A Trip Through New York City in 1911: Vintage Video of NYC Gets Colorized & Revived with Artificial Intelligence

Denis Shi­rayev is at it again! The man who only a few weeks ago put one of the most famous pieces of film his­to­ry–the Lumiere Bros. footage of a train arriv­ing at La Cio­tat sta­tion–through a neur­al net­work to bring it “to life,” so to speak, has turned to anoth­er fas­ci­nat­ing slice of his­to­ry.

For his next install­ment, he has tak­en footage of New York City dai­ly life in 1911, eight min­utes of tram rides, horse-drawn wag­ons, the ele­vat­ed train, and the rush of crowd­ed streets, and applied the same deep learn­ing algo­rithms to make it all look like it was shot yes­ter­day. This time he had a bit of help from anoth­er YouTube historian/technician Guy Jones, who had already speed cor­rect­ed and tweaked the footage, as well as adding envi­ron­men­tal sounds. Shi­rayev has used AI to upscale the footage to 4K and to 60p.

The orig­i­nal footage was shot by Sven­s­ka Biografteatern, a Swedish news­reel com­pa­ny, and begins with a shot of the Stat­ue of Lib­er­ty as if seen through a spy­glass. The film con­tin­ues as trav­el­ogue and as an intro­duc­tion to the immi­grant expe­ri­ence, as the cam­era shows boats dock­ing, pas­sen­gers dis­em­bark­ing, and then the over­whelm­ing expe­ri­ence of New York City.

The footage is clear enough to take in store­fronts and adver­tis­ing on trams and the sides of build­ings. But the atmos­phere is too clogged with dai­ly smoke to get a real clear vista of the sky­line from the Brook­lyn Bridge.

At the time, Man­hat­tan had a pop­u­la­tion about 2 mil­lion. Inter­est­ing­ly, that was its height. Over a hun­dred years lat­er, that has declined to 1.6 mil­lion, with a sig­nif­i­cant decrease in pop­u­la­tion den­si­ty. This Observ­er arti­cle ascribes that to gen­tri­fi­ca­tion, and a change of res­i­den­tial areas to com­mer­cial ones.

And let’s repeat what we said about Shirayev’s pre­vi­ous 4K footage: this is not a “remas­ter”. This is not a “restora­tion.” This is using the pow­er of com­put­ing to inter­pret frames of film and cre­ate in between frames, as well as cre­ate detail from blur­ry footage. (I’m not too sure about the colorization–it doesn’t real­ly work as well as all the oth­er software…yet).

Now we know that Shi­rayev is mak­ing this a thing, please note his pinned mes­sage in the YouTube com­ments: he’s tak­ing requests.

via Laugh­ing Squid

Relat­ed Con­tent:

Icon­ic Film from 1896 Restored with Arti­fi­cial Intel­li­gence: Watch an AI-Upscaled Ver­sion of the Lumière Broth­ers’ The Arrival of a Train at La Cio­tat Sta­tion

Pris­tine Footage Lets You Revis­it Life in Paris in the 1890s: Watch Footage Shot by the Lumière Broth­ers

Immac­u­late­ly Restored Film Lets You Revis­it Life in New York City in 1911

Ted Mills is a free­lance writer on the arts who cur­rent­ly hosts the artist inter­view-based FunkZone Pod­cast and is the pro­duc­er of KCR­W’s Curi­ous Coast. You can also fol­low him on Twit­ter at @tedmills, read his oth­er arts writ­ing at tedmills.com and/or watch his films here.

by | Permalink | Comments (3) |

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!

Comments (3)
You can skip to the end and leave a response. Pinging is currently not allowed.

Leave a Reply

Open Culture was founded by Dan Colman.