Seems Apple solved monocular SLAM problem before attempting VR.
Wonder if they used FPGA or anything to speed up the algorithm? In my experience SLAM takes too much computation power and/or release too much heat.
Wonder if they used FPGA or anything to speed up the algorithm? In my experience SLAM takes too much computation power and/or release too much heat.
Detailed monocular SLAM needs a lot of CPU, but that is changing...