Quantcast
Connect with us

Deepfakes: Here’s how to detect fake online political videos with one simple tell

Published

on

A new form of misinformation is poised to spread through online communities as the 2018 midterm election campaigns heat up. Called “deepfakes” after the pseudonymous online account that popularized the technique – which may have chosen its name because the process uses a technical method called “deep learning” – these fake videos look very realistic.

So far, people have used deepfake videos in pornography and satire to make it appear that famous people are doing things they wouldn’t normally. But it’s almost certain deepfakes will appear during the campaign season, purporting to depict candidates saying things or going places the real candidate wouldn’t.

ADVERTISEMENT

It’s Barack Obama – or is it?

Because these techniques are so new, people are having trouble telling the difference between real videos and the deepfake videos. My work, with my colleague Ming-Ching Chang and our Ph.D. student Yuezun Li, has found a way to reliably tell real videos from deepfake videos. It’s not a permanent solution, because technology will improve. But it’s a start, and offers hope that computers will be able to help people tell truth from fiction.

What’s a ‘deepfake,’ anyway?

Making a deepfake video is a lot like translating between languages. Services like Google Translate use machine learning – computer analysis of tens of thousands of texts in multiple languages – to detect word-use patterns that they use to create the translation.

Deepfake algorithms work the same way: They use a type of machine learning system called a deep neural network to examine the facial movements of one person. Then they synthesize images of another person’s face making analogous movements. Doing so effectively creates a video of the target person appearing to do or say the things the source person did.

How deepfake videos are made.

Before they can work properly, deep neural networks need a lot of source information, such as photos of the persons being the source or target of impersonation. The more images used to train a deepfake algorithm, the more realistic the digital impersonation will be.

ADVERTISEMENT

Detecting blinking

There are still flaws in this new type of algorithm. One of them has to do with how the simulated faces blink – or don’t. Healthy adult humans blink somewhere between every 2 and 10 seconds, and a single blink takes between one-tenth and four-tenths of a second. That’s what would be normal to see in a video of a person talking. But it’s not what happens in many deepfake videos.

A real person blinks while talking.
A simulated face doesn’t blink the way a real person does.

When a deepfake algorithm is trained on face images of a person, it’s dependent on the photos that are available on the internet that can be used as training data. Even for people who are photographed often, few images are available online showing their eyes closed. Not only are photos like that rare – because people’s eyes are open most of the time – but photographers don’t usually publish images where the main subjects’ eyes are shut.

Without training images of people blinking, deepfake algorithms are less likely to create faces that blink normally. When we calculate the overall rate of blinking, and compares that with the natural range, we found that characters in deepfake videos blink a lot less frequent in comparison with real people. Our research uses machine learning to examine eye opening and closing in videos.

ADVERTISEMENT

This gives us an inspiration to detect deepfake videos. Subsequently, we develop a method to detect when the person in the video blinks. To be more specific, it scans each frame of a video in question, detects the faces in it and then locates the eyes automatically. It then utilizes another deep neural network to determine if the detected eye is open or close, using the eye’ appearance, geometric features and movement.

We know that our work is taking advantage of a flaw in the sort of data available to train deepfake algorithms. To avoid falling prey to a similar flaw, we have trained our system on a large library of images of both open and closed eyes. This method seems to work well, and as a result, we’ve achieved an over 95 percent detection rate.

This isn’t the final word on detecting deepfakes, of course. The technology is improving rapidly, and the competition between generating and detecting fake videos is analogous to a chess game. In particular, blinking can be added to deepfake videos by including face images with closed eyes or using video sequences for training. People who want to confuse the public will get better at making false videos – and we and others in the technology community will need to continue to find ways to detect them.The Conversation

ADVERTISEMENT

Siwei Lyu, Associate Professor of Computer Science; Director, Computer Vision and Machine Learning Lab, University at Albany, State University of New York

This article was originally published on The Conversation. Read the original article.

Enjoy this piece?

… then let us make a small request. Like you, we here at Raw Story believe in the power of progressive journalism — and we’re investing in investigative reporting as other publications give it the ax. Raw Story readers power David Cay Johnston’s DCReport, which we've expanded to keep watch in Washington. We’ve exposed billionaire tax evasion and uncovered White House efforts to poison our water. We’ve revealed financial scams that prey on veterans, and legal efforts to harm workers exploited by abusive bosses. We’ve launched a weekly podcast, “We’ve Got Issues,” focused on issues, not tweets. And unlike other news outlets, we’ve decided to make our original content free. But we need your support to do what we do.

