An efficient unsupervised diffusion clustering algorithm with application to shape decomposition based on visibility context

作者：

Highlights：

• DCA: a fast clustering algorithm for 2D and 3D shape decomposition.

• Graph based visibility is efficiently diffused, capturing nearly convex parts.

• Closely related to clique graphs, perceptual rule for convexity is also implied.

摘要

Highlights•DCA: a fast clustering algorithm for 2D and 3D shape decomposition.•Graph based visibility is efficiently diffused, capturing nearly convex parts.•Closely related to clique graphs, perceptual rule for convexity is also implied.

论文关键词：Graph clustering,Shape decomposition,Visibility,Diffusion

论文评审过程：Received 18 June 2016, Revised 29 December 2016, Accepted 30 December 2016, Available online 1 January 2017, Version of Record 18 January 2017.

论文官网地址：https://doi.org/10.1016/j.image.2016.12.012