Can We Detect Neural Image Generators?

📝 The paper "CNN-generated images are surprisingly easy to spot...for now" is available here:

Our Discord server is now available here and you are all invited!

🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible:
Alex Haro, Alex Paden, Andrew Melnychuk, Angelos Evripiotis, Anthony Vdovitchenko, Benji Rabhan, Brian Gilman, Bryan Learn, Daniel Hasegan, Dan Kennedy, Dennis Abts, Eric Haddad, Eric Martel, Evan Breznyik, Geronimo Moralez, James Watt, Javier Bustamante, Kaiesh Vohra, Kasia Hayden, Kjartan Olason, Levente Szabo, Lorin Atzberger, Lukas Biewald, Marcin Dukaczewski, Marten Rauschenberg, Maurits van Mastrigt, Michael Albrecht, Michael Jensen, Nader Shakerin, Owen Campbell-Moore, Owen Skarpness, Raul Araújo da Silva, Rob Rowe, Robin Graham, Ryan Monsurate, Shawn Azman, Steef, Steve Messina, Sunil Kim, Taras Bobrovytsky, Thomas Krcmar, Torsten Reil, Tybie Fitzhugh.

Károly Zsolnai-Fehér's links:

#DeepFake #DeepFakes

[quantuminfinity]

This is a very impressive paper, thank you for covering it! I am very impressed by how well it does with all of the different Gans. I expected that it would have some troubles with of the other ones, but this is very cool!

[Dedjkeorrn]

I'm still not used to the doctor.

[carltheshivan]

I'm glad you put in an icon that appears when it's time to hold on to our papers so we know when to hold on to our papers.

[peetiegonzalez]

"Doctor" Congrats, man.

[bright-noise]

Can't the detector just be used as part of a new GAN that obscures/fixes the tells for a generated image?

[Qubot]

Using an AI to detect AI generated images, and after we will create an AI who detect if an AI can detect an AI generated image...

[beskamir]

Next month's paper: "GAN's have been improved using last month's detection network until that detection no longer works"

[essigautomat]

And now the AIs are trained against this fake detector

[qwerty_and_azerty]

Next month: a paper training a neural image generator adversarially against this new detector, rendering it useless

[saurabhsswami]

Thank you Dr. Fehér for your amazing assessment of these papers. Very interesting work!

[kebakent]

I don't understand why this technique cannot simply be included in the GAN training phase, and optimized for. Essentially, it should be a game of ad vs. adblocker, or a never-ending tug of war if you will.

[Dedjkeorrn]

Something like this should be added to search engines and social media to prevent the spread of misinformation.

[scruff]

Yes - as others have said - what about using a GAN with this detector? Surely it will be possible to create generated images that are extremely difficult to distinguish from real images.

[DammerGuy]

did I hear... Doctor? Congratulations!!

[TheJaviJB]

Nice, now we can use that detector as adversarial network to create even better images with the generative one

[bugagometr]

You are a Doctor! Wow! Congrats bro)))

[hallowsseeker]

Lovely to hear the phrase, " Dear Fellow Scholars, This is Two Minute Papers with Dr. Károly Zsolnai-Fehér"

[PatanxD]

If you generated a fake image then displayed it on a monitor and took a picture of the monitor displaying the image and you gave this algorithm the image from the camera, would this technique be able to spot that it was a picture of a fake generated image in that case? I doubt it.

[Jone]

The deep fake detector seems like a good loss function for a deep fake generator

[Sakatland]

What if you train a neural network to learn from what the one used to detect a image generated by another neural network to finally correct the original neural network so that the one which detects real than generated image fails?