Computational Photography with Panoramic Sensors that Have Uniform Resolution with Respect to Unwarping Transformations

  • Authors:
  • R. Andrew Hicks;Meredith Coletta

  • Affiliations:
  • Department of Mathematics, Drexel University, Philadelphia, USA 19104;Gateway Systems, Boyertown, USA 19512

  • Venue:
  • Journal of Mathematical Imaging and Vision
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

We address the problem of designing a computational photographic system for panoramic imaging, which employs a rotationally symmetric reflector and yet produces uniform resolution images. Our design is a catadiotpric sensor, whose key component is a rotationally symmetric mirror. Catadioptric sensors have already been designed that are equiresolution with respect to solid angle, but such sensors do not take into account the effect of unwarping transformations that are typically applied to the initially recorded images. The family of catadioptric sensors that we describe may be considered as realizing area preserving/scaling cylindrical projections. We present the derivations of the equations that describe such sensors and demonstrate in simulation their properties with respect to unwarping.