Watertight Planar Surface Meshing of Indoor Point-Clouds with Voxel Carving

1. EECS 3DV - June 29, 2013 Eric Turner Avideh Zakhor Watertight Planar Surface Meshing of Indoor Point-Clouds with Voxel Carving University of California, Berkeley

2. EECS Surface Meshing of Buildings University of California, Berkeley

3. EECS Surface Meshing of Buildings University of California, Berkeley • Indoor Navigation • Energy Simulation • Texture Mapping • Virtual Reality • Augmented Reality • Video Games

4. EECS Indoor Modeling University of California, Berkeley • Acquisition System

5. EECS Indoor Modeling University of California, Berkeley • Point-cloud Generation

6. EECS Example Point-cloud University of California, Berkeley

7. EECS Example Point-cloud University of California, Berkeley Example Mesh

8. EECS Motivation University of California, Berkeley • Why is meshing useful? • Why do we want it to be planar? • Why do we want it to be watertight?

9. EECS Motivation University of California, Berkeley • Why is meshing useful? Example Point-cloud • 45 Million Pts • 3.5 GB on disk

10. EECS Motivation University of California, Berkeley • Why is meshing useful? Example Mesh • 985,000 Tris • 20 MB on disk

11. EECS Approach University of California, Berkeley Point Cloud Path Voxels labeled inside / outside Voxel Carving Region Growing Planar geometry fit to voxel boundary Triangulate Regions Watertight Mesh Adaptively sized triangles

12. EECS Voxel Carving University of California, Berkeley LiDAR scans Path of Scanner

13. EECS Voxel Carving University of California, Berkeley • Trace path of laser through space LiDAR scans Path of Scanner

14. EECS Voxel Carving University of California, Berkeley • Interpolate neighboring scans to define volume Ray-tracing

15. EECS Voxel Carving University of California, Berkeley • Define volume with voxels Carved Voxels

16. EECS Voxel Data Structure University of California, Berkeley • Only boundary voxels explicitly stored • Carving preserves watertightness of volume

17. EECS University of California, Berkeley

18. EECS Plane Fitting University of California, Berkeley • Goal: remove discretization artifacts on surface

19. EECS Plane Fitting University of California, Berkeley • Goal: remove discretization artifacts on surface • Combine voxel faces into planar regions

20. EECS Plane Fitting University of California, Berkeley • Goal: remove discretization artifacts on surface • Combine voxel faces into planar regions Initialize via flood-fill

21. EECS Plane Fitting University of California, Berkeley • Merge regions by computing best-fit plane

22. EECS Plane Fitting University of California, Berkeley • Final merged regions

23. EECS Plane Fitting University of California, Berkeley • Final merged regions • Identified locations of planar regions • Now need to mesh

24. EECS Triangulation of Regions University of California, Berkeley • Example planar region

25. EECS Triangulation of Regions University of California, Berkeley • Example planar region • Dominant axis-aligned plane

26. EECS Triangulation of Regions University of California, Berkeley • Triangulate 2D projection

27. EECS Triangulation of Regions University of California, Berkeley • Example

28. EECS Triangulation of Regions University of California, Berkeley • Boundaries snapped to plane intersections

29. EECS Results University of California, Berkeley

30. EECS Results University of California, Berkeley

31. EECS University of California, Berkeley Constructed mesh of hallway Texture-mapped mesh of hallway

32. EECS University of California, Berkeley

33. EECS University of California, Berkeley

34. EECS University of California, Berkeley

35. EECS University of California, Berkeley

36. EECS University of California, Berkeley

37. EECS University of California, Berkeley Mesh, colored by region Close up of hotel hallway Viewing triangulation and planar regions Point-cloud

38. EECS Large Retail Shopping Center University of California, Berkeley 2.7 million triangles from 220 million points 112 m x 78 m

39. EECS Thank You University of California, Berkeley

40. EECS Supplemental University of California, Berkeley

41. EECS Model Statistics University of California, Berkeley

42. EECS Preserving Fine Detail in Voxels University of California, Berkeley Scan points Path of scanner

43. EECS Preserving Fine Detail in Voxels University of California, Berkeley Scan points Lasers Path of scanner

44. EECS Preserving Fine Detail in Voxels University of California, Berkeley Exterior Space Interior Space

45. EECS Preserving Fine Detail in Voxels University of California, Berkeley Scan points Path of scanner Lasers stopped at any occupied voxel

46. EECS Preserving Fine Detail in Voxels University of California, Berkeley Exterior Space Interior Space

47. EECS Preserving Fine Detail in Voxels University of California, Berkeley Preserving fine detail

48. EECS Preserving Fine Detail in Voxels University of California, Berkeley Without preserving fine detail