Alice in Wonderland Gets Re-Envisioned by a Neural Network in the Style of Paintings By Picasso, van Gogh, Kahlo, O’Keeffe & More

An artist just start­ing out might first imi­tate the styles of oth­ers, and if all goes well, the process of learn­ing those styles will lead them to a style of their own. But how does one learn some­thing like an artis­tic style in a way that isn’t sim­ply imi­ta­tive? Arti­fi­cial intel­li­gence, and espe­cial­ly the cur­rent devel­op­ments in mak­ing com­put­ers not just think but learn, will cer­tain­ly shed some light in the process — and pro­duce, along the way, such fas­ci­nat­ing projects as the video above, a re-envi­sion­ing of Dis­ney’s Alice in Won­der­land in the styles of famous artists: Pablo Picas­so, Geor­gia O’Ke­effe, Kat­sushi­ka Hoku­saiFri­da Kahlo, Vin­cent van Gogh and oth­ers.

The idea behind this tech­no­log­i­cal process, known as “style trans­fer,” is “to take two images, say, a pho­to of a per­son and a paint­ing, and use these to cre­ate a third image that com­bines the con­tent of the for­mer with the style of the lat­er,” says an explana­to­ry post at the Paper­space Blog.

“The cen­tral prob­lem of style trans­fer revolves around our abil­i­ty to come up with a clear way of com­put­ing the ‘con­tent’ of an image as dis­tinct from com­put­ing the ‘style’ of an image. Before deep learn­ing arrived at the scene, researchers had been hand­craft­ing meth­ods to extract the con­tent and tex­ture of images, merge them and see if the results were inter­est­ing or garbage.”

Deep learn­ing, the fam­i­ly of meth­ods that enable com­put­ers to teach them­selves, involves pro­vid­ing an arti­fi­cial intel­li­gence sys­tem called a “neur­al net­work” with huge amounts of data and let­ting it draw infer­ences. In exper­i­ments like these, the sys­tems take in visu­al data and make infer­ences about how one set of data, like the con­tent of frames of Alice in Won­der­land, might look when ren­dered in the col­ors and con­tours of anoth­er, such as some of the most famous paint­ings in all of art his­to­ry. (Oth­ers have tried it, as we’ve pre­vi­ous­ly fea­tured, with 2001: A Space Odyssey and Blade Run­ner.) If the tech­nol­o­gy at work here piques your curios­i­ty, have a look at Google’s free online course on deep learn­ing or this new set of cours­es from Cours­era— it prob­a­bly won’t improve your art skills, but it will cer­tain­ly increase your under­stand­ing of a devel­op­ment that will play an ever larg­er role in the cul­ture and econ­o­my ahead.

Here’s a full list of painters used in the neur­al net­worked ver­sion of Alice:

Pablo Picas­so
Geor­gia O’Ke­effe
S.H. Raza
Hoku­sai
Fri­da Kahlo
Vin­cent van Gogh
Tar­si­la
Saloua Raou­da Chou­cair
Lee Kras­ner
Sol Lewitt
Wu Guanzhong
Elaine de Koon­ing
Ibrahim el-Salahi
Min­nie Pwer­le
Jean-Michel Basquiat
Edvard Munch
Natalia Gon­charo­va

via Kot­tke

Relat­ed Con­tent:

Kubrick’s 2001: A Space Odyssey Ren­dered in the Style of Picas­so; Blade Run­ner in the Style of Van Gogh

What Hap­pens When Blade Run­ner & A Scan­ner Dark­ly Get Remade with an Arti­fi­cial Neur­al Net­work

Google Launch­es Free Course on Deep Learn­ing: The Sci­ence of Teach­ing Com­put­ers How to Teach Them­selves

New Deep Learn­ing Cours­es Released on Cours­era, with Hope of Teach­ing Mil­lions the Basics of Arti­fi­cial Intel­li­gence

The First Film Adap­ta­tion of Alice in Won­der­land (1903)

Based in Seoul, Col­in Mar­shall writes and broad­casts on cities and cul­ture. He’s at work on the book The State­less City: a Walk through 21st-Cen­tu­ry Los Ange­les, the video series The City in Cin­e­ma, the crowd­fund­ed jour­nal­ism project Where Is the City of the Future?, and the Los Ange­les Review of Books’ Korea Blog. Fol­low him on Twit­ter at @colinmarshall or on Face­book.


by | Permalink | Comments (0) |

Sup­port Open Cul­ture

We’re hop­ing to rely on our loy­al read­ers rather than errat­ic ads. To sup­port Open Cul­ture’s edu­ca­tion­al mis­sion, please con­sid­er mak­ing a dona­tion. We accept Pay­Pal, Ven­mo (@openculture), Patre­on and Cryp­to! Please find all options here. We thank you!


Leave a Reply

Quantcast
Open Culture was founded by Dan Colman.