Making AI imagine stuff: Research on generating Image from Text



This image completely explains the work done by the authors in this paper:

I find the paper particularly interesting, because it can give way to information retrieval we have never seen before. Just consider this scenario; there are times when you remember the description of the object but when explaining this to the other person, a visual image is more interpretable. So what you can do is ask Siri to give you images of objects you describe and show it to the other fellow. As simple as that!

Definitely an interesting read. If you do please share your thoughts here!

PS: This is a continuation of My Random Nonsense!. Do check out the other ones