TrakMark - Panel14

Last Update : October 6, 2014

TrakMark Panel Session in ISMAR 2014 (Wednesday 10 Sep, 2014)

TrakMark: How to Benchmark AR/MR Geometric Registration and Tracking Methods

Co-chairs: Hirokazu Kato, Nara Institute of Science and Technology
                Fumihisa Shibata, Ritsumeikan University
Panelists: Nassir Navab, TU Munich
               Jürgen Sturm, Metaio
               Takeshi Kurata, AIST
               Koji Makita, AIST
               Fumihisha Shibata, Ritsumeikan University
Description: In the research fields of Augmented Reality (AR) and Mixed Reality (MR), tracking and geometric registration methods are still one of the most important topics. The tracking research field is highly active, and every year, “Tracking” session is scheduled in the ISMAR. The “TrakMark” working group (WG) was established in 2009, as a lower branch of the Special Interest Group on Mixed Reality (SIG-MR), the Virtual Reality Society of Japan (VRSJ), to create the benchmark of AR/MR geometric registration and tracking methods, named “TrakMark”, that permits objective evaluation and comparison of diverse registration and tracking methods.

In 2009, we held a special session in ISMAR 2009 to launch the TrakMark activity worldwide, where we provided a kind of a policy speech. After that, we held the following four workshops.

  • The 1st International Workshop (October 12, 2010, in Seoul, closed style)style)
  • The 2nd International Workshop (October 26, 2011, in association with ISMAR 2011)
  • The 3rd International Workshop (November 11, 2012, in association with ICPR 2012)
  • The 4th International Workshop (Octorber 1, 2013, Joint Workshop on Tracking Methods & Applications and TrakMark, in association with ISMAR 2013)

  • Based on the discussion in these workshops, we have created the benchmark test scheme, which is the release candidate version.

    In the first half of this session, we will make a presentation of the report as a compilation of these works. This report will include not only the introduction of Trakmark and the data sets we published but also the indicators of the benchmark and the trial results of the benchmark. Furthermore, we will introduce the uniform software environment, named “Casper Cartridge” for benchmarking, and the benchmark standardization activity in ISO/IEC JTC 1/SC 24/WG 9.

    In the second half of this session, we will run a panel discussion about the current issue and the future of TrakMark.


    [The first half: introduction by presentation, about 50-60 minutes]

    [The second half: panel discussion, about 30-40 minutes]

    Panelists: Nassir Navab, Jurgen Sturm, Koji Makita, Takeshi Kurata, Fumihisa Shibata