full transcript

From the Ted Talk by Doug Roble: Digital humans that look just like us

Unscramble the Blue Letters

You can see every pore, every wrinkle. But we had to have that. Reality is all about datiel. And without it, you miss it. We are far from done, though. This let us build a model of my face that lokeod like me. But it didn't really move like me. And that's where mhinace learning comes in. And machine learning needs a ton of data. So I sat down in front of some high-resolution motion-capturing device. And also, we did this traditional mitoon capture with markers. We created a whole bucnh of images of my face and mionvg point clouds that represented that shapes of my face. Man, I made a lot of expressions, I said different lines in different emotional states ... We had to do a lot of capture with this. Once we had this enormous amount of data, we built and trained deep neuarl networks. And when we were finished with that, in 16 milliseconds, the neural ntrewok can look at my image and figure out everything about my face. It can cutompe my expression, my wrinkles, my blood flow — even how my eyelashes move. This is then rendered and depliysad up there with all the detail that we captured previously.

Open Cloze

You can see every pore, every wrinkle. But we had to have that. Reality is all about ______. And without it, you miss it. We are far from done, though. This let us build a model of my face that ______ like me. But it didn't really move like me. And that's where _______ learning comes in. And machine learning needs a ton of data. So I sat down in front of some high-resolution motion-capturing device. And also, we did this traditional ______ capture with markers. We created a whole _____ of images of my face and ______ point clouds that represented that shapes of my face. Man, I made a lot of expressions, I said different lines in different emotional states ... We had to do a lot of capture with this. Once we had this enormous amount of data, we built and trained deep ______ networks. And when we were finished with that, in 16 milliseconds, the neural _______ can look at my image and figure out everything about my face. It can _______ my expression, my wrinkles, my blood flow — even how my eyelashes move. This is then rendered and _________ up there with all the detail that we captured previously.

Solution

  1. bunch
  2. neural
  3. network
  4. moving
  5. compute
  6. displayed
  7. looked
  8. machine
  9. detail
  10. motion

Original Text

You can see every pore, every wrinkle. But we had to have that. Reality is all about detail. And without it, you miss it. We are far from done, though. This let us build a model of my face that looked like me. But it didn't really move like me. And that's where machine learning comes in. And machine learning needs a ton of data. So I sat down in front of some high-resolution motion-capturing device. And also, we did this traditional motion capture with markers. We created a whole bunch of images of my face and moving point clouds that represented that shapes of my face. Man, I made a lot of expressions, I said different lines in different emotional states ... We had to do a lot of capture with this. Once we had this enormous amount of data, we built and trained deep neural networks. And when we were finished with that, in 16 milliseconds, the neural network can look at my image and figure out everything about my face. It can compute my expression, my wrinkles, my blood flow — even how my eyelashes move. This is then rendered and displayed up there with all the detail that we captured previously.

Frequently Occurring Word Combinations

ngrams of length 2

collocation frequency
real person 4
machine learning 3
digital human 2
motion capture 2
real time 2
visual effects 2
digital humans 2
enormous amount 2
blood flow 2

Important Words

  1. amount
  2. blood
  3. build
  4. built
  5. bunch
  6. capture
  7. captured
  8. clouds
  9. compute
  10. created
  11. data
  12. deep
  13. detail
  14. device
  15. displayed
  16. emotional
  17. enormous
  18. expression
  19. expressions
  20. eyelashes
  21. face
  22. figure
  23. finished
  24. flow
  25. front
  26. image
  27. images
  28. learning
  29. lines
  30. looked
  31. lot
  32. machine
  33. man
  34. markers
  35. milliseconds
  36. model
  37. motion
  38. move
  39. moving
  40. network
  41. networks
  42. neural
  43. point
  44. pore
  45. previously
  46. reality
  47. rendered
  48. represented
  49. sat
  50. shapes
  51. states
  52. ton
  53. traditional
  54. trained
  55. wrinkle
  56. wrinkles