Slashdot Mirror
Facebook To Fight Revenge Porn by Letting Potential Victims Upload Nudes in Advance (bleepingcomputer.com)
Catalin Cimpanu, writing for BleepingComputer: Facebook is testing new technology that is designed to help victims of revenge porn acts. It works on a database of file hashes, a cryptographic signature computed for each file. Facebook says that once an abuser tries to upload an image marked as "revenge porn" in its database, its system will block the upload process. This will work for images shared on the main Facebook service, but also for images shared privately via Messenger, Facebook's IM app. The weird thing is that in order to build a database of "revenge porn" file hashes, Facebook will rely on potential victims uploading a copy of the nude photo in advance. This process involves the victim sending a copy of the nude photo to his own account, via Facebook Messenger. This implies uploading a copy of the nude photo on Facebook Messenger, the very same act the victim is trying to prevent. The victim can then report the photo to Facebook, which will create a hash of the image that the social network will use to block further uploads of the same photo.
I already have a service that handles this, just send me the pic and I'll handle it....
Cheap storage VM.
I know they "claim" they will not keep the pictures, but only a hash of the image. But do you really trust Facebook that much?
First law of people: People are generally stupid.
What's wrong with putting all the nudes of every person on facebook on a database ?
What could go Equifax?
~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
The public reaction to this is understandably somewhat muted and off-put. Why upload nude photos to Facebook, indeed? The claim is that they will compute a hash of the image, and store that to prevent future uploads.
If that is really the case, when why not compute the hash locally on the user's machine, and upload only the hash? Surely that can be done on essentially all modern hardware from cell phone to desktop in a reasonable amount of time.
Put my fist through my alarm clock with its ding-dong death inside my ear. - The Blackjacks.
If you don't want your nudes to end up on the internet, don't send them to other people.
The prudishness endemic in American society is not universal. Thankfully, I live in a society that does not tie itself in hypocritical knots because a nipple was shown on TV.
Democracy is a sheep and two wolves deciding what to have for lunch. Freedom is a well armed sheep contesting the issue
Comment removed based on user account deletion
There are much better methods of hashing images than stupidly taking a file checksum, such as this one here:
https://pippy360.github.io/tra...
This algorithm here does not care about affine transformations applied to the image, so it can be scaled, rotated, skewed, and still be a match.
There are image similarity algorithms out there that do not care about the absolute hash of the file, and can detect the same image cropped, scaled, or rotated just fine.
Here is one such algorithm:
https://pippy360.github.io/tra...
First off, is there really a problem with revenge porn on facebook and if there is, it would seem that the easiest solution for
facebook is to block all porn. I've never seen nudes on facebook. I always assumed that it would be against facebook policy
as facebook is mostly a PG-13 kindof place.
Second, I would think that facial recognition would be the correct solution. Let someone upload a picture of their face and
facebook can make sure that that particular face doesn't appear in nudes. An unidentified nude without a face even if someone
says "this is so-in-so" is pretty harmless as if you can't see the face you could pretty much say it is anyone.
Lastly, google just came out with facial recognition for dogs so presumably you could also use that same technology for
tattoos, or specific body parts too.
But again, I would think revenge porn would be primarily a problem on other services not facebook.
If nobody looks at the image, or, as some have suggested, the hash is computed client side (so nobody would be able to look at the image) it would be ripe for abuse.
There is a very easy fix for this - the first time the hash matches the takedown requires human approval. This way someone only looks at the image if the image is already uploaded for people to look at and you can't abuse the system by filing takedowns for random pictures. This would even reduce Facebook's work because instead of checking every upload they only have to check ones which match.
Like audio fingerprinting this isn't a cryptographically accurate hash but more like a song fingerprint in that it only takes a small segment of the image for recognition.
Of course the real abuse would be to take a picture of the Eiffel tower and watch everyone's uploads fail for a few hours.
Velociraptor = Distiraptor / Timeraptor
/.!
That signature may be the funniest thing I've ever read on
I'm assuming, of course, that Distiraptor is a vector and Timeraptor is a scalar.
Circle the wagons and fire inward. Entropy increases without bounds.
The first step of hashing, per the linked article, involves finding keypoints in the image that are still detected as keypoints even in an affine transformed copy of an image. How is this done? Does it involve scale-invariant feature transform (SIFT) or some other feature detection means subject to a United States patent?
If Facebook is more trustworthy than your lovers then you've really made some poor choices in your life.