Feature sensitive mesh segmentation
Authors:
Open Access
- 6 June 2006
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 17-25
Abstract
Segmenting meshes into natural regions is useful for model under- standing and many practical applications. In this paper, we present a novel, automatic algorithm for segmenting meshes into meaning- ful pieces. Our approach is a clustering-based top-down hierarchi- cal segmentation algorithm. We extend recent work on feature sen- sitive isotropic remeshing to generate a mesh hierarchy especially suitable for segmentation of large models with regions at multiple scales. Using integral invariants for estimation of local characteris- tics, our method is robust and efficient. Moreover, statistical quanti- ties can be incorporated, allowing our approach to segment regions with different geometric characteristics or textures.Keywords
This publication has 24 references indexed in Scilit:
- Mesh segmentation using feature point and core extractionThe Visual Computer, 2005
- D‐Charts: Quasi‐Developable Mesh SegmentationComputer Graphics Forum, 2005
- Polyhedral surface decomposition with applicationsComputers & Graphics, 2002
- Metamorphosis of Polyhedral Surfaces using DecompositionComputer Graphics Forum, 2002
- Geometric Snakes for Triangular MeshesComputer Graphics Forum, 2002
- Partitioning 3D surface meshes using watershed segmentationIEEE Transactions on Visualization and Computer Graphics, 1999
- Reverse engineering of geometric models—an introductionComputer-Aided Design, 1997
- Salience of visual partsCognition, 1997
- 3D part segmentation using simulated electrical charge distributionsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Parts of recognitionCognition, 1984