Raw Story is independent. You won’t find mainstream media bias here. We’re not part of a conglomerate, or a project of venture capital bros. From unflinching coverage of racism, to revealing efforts to erode our rights, Raw Story will continue to expose hypocrisy and harm. Unhinged from billionaires and corporate overlords, we fight to ensure no one is forgotten.

We need your support to keep producing quality journalism and deepen our investigative reporting. Every reader contribution, whatever the amount, makes a tremendous difference. Invest with us in the future. Make a one-time contribution to Raw Story Investigates, or click here to become a subscriber. Thank you. Click to donate by check.

Enjoy this piece?

… then let us make a small request. Like you, we here at Raw Story believe in the power of progressive journalism — and we’re investing in investigative reporting as other publications give it the ax. Raw Story readers power David Cay Johnston’s DCReport, which we've expanded to keep watch in Washington. We’ve exposed billionaire tax evasion and uncovered White House efforts to poison our water. We’ve revealed financial scams that prey on veterans, and efforts to harm workers exploited by abusive bosses. We’ve launched a weekly podcast, “We’ve Got Issues,” focused on issues, not tweets. Unlike other news sites, we’ve decided to make our original content free. But we need your support to do what we do.

Raw Story is independent. You won’t find mainstream media bias here. We’re not part of a conglomerate, or a project of venture capital bros. From unflinching coverage of racism, to revealing efforts to erode our rights, Raw Story will continue to expose hypocrisy and harm. Unhinged from corporate overlords, we fight to ensure no one is forgotten.

We need your support to keep producing quality journalism and deepen our investigative reporting. Every reader contribution, whatever the amount, makes a tremendous difference. Invest with us in the future. Make a one-time contribution to Raw Story Investigates, or click here to become a subscriber. Thank you.



Report typos and corrections to: [email protected]. Send news tips to: [email protected].
READ COMMENTS - JOIN THE DISCUSSION
Continue Reading

Breaking Banner

Doctors treat Ruth Bader Ginsburg for tumor on her pancreas: SCOTUS spokesperson

Published

on

Supreme Court Justice Ruth Bader Ginsburg this week had underwent a procedure to treat a tumor from her pancreas, a court spokesperson tells Washington Post reporter Robert Barnes.

According to the Court, "the tumor was treated definitively" with radiation "and there is no evidence of disease elsewhere in the body."

the 86-year-old Ginsburg late last year underwent surgery to treat lung cancer and she returned to the bench this past February.

Continue Reading

Breaking Banner

REVEALED: Trump associate Felix Sater gave Osama bin Laden’s phone numbers to the feds

Published

on

The Russian businessman Felix Sater has been linked to Trump family enterprises for years. He was central to the effort to bring Trump Tower to Moscow. He also worked on the Trump SoHo condominium hotel in the mid-aughts.

During that time, Sater also worked as an informant for the FBI and CIA and aided prosecutors on various cases. He started working with the FBI in 1998, after he was caught in a stock-fraud scheme, reports the Wall Street Journal.

That's been previously reported. But what the WSJ found is that a letter just unsealed from the government praises Sater for being an especially good collaborator.

Continue Reading
 

Breaking Banner

Judges ‘frustrated’ after lawyers for Trump’s banks refuse to say if they have his tax returns: CNN

Published

on

Lawyers representing two of President Donald Trump's main banks upset a three-judge panel at an appellate court in New York when they refused to reveal whether they are in possession of President Donald Trump's tax returns.

CNN reports that lawyers for Deutsche Bank and Capital One "frustrated" the judges after they repeatedly cited "contractual obligations" to justify not divulging information about the president's taxes.

"The three-judge panel was so frustrated by the lawyers' refusal to answer that at one point one of the judges suggested the appellate court might seek an order for the information," reports CNN. "After several minutes of argument on the subject, the panel directed the lawyers to file letters under seal within 48 hours saying whether the banks have Trump's tax returns."

Continue Reading
 
 

Thank you for whitelisting Raw Story!

As a special thank you, from now until August 31st, we're offering you a discounted rate of $5.99/month to subscribe and get ad-free access. We're honored to have you as a reader. Thank you. :) —Elias, Membership Coordinator
LEARN MORE
close-link
close-image