Hardware settings

Hello, thank you for sharing such a great results and codes.
I`m trying to apply your algorithm for grasping (without suction) the occluded objects in a box and I have a question about camera setting after reading the paper. Is it necessary to put 2 cameras for height-map generation? Would it be also possible with 1 camera? Thanks in advance.