Face swap photo face for PC Archives

Will 2016 go down in history as the year we failed to heed the warning about artificial intelligence overthrowing humanity one stone at a time, because we were distracted by swapping faces with our friends, our pets or nearby breasts?

Perhaps. Face-swapping apps aren’t actually new – several have been around since 2013 – but they have become this year’s mobile craze, in a lineage that includes Draw Something, Dubsmash, Flappy Bird, FatBooth and (in the earliest days of Apple’s App Store) virtual pint-drinking and lightsaber apps.

The craze may only last as long as it takes for the next novelty app genre to emerge, but in the meantime, here are five apps for giving face-swapping a go.

MSQRD
Android / iOS

App crazes may come and go, but tech startups shunning vowels is a trend that will never die. MSQRD (i.e. Masquerade) made headlines this month when it was acquired by Facebook less than three months after launching. For now, the app remains available on Android and iOS.

MSQRD is pitched as a “video selfies” app for recording short clips of your digitally edited face – from adding virtual accessories through to celebrity face-masks, as well as swapping faces with friends. It’s easy to use, as you swipe between effects, with quick sharing to Instagram, Facebook and Twitter.

Face Swap Live
iOS

Face Swap Live may not have been acquired by a tech giant, but it has made plenty of headlines thanks to this feat of ingenuity:

What times we live in! For now, Face Swap Live is iOS-only, although its developer is getting Android users to sign up to be notified when it gets ported to Google-powered smartphones.

Live face swaps with friends (or body parts) are its key appeal, although Face Swap Live also works with photos, including pics of celebrities pulled from the web. The celebrity angle is fuelling the face-swap buzz: mapping your face on to Donald Trump’s infamous hairline is the new political satire. Possibly.

Snapchat
Android / iOS

With its 100 million daily active users, Snapchat has been one of the key ways face-swapping technology has been introduced to a mainstream audience. Face-swapping was added as one of the app’s “lenses” in February 2016, likely using technology from another acquisition – of startup Looksery the year before.

Like other lenses, the face-swapping feature is accessed by lining up a selfie, then pressing and holding on your face, before swiping along a carousel of special effects. Line up the on-screen markets with a friend’s face, and marvel at the horror. It’s not a reason to download Snapchat if you haven’t already, but for users it’s another freakish-but-fun reason to keep snapping.

Face Swap Booth
Android / iOS

It may be lacking the media buzz of MSQRD and Face Swap Live – not to mention the video aspects – but if you want to simply mix and match faces from photos, it’s a good free app to experiment with.

An accessible interface masks some pretty powerful editing features, as well as handy options like saving a face to swap in to other photos, and a preloaded bank of celebrity photos to use. For static face-swapping, the fine-tuning options make this well worth a try.

Face Stealer
iOS

Included as a reminder that Yahoo has managed to be one of the more inventive companies around mobile in recent years, even if it hasn’t had the best of luck capitalising on that cleverness. Yahoo’s Japanese division released this photo face-swapping app in February 2013, but three years on, newer startups have stolen its thunder.

The language here is about “stealing” other people’s faces rather than swapping with them, which may be one reason: it sounds a bit more horror movie. The app is fun although more limited in terms of subjects and scale: Einstein, the Mona Lisa and a few other options.

•‘I’m speechless. It’s revolting!’ Why I love faceswapping apps

^{This article contains affiliate links, which means we may earn a small commission if a reader clicks through and makes a purchase. All our journalism is independent and is in no way influenced by any advertiser or commercial initiative. By clicking on an affiliate link, you accept that third-party cookies will be set. More information.}

Thanks to the wonders of computer vision and a goofy new app called Face Swap Live, I am controlling Hilary's face - with nothing more than the expression on mine.

If you haven't yet experienced the viral, nightmarish joys of Face Swap Live, it's well worth the 99 cents it's currently selling for. Download the app and point your phone's camera at a friend, and it will convincingly map their face, in real time, onto someone else's: yours, a baby's, Beyonce's, Richard Nixon's.

