A simple, fast C++ library for getting the complete pose information from an Oculus Rift, with Python bindings. Meant to be an easy starting place for Rift-based projects that don't require the rendering functionality of the OculusSDK. Used in the OcuMol-Leap project, which adds VR support to PyMol. Tested on OSX and Windows, should work on Linux too.

• CMake
• Cython (for the Python bindings)
• OculusSDK (the official one)
• Oculus Runtime (make sure your runtime and SDK versions match)
• On Windows, you will also need
  • Python (doesn't come preinstalled like on OSX)
  • Visual Studio (for the compilers)

If are on OSX, I strongly recommend using Homebrew to install CMake and Python, and then using pip to install Cython

• Get the official OculusSDK and put it in a convenient directory
• cd to the ocudump root directory
• run OCULUS_SDK_ROOT_DIR=<path-to-your-oculusSDK> pip install -e .
  • Be sure to replace <path-to-your-oculusSDK> with the actual path to your copy of the Oculus SDK.
  • If you are running with the sudo command, you can pass environment variables by -E (at least on OS X), so you would run: OCULUS_SDK_ROOT_DIR=<path-to-your-oculusSDK> sudo -E pip install -e .
  • The above command will get pip to install ocudump in development mode, meaning that it will create a kind of soft link between your python module directory and the ocudump directory.
  • Eventually ocudump will also be available directly through pypi.

• Run the pip command as described in Build and Install
• cd to the newly-created build directory within your ocudump root dir
• run ./dumpTest

• Run the pip command as described in Build and Install
• cd to the newly-created build directory within your ocudump root dir
• run python dumpTest.py

• Follow the Build and Install instructions
• Import the Ocudump class from the ocudump module:
  • from ocudump import Ocudump
• Create an Ocudump instance (this will also initialize your Oculus Rift and it's tracking camera, so be sure to have them both plugged in at this point):
  • o = Ocudump()
• Update the pose information contained in your Ocudump instance:
  • o.getPose()
• If your tracking camera was tracking position at the time when you got the last pose, Ocudump member variable positionTracked will be set to true. Otherwise, it will be set to False:
  • print(o.positionTracked)
• After you run getPose(), the pose Ocudump member variable will contain a list with six floating point numbers. The first three numbers are the rotation angles of the Oculus Rift pose, and the last 3 are the xyz position:
  • o.pose -> [yaw, pitch, roll, x, y, z]
  • In the case where the tracking camera is not currently tracking your Rift, the last three entries, where xyz normally go, will instead just contain nan values.