x disparity determination in stereo vision
play

X Disparity Determination in Stereo Vision Lu Sang, Michael Haberl, - PowerPoint PPT Presentation

X Disparity Determination in Stereo Vision Lu Sang, Michael Haberl, Raphael Ullmann 22.07.2017 Lu Sang, Michael Haberl, Raphael Ullmann Overview Problem Description Algorithms Results Lu Sang, Michael Haberl, Raphael Ullmann Overview


  1. X Disparity Determination in Stereo Vision Lu Sang, Michael Haberl, Raphael Ullmann 22.07.2017 Lu Sang, Michael Haberl, Raphael Ullmann

  2. Overview Problem Description Algorithms Results Lu Sang, Michael Haberl, Raphael Ullmann

  3. Overview Problem Description Algorithms Results Lu Sang, Michael Haberl, Raphael Ullmann

  4. Problem Description Lu Sang, Michael Haberl, Raphael Ullmann

  5. Problem Description Lu Sang, Michael Haberl, Raphael Ullmann

  6. Problem Description • What is the distance to an object? • How to determine the distance by 2D images? 1https://upload.wikimedia.org/wikipedia/commons/4/49/Roboterhand.mit.Gluehbirne.png Lu Sang, Michael Haberl, Raphael Ullmann

  7. Problem Description • What is the distance to an object? • How to determine the distance by 2D images? Applications • Autonomous driving • Robotics • Object recognition 1 1https://upload.wikimedia.org/wikipedia/commons/4/49/Roboterhand.mit.Gluehbirne.png Lu Sang, Michael Haberl, Raphael Ullmann

  8. Stereo Cameras Lu Sang, Michael Haberl, Raphael Ullmann

  9. Corresponding Pixels Figure: Left Picture [1] Figure: Right Picture Lu Sang, Michael Haberl, Raphael Ullmann

  10. Corresponding Pixels Figure: Left Picture [1] Figure: Right Picture Lu Sang, Michael Haberl, Raphael Ullmann

  11. Corresponding Pixels Figure: Left Picture [1] Figure: Right Picture 1 Calculate for the pixels in the left image costs in the right image Lu Sang, Michael Haberl, Raphael Ullmann

  12. Corresponding Pixels Figure: Left Picture [1] Figure: Right Picture 1 Calculate for the pixels in the left image costs in the right image 2 Pixel with minimal cost is the corresponding pixel Lu Sang, Michael Haberl, Raphael Ullmann

  13. Disparity Lu Sang, Michael Haberl, Raphael Ullmann

  14. Disparity Disparity Pixel distance of related pixels. Lu Sang, Michael Haberl, Raphael Ullmann

  15. Getting distance from Disparity Lu Sang, Michael Haberl, Raphael Ullmann

  16. Getting distance from Disparity Lu Sang, Michael Haberl, Raphael Ullmann

  17. Getting distance from Disparity Lu Sang, Michael Haberl, Raphael Ullmann

  18. Getting distance from Disparity Lu Sang, Michael Haberl, Raphael Ullmann

  19. Distance z ✏ f ☎ b d • Distance z • Focal length of the camera f • Disparity d Lu Sang, Michael Haberl, Raphael Ullmann

  20. Overview Problem Description Algorithms Results Lu Sang, Michael Haberl, Raphael Ullmann

  21. Overview Pre-processing Cost Calculation Pyramid Scheme Force Local Consistency Penalty Terms Median Filter Lu Sang, Michael Haberl, Raphael Ullmann

  22. Overview Pre-processing Cost Calculation Pyramid Scheme Force Local Consistency Penalty Terms Median Filter Lu Sang, Michael Haberl, Raphael Ullmann

