TT_TrackableLocation error

HPx
Posts: 3
Joined: Tue May 03, 2011 3:55 am

TT_TrackableLocation error

Post by HPx » Tue May 03, 2011 4:13 am

Hello,

I'm using the dll NPTrackingTools.dll to track a rigid body.
It works well for a time and then, TT_IsTrackableTracked is still
working but TT_TrackableLocation returns undefined numbers until the application is restarted.

I also have the problem in Tracking Tool which sometimes lost my Trackable until restarted.

Any Ideas ?

beckdo
NaturalPoint Employee
NaturalPoint Employee
Posts: 520
Joined: Tue Jan 02, 2007 2:02 pm

Re: TT_TrackableLocation error

Post by beckdo » Tue May 03, 2011 10:43 am

Can you give us some more information on this issue? ie. How long till you have problems? What's your rigid body look like, how many markers, and specific settings besides the default? How many cameras, is the 2D data still updating? What devices are you using? And maybe some suggestions about how we can reproduce it?

HPx
Posts: 3
Joined: Tue May 03, 2011 3:55 am

Re: TT_TrackableLocation error

Post by HPx » Wed May 04, 2011 12:42 am

Thanks for answering me

My configuration is:
- 3 cams V100R2 + Optihub
- 3D Glasses with 5 reflectors
- A second tool with 4 reflectors

The problem happen when you have many interferences, from an other tool or the environment. If the world is not stressful, I don't think it's happening. But if your tools are very closed, it could happen in severals seconds.
It seams that it mostly occurs when the trackable enter or leave the frustum.
Cameras still update because, they are still tracking my second tool, when I have one.

Harold

beckdo
NaturalPoint Employee
NaturalPoint Employee
Posts: 520
Joined: Tue Jan 02, 2007 2:02 pm

Re: TT_TrackableLocation error

Post by beckdo » Wed May 04, 2011 11:15 am

Hey Harold,

You're describing a very difficult tracking scenario. Especially if your markers are on or within the frames of the glasses. To make things worse you have reflections from the ringlights typically within every camera's view making tracking challenging.

If possible, please record a timeline (.TIM) and send over the timeline and the calibration (.CAL) file so I can take a look at your scenario.

Thanks

HPx
Posts: 3
Joined: Tue May 03, 2011 3:55 am

Re: TT_TrackableLocation error

Post by HPx » Thu May 05, 2011 12:33 am

Hi Doug,

I've recreated my trackables and the camera positions and it works better. I'll try to record my scenario if it happens again.
Do you have advices to track 3D Glasses ?
Is it better to use active markers ? with which skeleton ?

Thanks.

bencrossman
Posts: 36
Joined: Wed Jun 06, 2012 10:02 pm

Re: TT_TrackableLocation error

Post by bencrossman » Mon Aug 27, 2012 8:23 pm

I too would like to know the best trackable markers arrangement for a pair of 3D Glasses.

Thanks

NaturalPoint - Mike
Posts: 1896
Joined: Tue Feb 01, 2011 8:41 am
Location: Corvallis, OR

Re: TT_TrackableLocation error

Post by NaturalPoint - Mike » Tue Aug 28, 2012 3:06 pm

It does depend heavily on the glasses as well as the synchronization system.

Active markers with no strobes on the cameras would be ideal, otherwise you would need to synchronize the camera exposure with the burst of the IR system in the 3d glasses.

Marker arrangement would ideally be 4 or 5 markers on the glasses around the frame where they can easily be seen by the cameras without occlusion.

bencrossman
Posts: 36
Joined: Wed Jun 06, 2012 10:02 pm

Re: TT_TrackableLocation error

Post by bencrossman » Tue Aug 28, 2012 4:50 pm

We are using DLP glasses so there is no IR problem.

Ok, will try 4 or 5 markers.

Thanks

bencrossman
Posts: 36
Joined: Wed Jun 06, 2012 10:02 pm

Re: TT_TrackableLocation error

Post by bencrossman » Mon Oct 22, 2012 8:32 pm

We are still having trouble getting a working configuration.
I tried this and got bad results:

Image

This is what we are trying to achieve:
- A room 4.8m x 2.4m
- We want a wearer of the glasses to be 6 DOF trackable in the whole room
- The user will typically be facing one of the 4.8m walls
- We have four wide cameras currently positioned in each corner but this is probably wrong. Would it be better for all 4 cameras to be along the facing wall?

Thanks
Ben

Seth Steiling
Posts: 1366
Joined: Fri Jun 27, 2008 11:29 am
Location: Corvallis, Oregon

Re: TT_TrackableLocation error

Post by Seth Steiling » Wed Oct 24, 2012 9:53 am

If the wearer is always going to be looking at the same wall, I would recommend mounting all cameras on that wall--perhaps along the top of the projector screen/display/etc. I would also move the markers to the front of the glasses, so they aren't separated by the wearer's head.

What cameras are you using? And what lenses?
Marketing Manager
TrackIR | OptiTrack

Post Reply