In 2017, robots can make art, our phones remember our faces, and computers, well… computers are able to recognize the motifs of Wes Anderson’s films. As originally reported by Mashable, software developer Yannick Assogba has used Google’s Inception V3 program to analyze and categorize Wes Anderson’s aesthetic in breakneck speed — no binge night required.
Google’s Inception V3 program is described as a deep neural network or, in n00b terms, a “machine learning” network that, after a training process, is able to focus in on specific information. Using the program, Assogba was able to quickly produce eagle-eyed findings about Wes’s stylistic choices that would require the average person multiple viewings.
The program found that shelves, tents, and TV screens were some of the most common objects in Wes’s films. And by studying every pixel in Moonrise Kingdom, it was able to find out that warm colors were the most popular hues.
How exactly was the program able to become an adept film critic? Assogba gives an a very technical explanation to Mashable that you can read in full here. But here’s the gist: “During training hundreds of thousands of labeled (i.e. categorized) images are fed into the network and the strengths of connections between nodes is continually adjusted until the network accurately predicts the labels for known images,” he says. We’d like to use the program to dissect Wes’s styling choices for his characters. For example, how many athleisure pieces can be found in The Life Aquatic?
Read More: Reddit Users Are Sharing Dreamy Wes Anderson Locations