Build Your Own 3D Scanner: Conclusion

Session II
Structured Lighting
Projector Calibration / Structured Light Reconstruction
Combining Point Clouds Recovered from Multiple Views
Surface Reconstruction from Point Clouds
Elementary Mesh Processing
Conclusion / Q & A
Course Schedule

Summary of Optical Triangulation Assign Texture (BRDF, etc.) Align 3D Point Clouds Reconstruct 3D Point Cloud(s) Data Capture Extract Isosurface (Polyhedral Mesh)

Summary of Optical Triangulation: Projector-Camera Calibration calibration board camera coordinate system projector coordinate system 0 500 1000 0 500 1000 1500 400 200 0 X c2 Y c2 O c2 Z c2 X p Y p X c Z p O p X c1 Z c1 Y c1 O c1 (mm) Z c (mm) Y c (mm)

Summary of Optical Triangulation: Post-Processing (Point Clouds  Meshes)
Register multiple point cloud scans
Initialize by selecting matching point pairs
Iteratively refine with ICP
Fit watertight implicit surface to point cloud
Automatically fills holes
Isosurface polygon mesh from regular hexahedral grid
Triangle mesh from adaptive tetrahedral mesh
Geometry processing
Denoising / Simplification / Refinement
Remeshing / Reparameterization

Spatio-Temporal Trade-offs for Active Imaging Number of Projected Patterns Temporal Resolution Spatial Resolution
Structured lighting patterns can be encoded spatially and/or temporally
Purely temporal encodings  high-resolution capture of static scenes
Purely spatial encodings  low-resolution capture of dynamic scenes
Hybrid codes adjust trade-off between spatial and temporal resolution
High-speed/synchronized systems allow high-resolution dynamic capture
single-shot swept-plane/flying spot hybrid low (static) high (dynamic) low high

Recent Work: Scanning Dynamic Scenes S.G. Narasimhan, S. J. Koppal, and S. Yamazaki. Temporal Dithering of Illumination for Fast Active Vision . European Conf. Comp. Vision, 2008

*See the SIGGRAPH 2009 course: Acquisition of Optically Complex Objects and Phenomena by Heidrich and Ihrke. Recent Work: Scanning Transparent Objects M. B. Hullin, M. Fuchs, I. Ihrke, H.-P. Seidel, and H. P. A. Lensch. Fluorescent Immersion Range Scanning . ACM SIGGRAPH, 2008
Conventional laser striping observes contour on an opaque object
Fluorescent immersion laser striping observes endpoint of laser sheet
Captures cross-section in a single image with an index-matching fluid

*See the SIGGRAPH 2009 course: Acquisition of Optically Complex Objects and Phenomena by Heidrich and Ihrke. Recent Work: Scanning Transparent Objects B. Trifonov, D. Bradley, and W. Heidrich. Tomographic Reconstruction of Transparent Objects . Eurographics Symposium on Rendering (2006)

Limitations of Structured Lighting
Only recovers mutually-visible surface (i.e., must be illuminated and imaged)
Complete model requires multiple scans or additional projectors/cameras
Often requires post-processing (e.g., ICP)
Multiple Views with Planar Mirrors
Trade spatial for angular resolution
Multiple views by including planar mirrors
What about illumination interference?
Use orthographic illumination
System Components
Multi-view: digital camera + planar mirrors
Orthographic: DLP projector + Fresnel lens
Recent Work: Multiplexed Views and Eliminating Moving Parts D. Lanman, D. Crispell, and G. Taubin. Surround Structured Lighting: 3-D Scanning with Orthographic Illumination . Computer Vision and Image Understanding , 2009 A. Griesser, T. P. Koninckx, and L. Van Gool. Adaptive real-time 3D acquisition and contour tracking within a multiple structured light system . Computer Graphics and Applications , 2004 Surround Lighting by Precise Alignment [Van Gool '04]

Applications: 3D Printing http://shapeways.com

Applications: DIY 3D Printing MakerBot 'Cupcake' DIY 3D printer [~$750] http://store.makerbot.com/

Applications: DIY 3D Printing RepRap (Replicating Rapid-prototyper) Project [~$650] http://reprap.org

Applications: DIY 3D Printing Fab@Home Fabber Model 1 [~$2,300] http://fabathome.org

Applications: DIY 3D Printing CandyFab 4000: 3D Freeform Fabrication [~$500] http://www.candyfab.org/

Applications: Entertainment (Leaving the “Uncanny Valley”) *See the SIGGRAPH 2009 course: The Digital Emily Project: Photoreal Facial Modeling and Animation by Debevec et al. Light Stage 5 Data Acquisition Color Fringes for Coarse Geometry Gradient Illumination for Fine Details

Applications: Entertainment (3D Cinematography)
3D Scanners in Film Production
“ House of Cards” by Radiohead
Uses phase-shifting and time-of-flight
GeoVideo Real-Time Motion Capture
Velodyne HDL-64E LIDAR
http://code.google.com/creative/radiohead/

International Conference on Computational Photography Papers due November 2, 2009 http://cameraculture.media.mit.edu/iccp10

Thank you for attending! Win a SIGGRAPH 2009 mug: http://ww.siggraph.org/courses_evaluation One winner (per course) notified by email tonight! http://mesh.brown.edu/byo3d

http://mesh.brown.edu	4460
http://translate.googleusercontent.com	37
http://www.slideshare.net	15
http://fanyi.youdao.com	3
http://207.46.192.232	2

Build Your Own 3D Scanner: Conclusion

by Douglas Lanman on Aug 20, 2009

Accessibility

Categories

Tags

Upload Details

Usage Rights

5 Embeds 4,517

Statistics

Build Your Own 3D Scanner: Conclusion — Presentation Transcript