Skip to main content Link Search Menu Expand Document (external link)
Hidale.com Docs

@skelcompute attribute

🔌dp.kinect3 dp.oak
signature
skelcompute ENGINE_STRING [ DEVICE_SUBSTRING | DEVICE_INDEX ]
values
directml 0 default
examples
@skelcompute directml        <- DirectML, first GPU
@skelcompute directml 1      <- DirectML, second GPU
@skelcompute directml "2070" <- DirectML, "2070" GPU
@skelcompute directml intel  <- DirectML, "intel" GPU
@skelcompute directml rtx    <- DirectML, "rtx" GPU
@skelcompute cuda            <- CUDA, first Nvidia GPU
@skelcompute tensor          <- TensorRT, first Nvidia GPU
@skelcompute tensor_fp16     <- TensorRT, Nvidia GPU, float16

Skeleton tracking compute engine and device. Enable multiple compute devices for different tasks using attributes like @skelcompute, @transcoder, and @opencl. For example…

  • Track skeleton joints on the discrete Nvidia GPU @skelcompute nvidia
  • Decode color frames on the Intel CPU harware decoder @transcoder intelmedia
  • Flip and undistort frames on integrated Intel GPU @opencl intel
  • and the remaining features run on your CPU

Test to discover which settings meet your needs for hardware, latency, and throughput. You can have significant performance improvements! 🙂

DirectML is built into Windows and requires no downloads. Other compute engines like CUDA and TensorRT have additional download requirements. Consult your plugin setup documentation for details.

📝 The second parameter of @skelcompute is the name of the device or the numeric index (starting with 0) of the device. Name of the device is recommended. The index may change because the Windows Graphics Performance Preference default is “Let Windows decide”. Windows may decide to change the order of GPUs and therefore the indices change. The index is consistent when you choose a Graphics Performance Preference. This setting is in Windows settings, System, Display, Graphics settings.

Performance

  • DirectML works very well across most GPUs. It is the recommended choice.
  • CUDA and TensorRT are optional providers for body tracking. They require additional setup, are for unusual situations, and performed no better than DirectML during testing.
  • TensorRT usually performs better than CUDA.
  • @skelcompute tensor_fp16 will use the Tensor Runtime and optimize the chosen body tracking model with 16-bit floating point calculations. It is slightly less accurate and uses less resources.
  • TensorRT choices will have a very long first-time startup. For example, it is almost 5 minutes on a laptop with a RTX2070-Super GPU and choosing tensor_fp16. Later startups are only a few seconds due to caching.
  • TensorRT caches the first-time startup optimizations at %TEMP%\PLUGIN_NAME\tensor.cache. You or Windows can delete these cache folders. TensorRT will re-create a cache on the next startup.
  • Each TensorRT cache folder is dependent on your DLLs, body tracking model, compute device, fp16 optimization, etc. When you change configuration, a new cache folder will be created and new optimizations cached. These cache folders can grow to be hundreds of megabytes.

Known Issues

  • Microsoft’s Azure Kinect bug https://github.com/microsoft/Azure-Kinect-Sensor-SDK/issues/1783 sometimes crashes Max when closing the application. It involves some combinations of compute choices. The crash may not be apparent; it happens as Max is closing. It may be no real-world harm.
  • Azure Kinect bug may freeze or crash with the below scenario. It may be safer to change @skelcompute by typing that attribute on the (dp.kinect3 @skelcompute directml ...) box itself.
    1. Output visual data like depth, color, ir, or playermap
    2. Stop dp.kinect3
    3. Change @skelcompute using a message to dp.kinect3 or an (attrui)
    4. Start dp.kinect3, the crash/freeze usually happens within a few seconds.