Train an AI to infer accurate mathematical calculations by simply “looking” at images of shapes/objects


I’d like to build a model that has an understanding of geometry, where it can be applied to question and answering system. Specifically, it would be nice if it could determine the volume of an object by simply looking at pictures of it.

If there are any pre-trained models out there that I can utilize that would certainly make things easier.

Otherwise, are there any suggestions on the kind of model(s) I should use to do this?

Also, I read something online about how Facebook trained an AI to solve complex math problems just by looking at them. They approached the problem as a language translation problem, not as a math problem. I wonder if this is the way to go?


Posted 2020-03-03T15:15:14.750

Reputation: 231

No answers