  23. Pre-processing: Undistortion Original Picture Undistorted Picture Lu Sang, Michael Haberl, Raphael Ullmann

  24. Pre-processing: Undistortion Original Picture Undistorted Picture Lu Sang, Michael Haberl, Raphael Ullmann

  25. Pre-processing: Rectification Lu Sang, Michael Haberl, Raphael Ullmann

  26. Pre-processing: Rectification Lu Sang, Michael Haberl, Raphael Ullmann

  27. Pre-processing: Rectification Lu Sang, Michael Haberl, Raphael Ullmann

  28. Pre-processing: Rectification Left Picture Right Picture Lu Sang, Michael Haberl, Raphael Ullmann

  29. Pre-processing: Rectification Left Picture Right Picture Lu Sang, Michael Haberl, Raphael Ullmann

  30. Modelling: Energy Function Energy Function argmin d C ♣ p , d q � S ♣ p , d q • Cost C ♣ p , d q for every pixel p and disparities d ✏ 1 , ..., D • Regularization S ♣ p , d q • E.g. penalty for deviation of neighbouring pixels Lu Sang, Michael Haberl, Raphael Ullmann

  31. Overview Pre-processing Cost Calculation Pyramid Scheme Force Local Consistency Penalty Terms Median Filter Lu Sang, Michael Haberl, Raphael Ullmann

  32. Cost Calculation: Comparing Windows Lu Sang, Michael Haberl, Raphael Ullmann

  33. Cost Calculation: Comparing Windows Lu Sang, Michael Haberl, Raphael Ullmann

  34. ❞ ✏ ñ ✏ ✁ ❞ ✏ ñ ✏ ✁ r✁ s Cost Calculation: Cross Correlation Lu Sang, Michael Haberl, Raphael Ullmann

  35. ñ ✏ ✁ ❞ ✏ ñ ✏ ✁ r✁ s Cost Calculation: Cross Correlation ❞ ✏ Lu Sang, Michael Haberl, Raphael Ullmann

  36. ñ ✏ ✁ ❞ ✏ ñ ✏ ✁ r✁ s Cost Calculation: Cross Correlation ❞ ✏ Lu Sang, Michael Haberl, Raphael Ullmann

  37. ❞ ✏ ñ ✏ ✁ r✁ s Cost Calculation: Cross Correlation ❞ ✏ Total sum = 2 ñ Costs ✏ ✁ 2 Lu Sang, Michael Haberl, Raphael Ullmann

  38. ❞ ✏ ñ ✏ ✁ r✁ s Cost Calculation: Cross Correlation ❞ ✏ Total sum = 2 ñ Costs ✏ ✁ 2 Lu Sang, Michael Haberl, Raphael Ullmann

  39. ñ ✏ ✁ r✁ s Cost Calculation: Cross Correlation ❞ ✏ Total sum = 2 ñ Costs ✏ ✁ 2 ❞ ✏ Lu Sang, Michael Haberl, Raphael Ullmann

  40. r✁ s Cost Calculation: Cross Correlation ❞ ✏ Total sum = 2 ñ Costs ✏ ✁ 2 ❞ ✏ Total sum = 4 ñ Costs ✏ ✁ 4 Lu Sang, Michael Haberl, Raphael Ullmann

  41. Cost Calculation: Cross Correlation ❞ ✏ Total sum = 2 ñ Costs ✏ ✁ 2 ❞ ✏ Total sum = 4 ñ Costs ✏ ✁ 4 Normalization and Zero Mean: Values in r✁ 1 , 1 s Lu Sang, Michael Haberl, Raphael Ullmann

  42. Cost Calculation: Result Lu Sang, Michael Haberl, Raphael Ullmann

  43. Cost Calculation: Result Lu Sang, Michael Haberl, Raphael Ullmann

  44. Cost Calculation: Result • Error Rate of NCC: 33 . 09% Lu Sang, Michael Haberl, Raphael Ullmann

  45. Test Data Figure: Left Image Figure: Right Image Lu Sang, Michael Haberl, Raphael Ullmann

  46. Test Data • Middlebury Dataset • Ground Truth • Leaderboard Lu Sang, Michael Haberl, Raphael Ullmann

  47. Algorithm Defect Lu Sang, Michael Haberl, Raphael Ullmann

  48. Overview Pre-processing Cost Calculation Pyramid Scheme Force Local Consistency Penalty Terms Median Filter Lu Sang, Michael Haberl, Raphael Ullmann

