As a side project I'm building a computer vision framework and hardware sensor for mobile robots. The target users are people who want to build applications for vision based autonomy, but don't want to go through the hassle of building the full 'stack' themselves.

The sensor is designed to be plug-and-play, so like the Kinect it just works out of the box.

The basic building blocks are in place now, and I'm currently working on C++ and Python APIs, and considering a Swift API.