3D Reconstruction & Scanning
Transcript of 3D Reconstruction & Scanning
3D Reconstruction & Scanning
http://graphics.stanford.edu/projects/mich/
The Digital Michelangelo Project 1997~9
Agenda
3/15/2016 [email protected] 2
Multiple View Geometry
Depth Camera
Structured Light
Camera Calibration
Projector Calibration
http://www.ten24.info/
http://www.youtube.com/watch?v=j4waCVRgBWI
http://www.youtube.com/watch?v=Wgp0Cg3UFA4
Multiple View Geometry
in Computer Vision
2015.1
http://www.cs.unc.edu/~marc/
Multiple View Geometry in Computer Vision Second Edition
Richard Hartley and Andrew Zisserman, Cambridge University Press, March 2004.
http://users.cecs.anu.edu.au/~hartley/ http://www.robots.ox.ac.uk/~vgg/index.html
3/15/2016 [email protected] 4
M. Pollefeys, L. Van Gool, M. Vergauwen, F. Verbiest, K. Cornelis, J. Tops, R. Koch, Visual modeling with a hand-held camera, International Journal of
Computer Vision 59(3), 207-232, 2004. https://www.youtube.com/watch?feature=player_embedded&v=2mvzHvPYX0k
3/15/2016 [email protected] 5
3/15/2016 [email protected] 6
3/15/2016 [email protected] 7
Visual 3D Modeling from Images
3/15/2016 [email protected] 8
3/15/2016 [email protected] 9
Projective Transformations
Camera Calibration
Epipolar Geometry
Feature Points
Correspondence Search
RANSAC Algorithm
3D Reconstruction
SIFT&ASIFT
Photo Tourism
3/15/2016 [email protected] 10
http://phototour.cs.washington.edu/
https://www.youtube.com/watch?v=IgBQCoEfiMs
Building Rome on a cloudless day
3/15/2016 [email protected] 11
Building Rome on a cloudless day, Jan-Michael Frahm, Pierre Georgel, David Gallup, Tim Johnson, Rahul Raguram, Changchang Wu, Yi-Hung Jen,
Enrique Dunn, Brian Clipp, Svetlana Lazebnik, Marc Pollefeys, ECCV 2010 https://www.youtube.com/watch?v=4cEQZreQ2zQ
Photosynth
3/15/2016 [email protected] 12
http://photosynth.net/
https://www.youtube.com/watch?v=s-DqZ8jAmv0
Scale-invariant feature transform (or
SIFT) is an algorithm in computer vision to
detect and describe local features in images.
The algorithm was published by David Lowe in
1999.
Applications include object recognition, robotic
mapping and navigation, image stitching, 3D
modeling, gesture recognition, video tracking,
and match moving.
The algorithm is patented in the US; the owner
is the University of British Columbia.
….
David Lowe
Computer Science Department
University of British Columbia
2016-03-15 13
Scale Invariant Feature Transform
Depth Camera
Motion sensing input device by Microsoft
Depth camera tech. developed by PrimeSense Invented in 2005
Software tech. developed by Rare
First announced at E3 2009 as “Project Natal”
Windows SDK Releases
15
Microsoft Kinect
http://www.microsoft.com/en-us/kinectforwindows/discover/features.aspx
[email protected] 3/15/2016
3/15/2016 [email protected] 16
ToF 3D Camera
3/15/2016 [email protected] 17
Structured Light + Range Imaging
http://www.cs.cmu.edu/afs/cs/academic/class/15385-s06/lectures/ppts/lec-17.ppt
http://www.sci.utah.edu/~gerig/CS6320-S2012/CS6320_3D_Computer_Vision.html
Example: Laser scanner
3/15/2016 [email protected] 21
Digital Michelangelo Project http://graphics.stanford.edu/projects/mich/
+ very accurate < 0.01 mm
− more than 10sec per scan
Cyberware® face and head scanner
Leica ScanStation C10
3/15/2016 [email protected] 23
Leica ScanStation C10
3/15/2016 [email protected] 24
http://www.youtube.com/watch?v=cUI6HD_d89k
http://www.youtube.com/watch?v=1lDO1UevAJI
http://www.youtube.com/watch?v=cxTYYce_d_8
Leica ScanStation C10
3/15/2016 [email protected] 25
LMS LIDAR scanner from SICK
3/15/2016 [email protected] 26
https://www.youtube.com/watch?v=yfQhEiq48YA
https://www.youtube.com/watch?v=aL-Ihvu0ugo
Structured Light 3D Surface Imaging
Continuum of Triangulation Methods
3/15/2016 [email protected] 28
Slow, robust Fast, fragile
Multi-stripe
Multi-frame
Single-frame Single-stripe
Faster Acquisition?