Since appearing in the app store about a month ago, the app hasn't strayed far outside of Apple's most-downloaded offerings, peaking in the United States. at No. 14. And while the technology isn't perfect - the app's first truly ubiquitous meme was a disastrous face-swap between a dad and his baby - the results are lifelike enough, enough of the time, that the "Today" show dubbed it "Kafkaesque" and the Daily Dot called it "creepy."

"It's a lot of fun," said Jason Laan, one of the app's two creators. "But behind the fun, there's some really amazing, hardcore technology."

Laan, a chemical engineer by training, has a long history of turning serious tech to more frivolous purposes: In the eight years since he founded his app development firm, Laan Labs, he estimates that they've launched around 50 products, from Tap DJ ("mix and add FX to your iPod music!") to Dog Vision HD ("see the world how your dog sees it!").

But for Laan, computer vision - the science of training computers to extract and understand information from pictures, the same way humans do - has always possessed a special intrigue. Researchers at places like Google and IBM, with their extraordinary 3-D cameras and lightning-fast processing speeds, had enabled computers to catalog objects, recognize faces and even interpret feelings. Laan and his partner, Will Perkins, began wondering if the iPhone's improving camera and processing capabilities would allow them to try out similar projects, albeit less seriously.

So late last winter, Face Swap Live was conceived. The app that has since launched a thousand YouTube videos, Imgur posts and nightmare memes.

Face-swapping makes a pretty ideal consumer application for the new computer vision tools, incidentally. While the technology is novel, the art form is not: Know Your Meme traces the first instances back to the early aughts, when the visages of an eccentric Vietnamese singer and and a 16-year-old Chinese kid began showing up on other bodies and in other places.

In 2004, the Something Awful forums fatefully began switching the faces of babies and their grandparents. It was an onerous Photoshop process, a labor of lolz, if you will: isolating the faces manually; copying, moving and rotating them; blending and feathering the mismatched edges until the heads and bodies fit. Even all that work made for some pretty unholy collages: The babies' heads pixelated, over-large; the grandparents' shrunken and neckless.

"Wasn't that (expletive) creepy?" exclaimed an SA writer in 2004. "Now I have to go to bed ... Oh, the dreams I'm going to have."

But the appeal of the face-swap has always been its weirdness - the degree to which it inverts and diverges from reality. The best face-swaps are also the most surreal: Tom Cruise as Jack Nicholson, Barack Obama as George Bush, Nicholas Cage as literally everybody.

"There's something about absurdity that gives Internet memes a lot of traction," said Britney Summit-Gil, a doctoral student at Rensselaer Polytechnic Institute who studies Reddit culture. "Absurdity has a long, storied history of entertaining humans." And it's not so different from awe, she says - one of our more viral emotions.

Oddly enough, however, we're moving closer to a world where face swaps are both less "awesome," in the sense of inspiring wonder, and less obviously absurd. Thanks to innovations like the ones that spurred Face Swap Live, face-swapping no longer requires any time at all, to say nothing of expensive editing software and human effort.

Just look at how fast Face Swap Live is, an accomplishment Laan and Perkins credit to a basket of cutting-edge algorithms. When they look at your face, they simply look for reference points - the corners of your eyes where the color changes, the curve of your chin - and then line them up with those points on another face, auto-smoothing and blending them in.

With better cameras, Laan and Perkins say (3D cameras, particularly, like the ones Intel just unveiled at CES), our smartphones could do far more than copy-paste a face. Already, Disney is working on a technology that can map your face down to its individual wrinkles. At Stanford University and Germany's Max Planck Institute, researchers have developed a technique that photorealistically transfers one person's facial expressions to another - not face-swapping, in the traditional sense, but face-hijacking via algorithm.

These researchers believe we're moving closer to a world where remote workers can Skype into meetings half-clothed, their faces mapped onto a body in a business suit. They suspect we'll be able to tweak actors' bad takes and zap unsuspecting bystanders from live TV news.

