{"id":147630,"date":"2021-01-05T17:56:22","date_gmt":"2021-01-05T14:56:22","guid":{"rendered":"https:\/\/en.buradabiliyorum.com\/watch-ai-transform-a-montage-of-lumiere-films-historians-will-hate-it\/"},"modified":"2021-01-05T17:56:22","modified_gmt":"2021-01-05T14:56:22","slug":"watch-ai-transform-a-montage-of-lumiere-films-historians-will-hate-it","status":"publish","type":"post","link":"https:\/\/buradabiliyorum.com\/en\/watch-ai-transform-a-montage-of-lumiere-films-historians-will-hate-it\/","title":{"rendered":"#Watch AI transform a montage of Lumi\u00e8re films (historians will hate it)"},"content":{"rendered":"

“#Watch AI transform a montage of Lumi\u00e8re films (historians will hate it)<\/strong>”
\n<\/p>\n

\n YouTube<\/a>r Denis Shiryaev has unveiled his most spectacular AI video restoration yet: a new montage of footage shot\u00a0by the Lumi\u00e8re brothers.<\/p>\n

Shiryaev spent months using machine learning to colorize,\u00a0upscale, and enhance\u00a0a collection of films<\/a> by the French cinema pioneers.<\/p>\n

He began by using an algorithm developed by his company,\u00a0Neural Love, to remove duplicate frames from the footage and restore the film\u2019s original frame rate.<\/p>\n

\u201cThis is a crucial step to achieve the effect of fluid movement when we do a version of the video at 60 frames per second (fps),\u201d\u00a0Shiryaev explained on YouTube<\/a>.<\/p>\n

[Read:\u00a0Meet the 4 scale-ups using data to save the planet]<\/em><\/p>\n

He also app<\/a>lied stabilization to the footage, removed noise where required, and corrected some of the damage to the films.<\/p>\n

Shiryaev then used neural networks to upscale every frame to 4K resolution,\u00a0generate more frames\u00a0until he reached 60 fps, and restore a realistic playback speed to the video.<\/p>\n

He also colorized the footage and added detail to the\u00a0faces\u00a0with a generative adversarial network (GAN) based algorithm.<\/p>\n

Finally, he added new sounds to the film:<\/p>\n

The sounds are not real, and sometimes I did a terrible job.\u00a0But I really like this ambient effect\u00a0when, for example, you can see and hear a dog starting to bark on screen and experience the moment as if you were there.<\/p>\n<\/blockquote>\n

You can check it out for yourself in the video below:
\nhttps:\/\/www.youtube.com\/watch?v=YZuP41ALx_Q<\/a>
\nThe video joins\u00a0Shiryaev\u2019s growing filmography of AI-powered restorations. But not everyone approves of his work.<\/p>\n

Academics warn<\/a> that it creates a distorted image of the past and undermines the choices of the original photographer. They fear that every added effect also takes something away from what was captured at the time.<\/p>\n

Shiryaev acknowledges that the videos are not historically accurate, and only intended for entertainment or artistic purposes:<\/p>\n

Neural networks can only guess colors when we\u2019re doing colorization, or new pixels when we\u2019re doing upscale, and faces when working with some facial restoration.<\/p>\n<\/blockquote>\n

Nonetheless, the artificial sights and sounds provide a new window to the past \u2014 even if it is a pretty cloudy one.<\/p>\n

\n Published January 5, 2021 \u2014 14:56 UTC<\/p><\/div>\n