In the works since end of 2019, it is the first open source attempt at traditional optical motion capture to a) have super fast onboard marker processing and b) have attempted the very complex algorithms that actually make such a system usable for e.g. VR, to my knowledge. It uses a Pi Zeros (1 or 2) QPU for ultra-fast low latency marker detection, which I finished in 2020. Since then I've worked on the Desktop Software, it's many complex algorithms, and System Integration side of things.