Let's talk about Fakes

Transcript

1. Let's talk about Fakes
   Let's talk about all the good things
   
   And the bad things
   
   that may be
   

   View Slide

2. Making Faces
   • Face Synthesis

   from nothing to face
   
   • Facial Attributes 

   modify age, gender, add glasses change hairstyle or color
   
   • Face Swap

   use the body and background of target, face of source
   
   • Facial Expression Transfer

   transfer only the expression or modify expression to
   match an impersonator or an audio track (reenactment)
   

   View Slide

3. View Slide

4. Face Synthesis
   https://research.nvidia.com/publication/2017-10_Progressive-Growing-of
   

   View Slide

5. https://thispersondoesnotexist.com/
   

   View Slide

6. Facial Attributes
   

   View Slide

7. https://www.faceapp.com/
   

   View Slide

8. https://www.faceapp.com/
   

   View Slide

9. Face Off (1997)
   

   View Slide

10. Nicolas Cage Everywhere
    

    View Slide

11. Reenactment
    Cafe 80s, Hill Valley, USA, (October 21, 2015)
    

    View Slide

12. Jordan Peele (2018)
    

    View Slide

13. Face2Face (2016)
    

    View Slide

14. RGB-D tracking
    

    View Slide

15. View Slide

16. View Slide

17. FaceSwapGAN (2018)
    

    View Slide

18. View Slide

19. View Slide

20. View Slide

21. View Slide

22. View Slide

23. View Slide

24. View Slide

25. View Slide

26. FaceShifter (Dec 2019)
    

    View Slide

27. View Slide

28. View Slide

29. View Slide

30. How to Spot a Fake
    

    View Slide

31. https://www.kaggle.com/c/deepfake-detection-challenge
    

    View Slide

32. uneven
    skin
    tone
    artifacts
    in hair
    says
    strange
    things
    warped
    edges
    How to Spot a Fake
    obstruction
    glitches
    

    View Slide

33. That was a real photo.
    

    View Slide

34. uneven
    skin
    tone
    artifacts
    in hair
    says
    strange
    things
    warped
    edges
    How to Spot a Fake
    obstruction
    glitches
    

    View Slide

35. https://people.csail.mit.edu/mrub/vidmag/
    

    View Slide

36. Why this is important
    

    View Slide

37. We are hardwired to
    react to faces.
    

    View Slide

38. View Slide

39. Especially to moving faces.
    

    Even more so, emotional.
    

    View Slide

40. View Slide

41. We have a tendency to fall for
    and spread shocking images.
    

    View Slide

42. View Slide

43. View Slide

44. We react to images
    that speak to our tribal
    instincts or align with
    our political world view
    

    View Slide

45. View Slide

46. View Slide

47. No matter how we try, how
    many tools we create to detect
    fakes, we WILL BE AFFECTED...
    

    View Slide

48. ...the question is:
    HOW?
    

    View Slide

49. View Slide