I'm surprised that binocular cameras aren't more of a thing for robots. It seems technically simple and cheap, and I feel like it should be very amenable to deep learning image approaches (using a six-layer RGB+RGB image input, overlaying the two images on each other).
The only things I've found are niche and expensive bits of hardware.
I guess it's easy enough to hack together, and hope there's no latency issues between two cameras, but this RPI has a limited number of ports...
This Mastodon instance is for people interested in technology. Discussions aren't limited to technology, because tech folks shouldn't be limited to technology either!