AI’s Halloween – Can machines learn to scare us?
by Chahira Akel
AI’s Halloween – Can machines learn to scare us? It is late October and the artificial intelligence (AI) experts at MIT have turned their attentions to the stuff of nightmares and are building an intelligent machine that can learn how to scare the pants off of people, just in time for Halloween.
According to The Verge, a three-person team has created something called the Nightmare Machine that has the sole purpose of taking regular images and turning them into something that might make you leap behind a sofa.
The Nightmare Machine is online, and if you have a go at it you will be helping to inform its learning and help it create better, more ghastly images that are as scary as they could be.
The machine creates nightmare faces and nightmare places, there are three stages involved and in each the nightmare element ramps up. There are a few filters; one adds tentacle monsters to the mix, while another makes things look like a ghost town. You can see the effect that the system has on the Sydney Opera House in the image.
There are a whole range of images on the Nightmare Machine Instagram, and if you like that sort of thing you might want to check them out. The important thing is that you don’t have nightmares, although if you did you would probably be proving how good it was.
“Since centuries, across geographies, religions, and cultures people try to innovate ways of scaring each other. Creating a visceral emotion such a fear remains one of the cornerstones of human creativity,” says the MIT trio by way of introduction.
“This challenge is especially important in a time where we wonder what the limits of Artificial Intelligence are: Can machines learn to scare us? Towards this goal, we present you Haunted Faces and Haunted Places: computer generated scary imagery powered by deep learning algorithms and evil spirits!”
Source: theinquirer.net
Also read more about the project here
The post AI’s Halloween – Can machines learn to scare us? appeared first on Internet of Things Event.
Comments
Post a Comment