[Discussion] [computer vision] Instant NeRF create quality depth maps?

Surprised I haven't seen more chatter about this. What do you think about Nvidia's instant Nerf which turns 2d into 3d based on these techniques [https://arxiv.org/abs/2003.10016](https://arxiv.org/abs/2003.10016)

Does the output of a NeRF give a depth map that's comparable to what you'd get from a Kinect?

Can these be used to create 3D models one would use in Unreal or Blender?

computer computer vision instant nerf machinelearning maps nerf quality vision