Punctuated simplification of man-made objects

  • Authors:

  • Justin Jang;Peter Wonka;William Ribarsky;Christopher D. Shaw

  • Affiliations:

  • GVU Center and College of Computing, Georgia Institute of Technology, 30332, Atlanta, GA, USA;PRISM and Dept. of Computer Science and Engineering, Arizona State University, 85287, Tempe, AZ, USA;CS Dept. and Charlotte Visualization Center, UNC Charlotte, 28223, Charlotte, NC, USA;School of Interactive Arts and Technology, Simon Fraser University Surrey, 14-660 Central City Tower, 13450 102 Ave, 28223, Surrey, BC, Canada

  • Venue:
  • The Visual Computer: International Journal of Computer Graphics
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a simplification algorithm for manifold polygonal meshes of plane-dominant models. Models of this type are likely to appear in man-made environments. While traditional simplification algorithms focus on generality and smooth meshes, the approach presented here considers a specific class of man-made models. By detecting and classifying edge loops on the mesh and providing a guided series of binary mesh partitions, our approach generates a series of simplified models, each of which better respects the semantics of these kinds of models than conventional approaches do. A guiding principle is to eliminate simplifications that do not make sense in constructed environments. This, coupled with the concept of “punctuated simplification”, leads to an approach that is both efficient and delivers high visual quality. Comparative results are given.