International Journal of Computer Vision

, Volume 38, Issue 3, pp 199–218

A Theory of Shape by Space Carving

Authors

  • Kiriakos N. Kutulakos
    • Department of Computer Science and Department of DermatologyUniversity of Rochester
  • Steven M. Seitz
    • The Robotics InstituteCarnegie Mellon University
Article

DOI: 10.1023/A:1008191222954

Cite this article as:
Kutulakos, K.N. & Seitz, S.M. International Journal of Computer Vision (2000) 38: 199. doi:10.1023/A:1008191222954

Abstract

In this paper we consider the problem of computing the 3D shape of an unknown, arbitrarily-shaped scene from multiple photographs taken at known but arbitrarily-distributed viewpoints. By studying the equivalence class of all 3D shapes that reproduce the input photographs, we prove the existence of a special member of this class, the photo hull, that (1) can be computed directly from photographs of the scene, and (2) subsumes all other members of this class. We then give a provably-correct algorithm, called Space Carving, for computing this shape and present experimental results on complex real-world scenes. The approach is designed to (1) capture photorealistic shapes that accurately model scene appearance from a wide range of viewpoints, and (2) account for the complex interactions between occlusion, parallax, shading, and their view-dependent effects on scene-appearance.

scene modelingphotorealistic reconstructionmulti-view stereospace carvingvoxel coloringshape-from-silhouettesvisual hullvolumetric shape representationsmetameric shapes3D photography

Copyright information

© Kluwer Academic Publishers 2000