When is the Shape of a Scene Unique Given its Light-Field: A Fundamental Theorem of 3D Vision? - Robotics Institute Carnegie Mellon University

When is the Shape of a Scene Unique Given its Light-Field: A Fundamental Theorem of 3D Vision?

Simon Baker, Terence Sim, and Takeo Kanade
Journal Article, IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol. 25, No. 1, pp. 100 - 109, 2003

Abstract

The complete set of measurements that could ever be used by a passive 3D vision algorithm is the plenoptic function or light-field. We give a concise characterization of when the light-field of a Lambertian scene uniquely determines its shape and, conversely, when the shape is inherently ambiguous. In particular, we show that stereo computed from the light-field is ambiguous if and only if the scene is radiating light of a constant intensity (and color, etc) over an extended region.

BibTeX

@article{Baker-2003-8609,
author = {Simon Baker and Terence Sim and Takeo Kanade},
title = {When is the Shape of a Scene Unique Given its Light-Field: A Fundamental Theorem of 3D Vision?},
journal = {IEEE Transaction on Pattern Analysis and Machine Intelligence},
year = {2003},
month = {January},
volume = {25},
number = {1},
pages = {100 - 109},
keywords = {3D shape reconstruction, stereo, shape-from-silhouette, the plenoptic function, light-fields, uniqueness},
}