, where detailed feature appears on small scale and feature with more global shape information shows up on large scale. We demonstrate this multi-scale description of features accords with human perception and can be further used for several applications as feature classification and viewpoint selection. Experiments exhibit that our method as a multi-scale analysis tool is very helpful for studying 3D shapes.
KeywordsSalient Feature Principal Curvature Mesh Model Shape Retrieval Ball Neighborhood
Unable to display preview. Download preview PDF.
- 1.Tangelder, J.W., Veltkamp, R.C.: A survey of content based 3d shape retrieval methods. SMI, 145–156 (2004)Google Scholar
- 5.Chen, X., Saparov, A., Pang, B., Funkhouser, T.: Schelling points on 3D surface meshes. ACM Trans. Graph. 31(3) (2012)Google Scholar
- 6.Yang, Y.L., Lai, Y.K., Hu, S.M., Pottmann, H.: Robust principal curvatures on multiple scales. In: Symposium of Geometry Processing, pp. 223–226 (2006)Google Scholar
- 8.Johnson, A.: Spin-Images: A Representation for 3-D Surface Matching. PhD thesis, Robotics Institute, Carnegie Mellon University, Pittsburgh, PA (August 1997)Google Scholar