Posted on Jun 16, 2018
Google’s DeepMind develops AI that can render 3D objects from 2D pictures
861
9
4
5
5
0
Eliminate the need for AI researchers to annotate images in datasets.
Google subsidiary DeepMind today unveiled a new type of computer vision algorithm that can generate 3D models of a scene from 2D snapshots: the Generative Query Network (GQN).
The GQN, details of which were published in Science, can “imagine” and render scenes from any angle without any human supervision or training. Given just a handful of pictures of a scene — a wallpapered room with a colored sphere on the floor, for example — the algorithm can render opposite, unseen sides of objects and generate a 3D view from multiple vantage points, even accounting for things like lighting in shadows. PO1 William "Chip" Nagel Sgt Wayne Wood 1LT (Anonymous) COL Mikel J. Burroughs SMSgt Minister Gerald A. "Doc" Thomas SPC Douglas Bolton Col (Join to see) SGM Matthew Quick SGT (Join to see) SSG Michael O. Capt Christian D. Orr SP5 Mark Kuzinski SGM (Join to see) SGT David A. 'Cowboy' Groth Capt Seid Waddell (Join to see) SPC Margaret Higgins Maj Marty Hogan
Google subsidiary DeepMind today unveiled a new type of computer vision algorithm that can generate 3D models of a scene from 2D snapshots: the Generative Query Network (GQN).
The GQN, details of which were published in Science, can “imagine” and render scenes from any angle without any human supervision or training. Given just a handful of pictures of a scene — a wallpapered room with a colored sphere on the floor, for example — the algorithm can render opposite, unseen sides of objects and generate a 3D view from multiple vantage points, even accounting for things like lighting in shadows. PO1 William "Chip" Nagel Sgt Wayne Wood 1LT (Anonymous) COL Mikel J. Burroughs SMSgt Minister Gerald A. "Doc" Thomas SPC Douglas Bolton Col (Join to see) SGM Matthew Quick SGT (Join to see) SSG Michael O. Capt Christian D. Orr SP5 Mark Kuzinski SGM (Join to see) SGT David A. 'Cowboy' Groth Capt Seid Waddell (Join to see) SPC Margaret Higgins Maj Marty Hogan
Google’s DeepMind develops AI that can render 3D objects from 2D pictures
Posted from venturebeat.com
Artificial Intelligence (AI)
Photography
Technology
Defense
Strategy
Posted >1 y ago
Responses: 3
2
2
0
If you coupled this technology with advanced milling and 3D printing technology this would work toward solving long term problems caused by low production, out of production, or long lead time parts acquisition by using an existing copy to drive the production of a reverse engineered dimensioned drawing for producing these hard to get items.
(2)
(0)
1
1
0
I worry about this thing they call AI. It is only as good as the programming.
(1)
(0)
LTC Jason Mackay
Maj Robert Thornton I understand what you mean, but that is to be expected in the nacient stages. The more they use/develop it, the better it becomes. If this type of thing could help solving the problem of "what's happening now" humans can work on what is happening next.
(1)
(0)
0
0
0
Read This Next
Continue with Facebook 
Continue with Google