1. Structured light and active ranging techniques Class 11

2. per-pixel optimization per-scanline optimization full image optimization last Tuesday: stereo

3. polar rectification planar rectification original image pair

4. Plane-sweep multi-view matching Simple algorithm for multiple cameras No rectification necessary, but also no gain Doesn’t deal with occlusions Collins’96; Roy and Cox’98 (GC); Yang et al.’02/’03 (GPU)

5. Today’s class unstructured light structured light time-of-flight (some slides from Szymon Rusinkiewicz, Brian Curless)

6. A Taxonomy

7. A taxonomy

8. Unstructured light project texture to disambiguate stereo

9. Space-time stereo Davis, Ramamoothi, Rusinkiewicz, CVPR’03

10. Space-time stereo Davis, Ramamoothi, Rusinkiewicz, CVPR’03

11. Space-time stereo Zhang, Curless and Seitz, CVPR’03

12. Space-time stereo results Zhang, Curless and Seitz, CVPR’03

13. Triangulation

14. Triangulation: Moving the Camera and Illumination Moving independently leads to problems with focus, resolution Most scanners mount camera and light source rigidly, move them as a unit

15. Triangulation: Moving the Camera and Illumination

16. (Rioux et al. 87)

17. Triangulation: Extending to 3D Possibility #1: add another mirror (flying spot) Possibility #2: project a stripe, not a dot Object Laser CameraCamera

18. Triangulation Scanner Issues Accuracy proportional to working volume (typical is ~1000:1) Scales down to small working volume (e.g. 5 cm. working volume, 50  m. accuracy) Does not scale up (baseline too large…) Two-line-of-sight problem (shadowing from either camera or laser) Triangulation angle: non-uniform resolution if too small, shadowing if too big (useful range: 15  -30  )

19. Triangulation Scanner Issues Material properties (dark, specular) Subsurface scattering Laser speckle Edge curl Texture embossing

21. Space-time analysis Curless ‘95

22. Space-time analysis Curless ‘95

23. Projector as camera

24. Multi-Stripe Triangulation To go faster, project multiple stripes But which stripe is which? Answer #1: assume surface continuity e.g. Eyetronics’ ShapeCam

25. Real-time system Koninckx and Van Gool

26. Multi-Stripe Triangulation To go faster, project multiple stripes But which stripe is which? Answer #2: colored stripes (or dots)

27. Multi-Stripe Triangulation To go faster, project multiple stripes But which stripe is which? Answer #3: time-coded stripes

28. Time-Coded Light Patterns Assign each stripe a unique illumination code over time [Posdamer 82] Space Time

29. An idea for a project? Bouget and Perona, ICCV’98

30. Pulsed Time of Flight Basic idea: send out pulse of light (usually laser), time how long it takes to return

31. Pulsed Time of Flight Advantages: Large working volume (up to 100 m.) Disadvantages: Not-so-great accuracy (at best ~5 mm.) Requires getting timing to ~30 picoseconds Does not scale with working volume Often used for scanning buildings, rooms, archeological sites, etc.

32. Depth cameras 2D array of time-of-flight sensors e.g. Canesta’s CMOS 3D sensor jitter too big on single measurement, but averages out on many (10,000 measurements  100x improvement)

33. Depth cameras Superfast shutter + standard CCD cut light off while pulse is coming back, then I~Z but I~albedo (use unshuttered reference view) 3DV’s Z-cam

34. AM Modulation Time of Flight Modulate a laser at frequency m, it returns with a phase shift  Note the ambiguity in the measured phase!  Range ambiguity of 1 / 2 m n

35. AM Modulation Time of Flight Accuracy / working volume tradeoff (e.g., noise ~ 1 / 500 working volume) In practice, often used for room-sized environments (cheaper, more accurate than pulsed time of flight)

36. Shadow Moire

37. Depth from focus/defocus Nayar’95 Nov. 8, don’t miss Distinguished lecture!

38. Next class: structure from motion