this post was submitted on 05 Dec 2023
441 points (97.6% liked)

Not The Onion

12558 readers
761 users here now

Welcome

We're not The Onion! Not affiliated with them in any way! Not operated by them in any way! All the news here is real!

The Rules

Posts must be:

  1. Links to news stories from...
  2. ...credible sources, with...
  3. ...their original headlines, that...
  4. ...would make people who see the headline think, “That has got to be a story from The Onion, America’s Finest News Source.”

Comments must abide by the server rules for Lemmy.world and generally abstain from trollish, bigoted, or otherwise disruptive behavior that makes this community less fun for everyone.

And that’s basically it!

founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] Senokir 1 points 1 year ago* (last edited 1 year ago) (1 children)

That's a fair point. I have never seen it done from a video before specifically, but I am positive that it is a technique which is theoretically possible given that there is enough data in the image. Obviously if the image was grainy to begin with then it doesn't matter what you do to it, you won't get anything better than the original. And regardless of how the file is exported, as long as you can take a screenshot of the video afterwards and there is enough definition in the image I don't see how this technique couldn't be applied.

Edit: and to be clear, I don't know what specific transformation(s) are traditionally used in video editing. For all I know it could be a long list of transformations that are all coded to happen with the click of a button to make it more difficult to unblur. But even that isn't entirely safe. There is just literally no reason to not use a black box/elipse or whatever in cases where the data is actually sensitive.

[–] ericisshort 7 points 1 year ago* (last edited 1 year ago) (2 children)

I hate to break it to you, but I’m pretty sure the only way unblurring would be possible is if you have the original unblurred source video. Blurred areas simply contain too little data to reconstruct the original. The artifacts of the unblurred face are not stored in the resulting video file for literally every video file type in existence today.

Even using AI to unblurr a video, you would need to show it at least one unblurred picture of the face to properly reconstruct the same face in the original video. Otherwise the AI would just guess and put a random face in place of the original. The regeneration technique you are describing is still science fiction and only exists in movies and television shows.

Edit: your edit is still incorrect. There is no such video editing tool called “unblur.” You can sharpen an image, but there is no way to undo a blur unless you still have the data from the original, which a final exported video does not. The act of blurring removes information, and you cannot rebuild a photo from information that no longer exists. Blacking out faces is in no way necessary for the reasons that you are suggesting.

[–] [email protected] 1 points 1 year ago

Has anyone trained an AI with mug shots? 😅