I was searching around today for Full Body Tracking solutions and I came across this website. There, the writer talks about a company called “Extreme Reality” which has a free SDK that supposedly turns your average webcam into something similar to a Kinect with “the power to capture a three-dimensional range of movement”.
At first, I was considering undertaking the project of using the SDK to create Full Body Tracking for VR but I’m a game developer – not a software engineer. So, I considered dropping the idea on Reddit or Discord but then I thought why not throw the idea to the creator of Driver4VR, whom, would already have experience in this field.
The idea is that, a single webcam is inferior to a Kinect. Even with my local prices of $80ea a Kinect 2.0 would be far better. However, according to this post multiple Kinects are not a reasonable solution. (At the very least you would need multiple USB 3.0 cards, one for each Kinect – as well as a beast PC to boot) However Webcams are different. They have far less overhead, and your average system could run several of them without too much trouble.
I would love to hear your feedback on this topic. Specifically why it is or isn’t reasonable, or even possible.