Tuesday, July 7, 2015 | 1:17 p.m.
In case you thought the imagery in the movie “Fear and Loathing in Las Vegas” wasn’t strange enough, a hallucinatory effect from Google has taken the visuals to an even more trippy level.
A user on GitHub, a code-sharing service, posted directions for running video through an artificial neural network created by Google to identify the contents of different images to better classify them. To demonstrate, the GitHub user ran a scene from the 1998 film adaptation of Hunter S. Thompson’s book.
Seen through the eyes of a computer, Fear and Loathing is a dreamscape of morphed figures and half-armadillo, half-dog figures.
The Google neural networks are trained to interpret and amplify images, often making them stranger and more psychedelic.
The Fear and Loathing clip turns into an awesome nightmare. It’s a swirl of surreal, impressionistic renditions of animals: terriers, primates, birds, some sort of freaky beetle.
If you look hard enough, the cigarette holder dangling from Johnny Depp’s mouth might appear to turn into an elongated anchovy. That happens about the same time his head sprouts a snout.
Join the Discussion:
Check this out for a full explanation of our conversion to the LiveFyre commenting system and instructions on how to sign up for an account.
Full comments policy