YouTube:
- HTC Vive: https://youtu.be/3XEQG9NMxoM
- PlayStation VR: https://youtu.be/SxlgeaNkngw
CAVR is a general, future-proof design that exports multiple types of context information that are sent over anetwork interface. This design works for basically all VR-HMDs, with the help of community-supported tools.
Details around this project can be found in: CAVR
The theoretical design and used software to implement this design:
First clone this repository:
git clone https://github.com/Charles-Derek/Context-Aware-VR.git
To build the Context-Extraction system, navigate to the third-party directory under vr_tracking:
cd Context-Aware-VR/vr_tracking/third-party
Make the three main install scripts executable:
chmod +x packages.sh cppzmq.sh mqtt.sh
First we install all necessary packages, check if any errors occur!
./packages.sh
Next, we install the ZMQ library, check if any errors occur!
./cppzmq.sh
Finally, we install the MQTT library, check if any errors occur!
./mqtt.sh
All necessary software should be installed and we can test the Context-Extraction system. To build it, use the CMakeLists.txt under Context-Aware-VR/vr_tracking/CMakeLists.txt
To test the system, don't forget to first install Steam & SteamVR, once SteamVR detects the HMD correctly the system should work.
The subscriber that shows the visualization is located in vr_subscriber called subscriber.py. For MQTT you need to run an additional instance broker.py
Future implimentations include a web interface for the subscriber to run on a remote machine using Docker with usb passthrough to the host machine and motion capture system.