Tracking Failure Detection and Recovery

Here are few words on a new feature we’ve added these days to our main development trunk: Tracking failure detection and recovery. Using that feature the system is capable of detecting various tracking failures and recover from that. Tracking failures occur, for example, when the user points the sensor in a direction that is not being covered by the volume currently reconstructed, or when the sensor is accelerated too fast. Once a tracking error is detected, the system switched to a safety position and requires the user to roughly align viewpoints. It automatically continues from the safety position when the viewpoints are aligned.

Here’s a short video demonstrating the feature at work.

We are considering to integrate this feature into the final beta phase, since stability of the system and its usability increase. Be warned, however there are still cases when the systems fails to track and fails to detect that the tracking was lost, causing the reconstruction to become messy.

9 thoughts on “Tracking Failure Detection and Recovery

  1. Pingback: Did you know … ? | ReconstructMe

  2. Pingback: Did you know … ? | ReconstructMe

  3. mark

    I’m hoping you’ve looked at ptam to see how SLAM can help you with that…
    or for more recent and uptodate approach these links are excellent.
    The links branching out from this page are very relevant
    and CSAIL can’t be underestimated…

    1. Christoph Heindl Post author

      Thanks for the links. I knew the first and the second one. The approach we’ve taken fits nicely into our processing pipeline and comes nearly at zero overhead cost in runtime and implementation. I hope we get the algorithm submitted to a journal by summer (publication is 2013 then).


  4. Pingback: Did you know … ? | ReConstructMe

  5. BRIAN

    I have latest version and this does not work .

    everytime i start to move the kinect 360 sensor .
    i get

    tracking lost move to recovery position (new scan) (finish)

    no matter where i move the sensor to the screen remains frozen and same warning .

    does not work like in video

Leave a Reply