Dear body trackers!

There are many options for skeleton tracking you can use with vvvv:

• Kinect (the original one)
• Kinect 2
• Azure Kinect
• Femto Bolt & Mega
• ZED
• Nuitrack
• Ultraleap

But those options all require the according specific depth camera, which you might not always at hand. Now here is a new option that doesn’t require additional hardware and tracks a single 3d skeleton (including fingers and face!) even from a simple webcam:

Introducing: Dollars MONO by Dollars MoCap

Watch this demo of their technology to get an idea of the quality of tracking you can expect:

When running Dollars MONO on the same PC as vvvv, all you have to do is activate the “vvvv Streaming” option you can find under Settings > Stream:

In vvvv you need the VL.DollarsMoCap NuGet that comes with a node called DollarsMoCap that receives the data.

To control an avatar with that data, all you need to do is load a rigged model and provide a joint mapping, ie. a spread of joint names the model comes with, that aligns with the joint names as returned by DollarsMoCap. This takes about 5 minutes to set up and then works for all models that use the same skeleton.

Many thanks go out to Kan Feng, the creator of Dollars MoCap, for the great support and helping us along the way of implementing this for vvvv!

How do i get it?

As of this writing, requires latest vvvv gamma 7.0 preview builds. Then install as usual:

nuget install VL.DollarsMoCap

If you have a request for custom development, please get in touch!