The Earliest Known Motion Picture, 1888’s Roundhay Garden Scene, Restored with Artificial Intelligence

No image is more close­ly asso­ci­at­ed with the birth of the motion pic­ture than a train pulling into the French coastal town of La Cio­tat. Cap­tured by cin­e­ma pio­neers Auguste and Louis Lumière, the 50-sec­ond clip fright­ened the audi­ence at its first screen­ing in 1896, who thought a real loco­mo­tive was hurtling toward them — or so the leg­end goes. Those ear­ly view­ers may sim­ply have felt a tech­no­log­i­cal aston­ish­ment we can no longer muster today, and cer­tain­ly not in response to such a mun­dane sight. That goes dou­ble for the slight­ly short­er and old­er Lumière Broth­ers pro­duc­tion La Sor­tie de l’U­sine Lumière a Lyon. Though it depicts noth­ing more than work­ers leav­ing a fac­to­ry at the end of the day, it has long been referred to as “the first real motion pic­ture ever made.”

That qual­i­fi­er “real,” of course, hints at the exis­tence of a pre­de­ces­sor. Where­as La Sor­tie de l’U­sine Lumière a Lyon pre­miered in 1895, Louis Le Prince’s Round­hay Gar­den Scene dates to 1888. With its run­time under two sec­onds, this depic­tion of a moment in the life of four fig­ures, a younger man and woman and an old­er man and woman, would even by the stan­dards of the Lumière Broth­ers’ day bare­ly count as a movie at all.

Equal­ly dis­qual­i­fy­ing is its low frame rate: just sev­en to twelve per sec­ond (which one it is has been a mat­ter of some dis­pute), which strikes our eyes more as a rapid sequence of still pho­tographs than as con­tin­u­ous motion. Even so, it must have been a thrill of a result for Le Prince, an Eng­land-based French artist-inven­tor who had been devel­op­ing his motion-pho­tog­ra­phy sys­tem in secre­cy since ear­ly in the decade.

We now have a clear­er sense of the action cap­tured in Round­hay Gar­den Scene thanks to the efforts Youtube-based film restora­tionist Denis Shiryaev, who’s used neur­al net­works to bring the his­toric film more ful­ly to life. Tak­ing a scan of Le Prince’s orig­i­nal paper film, Shiryaev “man­u­al­ly cut this scan into indi­vid­ual frames and cen­tered each image in the frame,” he says in the video at the top of the post. He then “added a sta­bi­liza­tion algo­rithm and applied an aggres­sive face recog­ni­tion neur­al net­work in order to add more details to the faces.” There fol­lowed adjust­ments for con­sis­ten­cy in bright­ness, dam­age repairs, and the work of “an ensem­ble of neur­al net­works” to upscale the footage to as high a res­o­lu­tion as pos­si­ble, inter­po­lat­ing as many frames as pos­si­ble. We may feel star­tled by the life­like qual­i­ty of the result in much the same way as 19th-cen­tu­ry view­ers by the Lumière Broth­ers’ train — which, as we’ve pre­vi­ous­ly fea­tured here on Open Cul­ture, has also received the Shiryaev treat­ment.

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

Watch Scenes from Belle Époque Paris Vivid­ly Restored with Arti­fi­cial Intel­li­gence (Cir­ca 1890)

Watch Scenes from Czarist Moscow Vivid­ly Restored with Arti­fi­cial Intel­li­gence (May 1896)

Watch AI-Restored Film of Labor­ers Going Through Life in Vic­to­ri­an Eng­land (1901)

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

Based in Seoul, Col­in Mar­shall writes and broad­casts on cities, lan­guage, and cul­ture. His projects include the book The State­less City: a Walk through 21st-Cen­tu­ry Los Ange­les and the video series The City in Cin­e­ma. Fol­low him on Twit­ter at @colinmarshall, on Face­book, or on Insta­gram.


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