  49. Pyramid Scheme Lu Sang, Michael Haberl, Raphael Ullmann

  50. Pyramid Scheme Lu Sang, Michael Haberl, Raphael Ullmann

  51. Pyramid Scheme Lu Sang, Michael Haberl, Raphael Ullmann

  52. Pyramid Scheme: Results Figure: Results of NCC Figure: Results of Pyramid Scheme • Error Rate of NCC: 33 . 09% • Error Rate of Pyramid Scheme: 28 . 1% Lu Sang, Michael Haberl, Raphael Ullmann

  53. Overview Pre-processing Cost Calculation Pyramid Scheme Force Local Consistency Penalty Terms Median Filter Lu Sang, Michael Haberl, Raphael Ullmann

  54. Force Local Consistency • Interpret Cross Correlation as confidence indicator • Use only pixels with high confidence • Replace low confidence by values with high confidence in the window Lu Sang, Michael Haberl, Raphael Ullmann

  55. Force Local Consistency: Confidence Map Figure: Results of NCC Figure: Confidence Map • Black point: trustworthy pixel with correct disparity ( C ♣ p q ➙ T ). • Red point: unreliable pixel with violated disparity ( C ♣ p q ➔ T ). Lu Sang, Michael Haberl, Raphael Ullmann

  56. Force Local Consistency Original C ♣ p ✶ q P W Disparity all violated Violated Unique p ✶ Disparity unique ➔ T max p ✶ P W C ♣ p ✶ q C ♣ p q C ♣ p ✶ q not unique ➙ T Correct Closest p ✶ Disparity 1 Mark all violated pixels. Lu Sang, Michael Haberl, Raphael Ullmann

  57. Force Local Consistency Original C ♣ p ✶ q P W Disparity all violated Violated Unique p ✶ Disparity unique ➔ T max p ✶ P W C ♣ p ✶ q C ♣ p q C ♣ p ✶ q not unique ➙ T Correct Closest p ✶ Disparity 1 Mark all violated pixels. 2 Find the pixel with max NCC coefficient. Lu Sang, Michael Haberl, Raphael Ullmann

  58. Force Local Consistency Original C ♣ p ✶ q P W Disparity all violated Violated Unique p ✶ Disparity unique ➔ T max p ✶ P W C ♣ p ✶ q C ♣ p q C ♣ p ✶ q not unique ➙ T Correct Closest p ✶ Disparity 1 Mark all violated pixels. 2 Find the pixel with max NCC coefficient. 3 Replace the disparity value by using the disparity of new pixel. Lu Sang, Michael Haberl, Raphael Ullmann

  59. Force Local Consistency: Results Figure: Results of NCC Figure: Results of FLC • Error Rate of NCC: 33 . 09% • Error Rate of FLC: 26 . 90% Lu Sang, Michael Haberl, Raphael Ullmann

  60. Overview Pre-processing Cost Calculation Pyramid Scheme Force Local Consistency Penalty Terms Median Filter Lu Sang, Michael Haberl, Raphael Ullmann

  61. Local Penalty Local Penalty 1 Smoothness check 2 Search four directions 3 Punish on NCC coefficient 4 Iterate Lu Sang, Michael Haberl, Raphael Ullmann

  62. Overview Pre-processing Cost Calculation Pyramid Scheme Force Local Consistency Penalty Terms Median Filter Lu Sang, Michael Haberl, Raphael Ullmann

  63. Post Processing: Median Filter ñ Lu Sang, Michael Haberl, Raphael Ullmann

  64. Final Results Figure: Results of FLC Figure: After Post-processing • Error Rate of FLC: 26 . 90% • Error Rate of FLC + Median filter: 21 . 10% Lu Sang, Michael Haberl, Raphael Ullmann

  65. Overview Problem Description Algorithms Results Lu Sang, Michael Haberl, Raphael Ullmann

Recommend


More recommend