Improved Rigid Body Tracking

Post Reply
beiyang
Posts: 2
Joined: Tue Mar 04, 2008 10:48 am

Improved Rigid Body Tracking

Post by beiyang » Tue Mar 04, 2008 10:57 am

We've got our OptiTrack 6 camera rigid body tracking working just fine. However I would like to know the time line for software development. Are there any planed improvements to the Rigid Body tracking?

Currently, if I define a rigid body with 4 markers. All 4 markers must be visible for the rigid body to be tracked. A better behavior is if one marker is covered it defaults the 3 remaining markers. So with more markers on a rigid body the more stable the tracking would be. However, with the current implementation the reverse is true.

We've hacked something together where instead of defining a rigid body with all 4 trackers, we've defined four 3-marker rigid bodies using the four possible permutations.

This is currently a hack and works only to an extent and gets complicated when we have 5 or 6 trackers on a rigid body. Is there any plans to get something like this implemented?

VincentG
Posts: 7728
Joined: Mon Jul 17, 2006 5:00 am
Location: Corvallis, Oregon

Re: Improved Rigid Body Tracking

Post by VincentG » Tue Mar 04, 2008 1:49 pm

What version of the software are you using?

Jim
NaturalPoint Employee
NaturalPoint Employee
Posts: 1395
Joined: Mon Oct 14, 2002 5:00 am
Location: Corvallis, Oregon
Contact:

Re: Improved Rigid Body Tracking

Post by Jim » Tue Mar 04, 2008 2:08 pm

Hello:

I understand the behavior you are describing. The latest software, build 34, should be robust enough to handle 4 markers and lose one and still track the rigid body. That is how it should work. The rigid bodies can be somewhat specific to the marker configuration, and some configurations work better than others for sure. You can upload your rigid body and calibration file or data file and we will take a look at it.

We appriciate comments on the functionality and feature requests, that is how we will make the sofware better. We will work to get the features in, but don't know the exact time frame, although we will do our best to give an estimate.

Finally, we have seen users combine the output of several 3 or 4 marker rigid bodies and average the data from them, so they can get really redundant tracking. So, on a headset, you could have 2 sets of 3 markers, each defined as its own RB, then average the data output between the sets. Just an idea.

Post Reply