R-Score: A Novel Approach to Compare Monte Carlo Renderings

Information

• Publication Type: Technical Report
• Workgroup(s)/Project(s):
  • A Test Suite for Photorealistic Rendering and Filtering
• Date: August 2020
• Number: TR-193-02-2020-4
• Open Access: no

Abstract

In this paper, we propose a new approach for the comparison and analysis of Monte Carlo (MC) rendering algorithms. It is based on a novel similarity measure called render score (RS) that is specically designed for MC rendering, statistically motivated, and incorporates bias and variance. Additionally, we propose a comparison scheme that alleviates the need for practically converged reference images (RIs). Our approach can be used to compare and analyze dierent rendering methods by revealing detailed (per-pixel) dierences and subsequently potential conceptual or implementation-related issues, thereby offering a more informative and meaningful alternative to commonly used metrics.

Additional Files and Images

Weblinks

No further information available.

BibTeX

@techreport{freude_2020_rs,
  title =      "R-Score: A Novel Approach to Compare Monte Carlo Renderings",
  author =     "Christian Freude and Hiroyuki Sakai and Karoly
               Zsolnai-Feh\'{e}r and Michael Wimmer",
  year =       "2020",
  abstract =   "In this paper, we propose a new approach for the comparison
               and analysis of Monte Carlo (MC) rendering algorithms. It is
               based on a novel similarity measure called render score (RS)
               that is specically designed for MC rendering, statistically
               motivated, and incorporates bias and variance. Additionally,
               we propose a comparison scheme that alleviates the need for
               practically converged reference images (RIs). Our approach
               can be used to compare and analyze dierent rendering methods
               by revealing detailed (per-pixel) dierences and subsequently
               potential conceptual or implementation-related issues,
               thereby offering a more informative and meaningful
               alternative to commonly used metrics.",
  month =      aug,
  number =     "TR-193-02-2020-4",
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  institution = "Research Unit of Computer Graphics, Institute of Visual
               Computing and Human-Centered Technology, Faculty of
               Informatics, TU Wien ",
  note =       "human contact: technical-report@cg.tuwien.ac.at",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2020/freude_2020_rs/",
}