Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Saturday June 08 2019, @05:55PM   Printer-friendly
from the Did-they-really-say-THAT? dept.

Stanford Engineers Make Editing Video as Easy as Editing Text:

In television and film, actors often flub small bits of otherwise flawless performances. Other times they leave out a critical word. For editors, the only solution so far is to accept the flaws or fix them with expensive reshoots.

Imagine, however, if that editor could modify video using a text transcript. Much like word processing, the editor could easily add new words, delete unwanted ones or completely rearrange the pieces by dragging and dropping them as needed to assemble a finished video that looks almost flawless to the untrained eye.

A team of researchers from Stanford University, Max Planck Institute for Informatics, Princeton University and Adobe Research created such an algorithm for editing talking-head videos – videos showing speakers from the shoulders up.

The work could be a boon for video editors and producers but does raise concerns as people increasingly question the validity of images and videos online, the authors said. However, they propose some guidelines for using these tools that would alert viewers and performers that the video has been manipulated.

"Unfortunately, technologies like this will always attract bad actors," said Ohad Fried, a postdoctoral scholar at Stanford. "But the struggle is worth it given the many creative video editing and content creation applications this enables."

There is a video on YouTube explaining the technique and containing numerous examples.

So, given enough sample data, you can type a transcript of what you want said, and a "talking head" version of the samplee can be created showing them saying exactly that.

Previously: House Intelligence Committee to Hold Hearing on "Deepfakes"


Original Submission

Related Stories

House Intelligence Committee to Hold Hearing on "Deepfakes" 28 comments

House Intelligence panel to examine 'deepfake' videos in June

The House Intelligence Committee has slated a hearing in June that will examine a series of national security matters, including the threat of videos manipulated by artificial intelligence (AI) that look strikingly real, a panel aide said.

The congressional hearing on June 13 will be one of the first to primarily focus on so-called deepfakes, which experts and lawmakers say pose a major disinformation threat heading into the 2020 election.

The hearing comes amid a spotlight on a fake video of Speaker Nancy Pelosi (D-Calif.) that made its way across Facebook and other social media platforms in late May.

This video, which was viewed millions of times, was edited in a way that made her appear to slur her words. While the video was not made with advanced AI technology, experts say that will be the next step in the disinformation threat.

Also at CNN and CNET.

See also: The 2020 campaigns aren't ready for deepfakes


Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 3, Funny) by Arik on Saturday June 08 2019, @06:04PM (3 children)

    by Arik (4543) on Saturday June 08 2019, @06:04PM (#853199) Journal
    Finally, we'll be able to edit video with our choice, EMACS /or/ VIM?
    --
    If laughter is the best medicine, who are the best doctors?
    • (Score: 2) by RS3 on Saturday June 08 2019, @10:09PM

      by RS3 (6367) on Saturday June 08 2019, @10:09PM (#853257)

      ed

    • (Score: 2) by darkfeline on Sunday June 09 2019, @09:00PM (1 child)

      by darkfeline (1030) on Sunday June 09 2019, @09:00PM (#853469) Homepage

      Emacs has had a video editor for decades: http://1010.co.uk/gneve.html [1010.co.uk]

      Vim is unfortunately lacking in this regard.

      --
      Join the SDF Public Access UNIX System today!
      • (Score: 2) by Arik on Sunday June 09 2019, @10:21PM

        by Arik (4543) on Sunday June 09 2019, @10:21PM (#853493) Journal
        Yes, that's why I put the emphasis on that /OR/.
        --
        If laughter is the best medicine, who are the best doctors?
  • (Score: 4, Insightful) by fyngyrz on Saturday June 08 2019, @06:31PM

    by fyngyrz (6567) on Saturday June 08 2019, @06:31PM (#853204) Journal

    These kinds of "guidelines" are only useful if:

    1. They are universally followed, or:
    2. You can tell they have not been followed and can address that.

    That is why guidelines for altering existing recordings, for limits on AI, for limits on genetic research, for limits on invasive surveillance, and a fair grab bag of other issues large and small all will have little to no effect upon those outside the public's eye and/or outside the law's reach who choose to operate outside them.

    That's also why government actors will engage in these things. Because these things are going to happen, and if they choose to eschew them themselves, they'll inevitably fall behind those who do not. In some cases, that will pose a significant threat to them.

    --
    I run like the winded.

  • (Score: 0) by Anonymous Coward on Saturday June 08 2019, @06:48PM (2 children)

    by Anonymous Coward on Saturday June 08 2019, @06:48PM (#853209)

    Enhance, enhance, rotate, through window, enhance!

  • (Score: 0) by Anonymous Coward on Saturday June 08 2019, @10:48PM (1 child)

    by Anonymous Coward on Saturday June 08 2019, @10:48PM (#853267)

    All vidoe is now deep fakes. Voices, what is said, what is in the background... VERY THING.

    This is great day for freedom. All CCD and Wiretaps are fakes.

    • (Score: 0) by Anonymous Coward on Sunday June 09 2019, @04:30AM

      by Anonymous Coward on Sunday June 09 2019, @04:30AM (#853309)

      Says the Russian troll... You should drink much more Vodak before posting comrade.

(1)