deepmind develops AI that imagines 3d scene from 2d images including parts not seen

Think of it like someone handing you a couple of pictures of a room, then asking you to draw what you’d see if you were standing in a specific spot in it. Again, this is simple enough for us, but computers have no natural ability to do it; their sense of sight, if we can call it that, is extremely rudimentary and literal, and of course machines lack imagination.

[link to techcrunch.com (secure)]

“However we found that sufficiently deep networks can learn about perspective, occlusion and lighting, without any human engineering. This was a super surprising finding.”

*****

wow sounds like we're really on the cusp of true human level AI

to me this seems like a step towards AI having an intuitive understanding of the 3d relationships of objects in the world

combine it with an intuitive understanding of physics and behaviors, many important applications, like autonomous driving