161–170 of 285 posts
Slightly more robust method... Ask the caller to move out of the frame and then back in again. You will see a noticable 'step' as the face that is partially in the frame suddenly gets detected as a face and the deepfake is applied. The only way around this is to crop the input video quite heavily - by at least one face diameter, which is a lot if the user is near the camera.
"profile view challenge" coming in 3, 2, 1 ...
"Hey! To make sure you stay secure, we require a short video. Please look straight into the camera and tap the screen." "You look great! We just need you to blink 5 times, and you're almost done!" "Almost done! Just show us your best side and turn your head to the left like shown above." "Of course, you only have best sides. Just turn your head to the right like displayed above, and we can continue." "You've almost g…
Is there a "client side" way to detect this? Similar to how we can detect photoshopped still images: checking edges, shadows, pixels, etc... The benefit is you would not have to rely on issuing commands to the remote party.
Hybrid Video/audio/semantic Captcha, perhaps? An audio prompt like 'Using your hand, repeat the numbers that I am signaling. Use set of fingers from what I am using'.
This is why I got really upset when my employer said the swimsuit competition segment of the interview was past its time. Its time is now!
Signed up for one of those 'neobanks' (that don't have physical branches) and part of the signup required me to turn my head sideways. I wondered why they wanted me to do that. Now I know.
www.hownormalami.eu
Was a fun project, but the cat-and-mouse feeling was inescapable. For those curious, look up the DARPA MediFor project. Siwei Lyu (in the article) did a bunch of work in this space. Also see Hany Farid and Shruti Agarwal. They've worked specifically with deep fake detection.
I do appreciate everyone on this site contributing to my knowledge of infosec. I don't work directly in the space, but I feel the contributions on this site help educate those of us not directly working the profession.
Source: I work in the field. This is a current limitation, and an artifact of the data+method but not something that should be relied upon. If we do some adversary modelling, we can find two ways to work around this: 1) actively generate and search for such data; perhaps expensive for small actors but not well equipped malicious ones. 2) wait for deep learning to catch up, e.g. by extending NERFs (neural radiance fie…
> This is a current limitation The thing with any AI/ML tech is that current limitations are always underplayed by proponents. Self-driving cars will come out next year, every year. I'd say that until the tech actually exists, this is a great way to detect live deepfakes. Not using the technique just because maybe sometime in the future it won't work isn't very sound. For an extreme opponent you may need additional s…
"Come out" could mean different things in different contexts. Deepfake defence context is analogous to something like: there are cars on public roads with no driver at the wheel. And this is already true in multiple places in the world.
Source: I work in the field. This is a current limitation, and an artifact of the data+method but not something that should be relied upon. If we do some adversary modelling, we can find two ways to work around this: 1) actively generate and search for such data; perhaps expensive for small actors but not well equipped malicious ones. 2) wait for deep learning to catch up, e.g. by extending NERFs (neural radiance fie…
> This is a current limitation The thing with any AI/ML tech is that current limitations are always underplayed by proponents. Self-driving cars will come out next year, every year. I'd say that until the tech actually exists, this is a great way to detect live deepfakes. Not using the technique just because maybe sometime in the future it won't work isn't very sound. For an extreme opponent you may need additional s…