Far outside the realm of face-swapping, real-time computer vision - particularly of human bodies and faces - will enable a million other technologies: self-driving cars, diagnostic computers, robots that understand emotions and react accordingly. We won't even delve into the more dystopian applications, like live video-manipulation or mass surveillance.

I ask Perkins and Laan about that, because it's seems odd: a silly app that advances a promising, and ominous, technology. Do they contemplate the juxtaposition at all, I wonder?

"We just want to have fun," Laan says. Then they both laugh nervously.

© 2016 The Washington Post

Image manipulation in this advanced stage of the digital age is not as much fun but a dicey weapon, in the shadows of fake news, to sway opinion and spark scandals.

Face-swapping, in particular, sounds like fun if you think of it as a chuckle at a family table while kids and adults try out different faces on different people. However, it's also a tool for far worse motives. Swapna Krishna in Engadget remarked that "People have, of course, taken advantage of this tool for some disturbing uses, including face-swapping people into pornographic videos—the ultimate revenge porn."

In MIT Technology Review, the "Emerging Technology from the arXiv" said, "pornographic videos called 'deepfakes' have emerged on websites showing famous individuals' faces superimposed onto bodies of actors."

Researchers, however, interested in exploring the tool and how to tell if it is used, have come up with an algorithm, say observers, that can outdo other techniques available. They figured out a way to detect a face swap via the algorithm, picking up on forged videos as soon as posted.

Analytics Vidhya commented, "Something akin to this algorithm was desperately required to wage the battle against face swaps being used for the wrong reasons. In releasing the research paper to the public, the researchers are hoping others also take up the baton and work on this study to make it more accurate and precise."

Andreas Rossler was team leader of the participants from Technical University of Munich, University Federico II of Naples and the University of Erlangen-Nuremberg.

They trained the algorithm, XceptionNet, using a large set of face swaps, said Engadget.

"We set a strong baseline of results for detecting a facial manipulation with modern deep-learning architectures," said Rossler and team in MIT Technology Review. Size mattered.

The size of this database was a significant improvement over what had been previously available. "We introduce a novel data set of manipulated videos that exceeds all existing publicly available forensic data sets by orders of magnitude," said Rossler.

In their paper, the authors said they introduced a face manipulation dataset, FaceForensics, "of about half a million edited images (from over 1000 videos)."

The paper is titled "FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces," on arXiv. Authors are Andreas Rössler, Davide Cozzolino, Luisa Verdoliva, Christian Riess, Justus Thies and Matthias Nießner.

The authors called attention to the difficulty—for humans and computers— in trying to distinguish between original and manipulated video, "especially when the videos are compressed or have low resolution, as it often happens on social networks."

They also called attention to the fact that "Research on the detection of face manipulations has been seriously hampered by the lack of adequate datasets."

There is a nuance in their success, though, that also merits attention. The "Emerging Technology from the arXiv" article called it the "sting in the tail." What is it? "The same deep-learning technique that can spot face-swap videos can also be used to improve the quality of face swaps in the first place—and that could make them harder to detect."

Engadget similarly said, "XceptionNet clearly outperforms its rival techniques in detecting this kind of fake video, but it also actually improves the quality of the forgeries. Rossler's team can use the biggest hallmarks of a face swap to make the manipulation more seamless. It doesn't fool XceptionNet, but in the long run, it could make it harder for other methods to detect faked videos."

Pranav Dar, in Analytics Vidhya, also weighed in on what he called "a caveat with this algorithm – it can also potentially be used to improve the quality of the face swaps which will make it harder to detect the fake. Also, as soon as a forgery detection algorithm is launched, the scammers always try to refine their model to stay a step ahead."

Nonetheless, the authors said, "our refiner mainly improves visual quality, but it only slightly encumbers forgery detection for deep learning method trained exactly on the forged output data."

© 2018 Tech Xplore