full transcript
From the Ted Talk "Supasorn Suwajanakorn: Fake videos of real people -- and how to spot them"

Unscramble the Blue Letters

It turns out this problem is very ngnigllhcae, but the key trick is that we are going to analyze a large hpoot collection of the person beforehand. For George W. buhs, we can just search on Google, and from that, we are able to build an average model, an iterative, ieendfr model to recover the expression in fnie details, like creases and wrinkles. What's fascinating about this is that the photo collection can come from your typical photos. It doesn't really matter what expression you're making or where you took those photos. What matters is that there are a lot of them. And we are still missing color here, so next, we develop a new blending technique that improves upon a single averaging method and rpdeocus phrsa facial textures and colors. And this can be done for any nerxispeos.

Open Cloze

It turns out this problem is very ___________, but the key trick is that we are going to analyze a large _____ collection of the person beforehand. For George W. ____, we can just search on Google, and from that, we are able to build an average model, an iterative, _______ model to recover the expression in ____ details, like creases and wrinkles. What's fascinating about this is that the photo collection can come from your typical photos. It doesn't really matter what expression you're making or where you took those photos. What matters is that there are a lot of them. And we are still missing color here, so next, we develop a new blending technique that improves upon a single averaging method and ________ _____ facial textures and colors. And this can be done for any __________.

Solution

  1. challenging
  2. produces
  3. fine
  4. bush
  5. photo
  6. sharp
  7. refined
  8. expression

Original Text

It turns out this problem is very challenging, but the key trick is that we are going to analyze a large photo collection of the person beforehand. For George W. Bush, we can just search on Google, and from that, we are able to build an average model, an iterative, refined model to recover the expression in fine details, like creases and wrinkles. What's fascinating about this is that the photo collection can come from your typical photos. It doesn't really matter what expression you're making or where you took those photos. What matters is that there are a lot of them. And we are still missing color here, so next, we develop a new blending technique that improves upon a single averaging method and produces sharp facial textures and colors. And this can be done for any expression.

ngrams of length 2

collocation frequency
video bo 3

Important Words

  1. analyze
  2. average
  3. averaging
  4. blending
  5. build
  6. bush
  7. challenging
  8. collection
  9. color
  10. colors
  11. creases
  12. details
  13. develop
  14. expression
  15. facial
  16. fascinating
  17. fine
  18. george
  19. google
  20. improves
  21. iterative
  22. key
  23. large
  24. lot
  25. making
  26. matter
  27. matters
  28. method
  29. missing
  30. model
  31. person
  32. photo
  33. photos
  34. problem
  35. produces
  36. recover
  37. refined
  38. search
  39. sharp
  40. single
  41. technique
  42. textures
  43. trick
  44. turns
  45. typical
  46. wrinkles