HDR Image Acquisition for Augmented Reality

Information

• Publication Type: Bachelor Thesis
• Workgroup(s)/Project(s):
  • Reciprocal Shading for Mixed Reality
• Date: September 2013
• Date (Start): 1. March 2012
• Date (End): 16. September 2013
• Matrikelnummer: 0925447
• First Supervisor:
  • Michael Wimmer
  • Martin Knecht
• Keywords: mixed reality, high dynamic range image

Abstract

In this thesis I present a method for calculating high dynamic range images in a mixed reality system. Cameras and monitors usually have a lower dynamic range than we encounter in the real world, e.g. the sun. While pictures have a maximal contrast of 1:500, real world scenes often have a contrast of 1:100 000. An image taken of a scene with a higher dynamic range than our camera can capture will have regions that are too bright or too dark. With a higher exposure time more details will be visible in dark regions and with a lower exposure time more details will be visible in bright regions. Since our camera cannot create an image preserving details in both dark and bright regions we have to calculate one using the images our camera can actually produce. The method described in this thesis is based on the work of Debevec and Malik. It takes several images taken with different exposure times and combines them to a high dynamic range image, leading to a better viewing experience in our RESHADE framework, a mixed reality framework, for which this method was implemented.

Additional Files and Images

Weblinks

No further information available.

BibTeX

@bachelorsthesis{rasch_martina-2013-HDRImage,
  title =      "HDR Image Acquisition for Augmented Reality",
  author =     "Martina Rasch",
  year =       "2013",
  abstract =   "In this thesis I present a method for calculating high
               dynamic range images in a mixed reality system. Cameras and
               monitors usually have a lower dynamic range than we
               encounter in the real world, e.g. the sun. While pictures
               have a maximal contrast of 1:500, real world scenes often
               have a contrast of 1:100 000. An image taken of a scene with
               a higher dynamic range than our camera can capture will have
               regions that are too bright or too dark. With a higher
               exposure time more details will be visible in dark regions
               and with a lower exposure time more details will be visible
               in bright regions. Since our camera cannot create an image
               preserving details in both dark and bright regions we have
               to calculate one using the images our camera can actually
               produce. The method described in this thesis is based on the
               work of Debevec and Malik. It takes several images taken
               with different exposure times and combines them to a high
               dynamic range image, leading to a better viewing experience
               in our RESHADE framework, a mixed reality framework, for
               which this method was implemented.",
  month =      sep,
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Institute of Computer Graphics and Algorithms, Vienna
               University of Technology ",
  keywords =   "mixed reality, high dynamic range image",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2013/rasch_martina-2013-HDRImage/",
}