It seems that fantasy worlds are everywhere, and are now making their way into the artificial intelligence space via AI painting technology. Take the instance of two popular British novel series. The Chronicles of Narnia made C.S Lewis known as a literary genius and the Harry Potter series turned J.K Rowling into a billionaire. The website Pottermore publishes weekly articles on the major characters sketches, discusses themes, and occasionally gives advice on how one can imitate Dumbledore’s humility and wit. The world is also spinning under the spell of Game Of Thrones, another fantasy adventure set in the imaginary land of Westeros. Regardless of the world our minds can conceive, it seems there’s always someone there to make it. However, these days, computers are able to do both the conception and the building.
All three series were popular novels before being adopted for screenplay. The leap from imagining new worlds in one’s mind to encountering visuals already prepared by some director is a gigantic one. Regardless of the world our minds can conceive, it seems there’s always someone there to make it. Along similar lines a Phd candidate at Stanford University along with Intel has created an AI) able to paint a completely new world from the ground up. However, these days, computers are able to do both the conception and the building.
The AI painting technology created by Qifeng Chen and Intel has been successful in creating a picture of a street view with the help of algorithms. The AI was fed with thousand of reference images and it educated itself regarding where a building, tree, road, car, sign and lane should be situated. Using these references the AI was able to paint a depiction of a street technologically that didn’t exist at all. This is similar to authors creating worlds which didn’t literally exist outside of their minds.
The scientists were able to use the template and the AI painting technology generated a colorful display which, with a little help from technicians, were made to look like images based on the real world. The AI uses its database and is able to create a completely distinctive image and video based on its memory. When the researcher directs the computer to produce an output or a certain visual, the AI processes the request and delivers the image.
This latest thrust in AI is expected to have a huge impact in the gaming world. If developers are able to create photo-realistic images via AI painting technology instead of the current blurry images then they can use this software to build city blocks, make virtual movie worlds, and everything else that could be done only with the help of coding. In fact, Chen has already used his AI to replace some images in Grand Theft Auto V.
Chen has said that the program could be used with voice commands in the future as “it would be really great fit if one could conjure up a photorealistic scene by just by describing it aloud.” It could also become a digital creator allowing one to create complete new virtual worlds, instead of just streets and buildings. If ever humans were to run short of fantasy genres and worlds, perhaps this AI painting technology could help us escape from our problem-ridden world. But that will be only temporary, just like the other fantasy worlds we’ve come to love.