Lens with mask
Deconvolved Image
Computed Distance Map
Detail of Deconvolved Image
A tri-lens mask inserted in a lens Structurally-deconvolved photograph Automatically computed distance map
Captured (top)  versus structurally deconvolved  (bottom)
Coding Depth through Mask Structure
Horacio E. Fortunato
hefortunato@inf.ufrgs.br

Manuel M. Oliveira
oliveira@inf.ufrgs.br

Instituto de Informática, UFRGS

Logo INFLogo UFRGS

Computer Graphics Forum.
Volume 31 (2012), Number 2, Proceedings of Eurographics 2012, pp. 459-468.   [DOI]


Contents

Abstract Downloads Tri-Lens Mask
Reference Acknowledgments

Abstract

We present a coded-aperture method based on a family of masks obtained as the convolution of one “hole” with a structural component consisting of an arrangement of Dirac delta functions. We call the arrangement of delta functions the structural component of the mask, and use it to efficiently encode scene distance information. We illustrate the potential of our approach by analyzing a family of masks defined by a circular hole component and a structural component consisting of a linear combination of three Dirac deltas. We show that the structural component transitions from well conditioned to ill conditioned as the relative weight of the central peak varies with respect to the lateral ones. For the well-conditioned structural components, deconvolution is efficiently performed by inverse filtering, allowing for fast estimation of scene depth. We demonstrate the effectiveness of our approach by constructing a mask for distance coding and using it to recover pairs of distance maps and structurally-deconvolved images from single photographs. For this application, we obtain significant speedup, and extended range and depth resolution compared to previous techniques.

Downloads

Paper

pdf
Full Paper (pre-print)

The final publication is available at the Computer Graphics Forum Journal

Eurographics Association Copyright Notice: Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than Eurographics must be honored. 

The Tri-Lens Mask


tri-lens



Reference

Citation

Fortunato, Horacio E. and Manuel M. Oliveira. "Coding Depth through Mask Structure". Computer Graphics Forum. Computer Graphics Forum, Volume 31 (2012), Number 2, pp. 459–468.

BibTeX

  @article{Fortunato2012CodingDepth,
  author  = {Horacio E. Fortunato and Manuel M. Oliveira},
  title   = {Coding Depth through Mask Structure},
  journal = {Computer Graphics Forum}
  volume  = {31},
  number  = {2},
  month   = {May},
  year    = {2012},
  pages   = {459-468},
  doi     = {10.1111/j.1467-8659.2012.03025.x}
  note    = {Proceedings of Eurographics}
}

Keywords

Computational Photography, Coded Aperture, Depth-Dependent Deconvolution, Depth Estimation.

Acknowledgments

CNPq  CNPq-Brazil fellowship 308936/2010-8.
REUNI


 Horacio Fortunato was sponsored by a postdoctoral fellowship(Bolsa REUNI, PROPG-UFRGS).

FAPERGS


  FAPERGS provided additional financial support.
Optolentes

  Optolentes (http://www.optolentes.com.br) kindly provided the contact lenses used in our tri-lens prototype.