Hello!
I’ve developed some novel algorithms for deployment in projector-camera systems (with varying degrees of maturity.) I am hoping to select a commercially available reference hardware to make it easier for others to replicate and improve upon my work (I built my own prototypes). I bought a Range in the hopes that I could SSH into it, as I hear is possible for the POP2. This is important, as I need to craft the projector output myself. I didn’t have any luck with SSH, but have been granted SDK access.
All the documentation I’ve seen so far focuses on receiving the analysed data, and camera feeds, but I haven’t seen any way to write to the projector directly.
I was wondering whether such a method exists (apologies if it’s obvious; I’m pursuing several projects simultaneously).
There are some really interesting applications being pursued, and the Range seems like it would be very suitable if I can use it (with the MIRACO being perfect if I could access its projector!), and I’m confident it would be very popular among researchers.
I am not an expert, but from my experience with the Revopoint scanners and some of the older Einscan/Shining3d equipment, the Revopoint uses a mechanical diffraction as apposed to a programable DLP/LCD projector.
I think you would be better served with a DLP projector and some cameras to test/develop your scanner. I think this is how David 3D Scanner started out, until it was acquired by HP (and effectively shut down). I think some of the older Shining 3d scanners had programmable projectors but I am pretty sure that this has all be locked down by now.
Once you have the “perfect” pattern, you can then verify that it is not copywrite by companies like PrimeSense (Apple), Intel (RealSense) etc.
Good luck and if you have a link to your project on Github, I would love to check it out.
Bruce
Thank you very much, Bruce, for such an informative and quick answer!
The 3D scanning application happens to be the weakest part of my project, but that method generates new patterns on the fly, so I’m not worried about copywrite!
The application for which I am seeking a demonstration platform is a cryptographic feedback loop in which a hash is generated from a camera image, projected onto the scene, a hash generated from the subsequent camera image, and projected etc. There are more details at https://poliebotics.com in a goofy demo video.
From infrared videos of scanner projections on YoutTube, it looks like there are some degrees of freedom in what is projected (could be a recordings artifact). If I can cram 256 bits of entropy into a projection, that would be good enough for me!