Right now, the code is developed to use one camera for pose estimation. What if we wanted to add a second, a third, or 10? 10 is a lot, but 3 isn't. We should develop the vision subsystem to take in multiple camera instances, fuse that data together in some way to get pose estimations
Right now, the code is developed to use one camera for pose estimation. What if we wanted to add a second, a third, or 10? 10 is a lot, but 3 isn't. We should develop the vision subsystem to take in multiple camera instances, fuse that data together in some way to get pose estimations