Turns Out There’s A Tool That Brings Back Sharpness Of Low-Res Photos And Some Results Are A Bit Creepy
InterviewAs soon as the internet was presented with a face depixelizer that takes advantage of top-of-the-line neural network technology to take super-pixelated photos and reconstruct them into high-quality portraits of what the AI thinks is the right person based on complex algorithms and programming ingenuity… some people began testing it with pictures that the AI was not really designed for, resulting in some pretty creepy creations.
And despite people knowing that it doesn’t work ideally all the time, they are still trying it out because it’s exciting to see what other bizarre ideas the AI will decide to make a reality with its AI brain, leading to a number of very interesting results.
Bored Panda invites you to check out some of the results that the internet has reached with its own experimenting on the face depixelizer. While you’re there, be sure to leave an upvote and a comment on the ones you think are great.
This post may include affiliate links.
Am I the only one who thinks he looks like a chonky Bill Hader??
Looks like the love child of Roel van Velzen and Jeroen van der Boom...
it looks like it already has been...multiple times
Load More Replies...Inside that pixelized face of a Creeper lies a screaming middle-aged man, probably in mid-life crisis. Makes one think. Kinda explains why it goes around screeching and destroying stuff that young people made.
The meat and bones of the tool were created by Alex Damian and his colleagues, while Denis Malimonov, a popularizer of Machine Learning in Art, gave it a simple and easy to use interface, to make it more accessible to a wider audience. He posted a tweet about it and people were engaged immediately, showing off their results. We’ve also got in touch with Malimonov for more information.
Malimonov used to be a designer, and now began dabbling in programming. Since his target audience (for the most part) consists of people who have never programmed before, he wanted to make sure that they do not experience difficulties in testing neural network technologies.
Looks like they're either very exhausted or got their a*s handed to them.
He looks like Ron weasley in the last deathly Hallows movie
The tool is based on the “PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Models” study. The way the algorithm works is the AI gets a pixelated image, which it compares to a bunch of proper quality portraits that it also pixelates down to the necessary quality to find the one that looks the most identical to the original picture. When it finds the right one, it adds several filters and adaptations to make it look even more like the original, and voila, you have a depixelated picture.
“Users immediately began testing the neural network using characters from computer games, and began wondering why the algorithms work so poorly,” explained Malimonov. “Then they began to try on real people's faces, the first of whom was Barack Obama. The result was a dark man of European appearance.”
He's got the face of a 74 year old, and the intelligence of a 7 year old.
That is a horrible thing to say about a seven year old.
Load More Replies...SCREECH BEFORE IT'S TOO LATE!SCREEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEECH!
guys i found the demon in charge of sending all the karens alerts and enstructions
And, naturally, one should have expected some discrepancies, not just because there is only so much a computer can do with no more than a hundred pixels, but also because the neural network was very limited in what it was, as explained Malimonov:
“The reason is very simple: this neural network uses the pre-trained CelebA-HQ model. The model was trained using faces of various celebrities. The majority of faces were of white people, amounting to over 7,000 pictures, while Asian, black, and Indian were around or under 1,000 pictures. As far as I know, there is no such model in which the distribution across races is uniform.”
Thanos (Snap to half the universe's population) to Thanis (snapping at a poetry slam)
Did-ja know that the name Thanos comes from a word that means desire to kill. Cool right?
Now, people have been putting up pictures of practically everything that’s effectively a face. Mario, the Creeper, Doom Guy, emojis, memes—anything that technically has a face. And while some resulted in pretty good renditions, others were rather questionable. Given the limited sample of the neural network, oddities were bound to happen.
oH NO FRISKY MY LAD WHAT HAVE THEY DONE TO YOU MY PRESHIS PACIFIST CHILD
Even pixelated like this, most people will recognize Barack Obama. The AI prefers to whitewash him though :-/
Am I tripping, is the left Former President Barack Obama? And if so, the AI made him into a white guy?!?
Algorithm probably uses publicly available content. Most of this is Caucasian, so the algorithm skews towards a lighter tint. This is important, to realize, because it's creates and even emphasizes existing bias.
Just an FYI: the term "Caucasian" to describe all white people is a misnomer created by the White supremacist pseudoscience of race categorization as decided by racist German Johann Blumenbach. Unless you're using it to describe people from the Georgia, Armenia, Azerbaijan area, please stop using it.
Load More Replies...Obama is very light skinned and has almost no black facial features, so the computer doesn't see that in those pixels. We see it because we know the guy.
Hahahahahaha, what? How is that accurate? It's gotten better results with the Minecraft looking pictures lol
That’s the lady from The Handmaid’s Tale, Invisible Man and Top of the Lake etc. 😂 100 %
Yes! Elizabeth Moss... Uh oh...we went from Science to Scientology! Boo!
Load More Replies...Dude, you play my singing monsters?!?!?!? THAT IS SO COOL I LOVE THAT GAME
Load More Replies...*smashes laptop* REEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE RID THE WORLD OF THIS MAKE IT STOP!!!!!!!!!!!!!!!!!1
I can't believe you didn't get an upvote! This is the EXACT thought I had when I saw this picture!!!
Load More Replies...Omg Peppy what have they done you’re gonna get kicked out of starfox
This is Putin. Without my glasses, with my bad eyesight I can tell who these are supposed to be. Its actually creepy and disturbing.
I knew it was him right away, too. I'm Russian so that's probably why :)
Load More Replies...Steve knows. Steve doesn't dare look. Steve knows he will not last long. Poor, poor steve.
Load More Replies...why is herobrine behind the man? it's gotta be herobrine, because the eyes are glowing
I see there might have been a small bias issue during the development faze.
Clearly... no data for black people was included in the AI.
Load More Replies...It’s legit the cat from the mirror mask movie like if you know what that is
Who's on the left? All I see is 19th century Josh Gad with loads of makeup.
how do u even get this app to work, i downloaded it, and have no idea how to run it
let face it - it's clickbait and the article fooled all of us (how else you would be here) the app simply finds the most similar face and possibly readjusts the colour gamut/colouring.
In my day we just held a magnifying lens in front of the screen.
Load More Replies...I wanted to see a pixelated pile of dog poo turned into the face/head of dRumpf...(don't really need AI for that transformation, I guess).
let face it - it's clickbait and the article fooled all of us (how else you would be here) the app simply finds the most similar face and possibly readjusts the colour gamut/colouring.
In my day we just held a magnifying lens in front of the screen.
Load More Replies...I wanted to see a pixelated pile of dog poo turned into the face/head of dRumpf...(don't really need AI for that transformation, I guess).