3/15/2016 [email protected] 29
Project multiple stripes simultaneously
Correspondence problem: which stripe is which?
Common types of patterns:
Binary coded light striping
Gray/color coded light striping
http://vimeo.com/37090394
http://www.youtube.com/watch?v=a6g17-4FmFo
Structured Light Projection Classification
3/15/2016 [email protected] 30
Jason Geng, "Structured-light 3D surface imaging: a tutorial," Adv. Opt. Photon. 3, 128-160 (2011)
Binary Coding
3/15/2016 [email protected] 31
Projected
over time
Pattern 1
Pattern 2
Pattern 3
Example: 7 binary
patterns proposed
by Posdamer &
Altschuler
…
Codeword of this píxel: 1010010
identifies the corresponding pattern stripe
J. L. Posdamer and M. D. Altschuler. Surface measurement by space-encoded projected beam systems. Computer Graphics and Image Processing, 18:1–17, 1982.
Space
Time
3/15/2016 [email protected] 32
3/15/2016 [email protected] 33
3/15/2016 [email protected] 34
3/15/2016 [email protected] 35
3/15/2016 [email protected] 36
3/15/2016 [email protected] 37
3/15/2016 [email protected] 38
3/15/2016 [email protected] 39
3/15/2016 [email protected] 40
3/15/2016 [email protected] 41
3/15/2016 [email protected] 42
3/15/2016 [email protected] 43
3/15/2016 [email protected] 44
3/15/2016 [email protected] 45
3/15/2016 [email protected] 46
Fringe Projection 3D Scanning
Agenda
3/15/2016 [email protected] 48
Fringe Projection Techniques
Three step phase shifting algorithm
http://www.jove.com/video/50421/high-resolution-high-speed-three-dimensional-video-imaging-with
Three-step Phase-shifting algorithm
3/15/2016 [email protected] 49
Wrapped Phase Image
3/15/2016 [email protected] 50
Unwrapped Phase Image
3/15/2016 [email protected] 51
Calibration Plane
3/15/2016 [email protected] 52
3D Result
3/15/2016 [email protected] 53
Phase shifting patterns
3/15/2016 [email protected] 54
Sinusoid patterns
3/15/2016 [email protected] 55
System Setup
3/15/2016 [email protected] 56
Large angle setting
3/15/2016 [email protected] 57
Small angle setting
3/15/2016 [email protected] 58
Calibration I
3/15/2016 [email protected] 59
Calibration II
3/15/2016 [email protected] 60
3/15/2016 [email protected] 61
3/15/2016 [email protected] 62
3/15/2016 [email protected] 63
http://www.jove.com/video/50421/high-resolution-high-speed-three-dimensional-video-imaging-with
Light Field Camera
Stanford Multi-Camera Array
3/15/2016 [email protected] 65
http://graphics.stanford.edu/projects/array/
640 × 480 pixels × 30 fps × 128 cameras
synchronized timing
continuous streaming
flexible arrangement
Adaptive Optics Microlens Array
3/15/2016 [email protected] 69
Adaptive Optics microlens array
125μ square-sided microlenses
4000 × 4000 pixels ÷ 292 × 292 lenses = 14 × 14 pixels per lens
Adobe LightField Camera
3/15/2016 [email protected] 70
http://lightfield-forum.com/light-field-camera-prototypes/adobe-lightfield-camera-protypes/