Fast Radial Search for Progressive Photon Mapping

Information

• Publication Type: Bachelor Thesis
• Workgroup(s)/Project(s):
  • Rendering and Modeling
• Date: October 2022
• Date (Start): 12. February 2022
• Date (End): 1. October 2022
• Matrikelnummer: 11777774
• First Supervisor: Michael Wimmer

Abstract

Global illumination is critical to realistic rendering and as such many different algorithms were developed to solve it, one of which – photon mapping – was designed to efficiently render caustics and indirect lighting. It achieves this by saving photons in a data structure during the first step and then using the result by collecting photons near specific search origins. This step is very time intensive due to the sheer amount of searches performed each iteration. In this paper, we will compare the performance of two spatial data structures and parallelized search algorithms written in CUDA for the GPU by execution time and memory usage for the photon-gathering use case. The algorithms were implemented in an open-source progressive photon mapping project [11] and are using parts of S. Reinwald’s fast-KNN as a basis [9].

Additional Files and Images

Weblinks

No further information available.

BibTeX

@bachelorsthesis{Rait2022,
  title =      "Fast Radial Search for Progressive Photon Mapping",
  author =     "Alexius Rait",
  year =       "2022",
  abstract =   "Global illumination is critical to realistic rendering and
               as such many different algorithms were developed to solve
               it, one of which – photon mapping – was designed to
               efficiently render caustics and indirect lighting. It
               achieves this by saving photons in a data structure during
               the first step and then using the result by collecting
               photons near specific search origins. This step is very time
               intensive due to the sheer amount of searches performed each
               iteration. In this paper, we will compare the performance of
               two spatial data structures and parallelized search
               algorithms written in CUDA for the GPU by execution time and
               memory usage for the photon-gathering use case. The
               algorithms were implemented in an open-source progressive
               photon mapping project [11] and are using parts of S.
               Reinwald’s fast-KNN as a basis [9]. ",
  month =      oct,
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Research Unit of Computer Graphics, Institute of Visual
               Computing and Human-Centered Technology, Faculty of
               Informatics, TU Wien ",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2022/Rait2022/",
}