forums.naturalpoint.com

Discussion and Support for the OptiTrack, SmartNav and TrackIR brands by NaturalPoint

Capturing rotate data with camera position constrained.

by Marcoma » Mon Nov 30, 2015 5:55 pm

Hey Guys - new here so bear with me.

I wanted to ask how was the best way create a large camera move in maya or motionbuilder then constrain the VCS to it and ONLY capture rotation information?

Thanks in advance.
Marcoma
 
Posts: 7
Joined: Fri Oct 16, 2015 3:25 pm

by steven.andrews » Thu Dec 03, 2015 11:27 am

Hello Marcoma,

The VCS Device's property list contains properties for Translation and Rotation. You can disable the Translation property so that only the rotation will drive the camera.

After disabling the Translation property, you can use a constraint to attach the camera to an animated node. This would provide the large camera move while leaving rotation in the hands of the person carrying the VCS.

I hope you find this suggestion to be useful. If we can do anything else to help, please feel free to open a support ticket with us at help.naturalpoint.com

Best regards,
Steven
--
Steven Andrews
OptiTrack | Customer Support Engineer
steven.andrews
NaturalPoint Employee
NaturalPoint Employee
 
Posts: 381
Joined: Mon Jan 19, 2015 11:52 am

by Marcoma » Fri Dec 04, 2015 2:54 pm

Perfect. Thanks (again) Steven.
Marcoma
 
Posts: 7
Joined: Fri Oct 16, 2015 3:25 pm

by Marcoma » Mon Dec 07, 2015 10:14 am

Ok - I spoke with Steven and we were able to constrain the camera's root translations to a moving node - however, we have had to settle on a less than eloquent solution to align the camera with the root.

Is there a way to create the camera in Mobu without it establishing the real-world offset between the root and camera position (in Motive)? i.e. Create a camera node and root at world zero co-ordinates?
Marcoma
 
Posts: 7
Joined: Fri Oct 16, 2015 3:25 pm

by steven.andrews » Mon Dec 07, 2015 1:16 pm

Hi Marco,

Have you tried disabling the Translation property on the device before going online?
I'm not sure if this will work, but it's worth a try.

I also see an OffsetTranslation property and a Camera 1 Position property. If nothing else is working, maybe you can get the current position of the camera and negate this in the OffsetTranslation. This could even be automated via Python.

Let me know if any of this helps. If we need, I will check with development to see if they have any tricks to suggest.

Cheers,
Steven
--
Steven Andrews
OptiTrack | Customer Support Engineer
steven.andrews
NaturalPoint Employee
NaturalPoint Employee
 
Posts: 381
Joined: Mon Jan 19, 2015 11:52 am

by danwarom » Sat Jan 23, 2016 9:47 am

Much like the suggestions here, this is my workflow for nodal VCS capture.

1) Create the move you want - I usually have this as a separate camera in Mobu so we have it for Before & After for the client. Bake it and save it.
2) Import the Optitrack VCS device, create the bind.
3) Open the Devices properties, untick all translation XYZ boxes.
4) Zero out any Tranform Offset Channels. Hell you can set the translation scale to 0,0,0 if you like but this wont have any effect.
5) Parent the VCS:Root to the previously moving camera.
6) Set the Optitrack Device to go online in Mobu.
7) Set your Orientation in Motive to the position of the camera man's most comfortable pose. For nodals, I usually have a stool on hand to make it easier.

You're now ready to go!

Cheers!
danwarom
 
Posts: 7
Joined: Wed Feb 13, 2013 7:08 pm
Location: Los Angeles


Return to Insight VCS