Fast 3D scanning with Automatic Motion Compensation
Thibaut Weise, Bastian Leibe and Luc Van Gool
Abstract:
We present a novel 3D scanning system combining stereo and active illumination based on phase-shift for robust and accurate scene reconstruction. Stereo overcomes the traditional phase discontinuity problem and allows for the reconstruction of complex scenes containing multiple objects. Due to the sequential recording of three patterns, motion will introduce artifacts in the reconstruction. We develop a closed-form expression for the motion error in order to apply motion compensation on a pixel level. The resulting scanning system can capture accurate depth maps of complex dynamic scenes at 17 fps and can cope with both rigid and deformable objects.
Paper:
[pdf]
Bibtex:
[bib]
@InProceedings{
  author = {T. Weise and B. Leibe and L. Van Gool},
  title = {Fast 3D Scanning with Automatic Motion Compensation},
  booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR'07)},
  year = {2007},
  month = {June}
}      
Videos:
Complex Scene Cloth Hand Gestures Speaking Face Scanning
(divx, 5.6 MB) (divx, 7.0 MB) (divx, 6.9 MB) (divx, 2.8 MB) (divx, 2.2 MB) (divx, 1.2 MB)
Data Repository:
Scanning data is publically available for research purposes. I would kindly ask you to reference my work in case you use it for your research. Further, I would be glad if you would send me an [email] about your work, so that I can place a link on this site. Feel free to contact me in case you want specific scanning data, and we will see what we can do. Note, though, that any generated data will be put on this site, so that other researchers can use it.
All 3D scans are provided as triangulated point clouds in obj-format. All measurements are in mm. The corresponding depthmaps can be generated using the following intrinsic parameters:
Focal Length:
 1180.63 (fx)  1182.99 (fy)
Principal Point:
  346.18 (cx)   282.29 (cy)
[sample]
(obj, 2.6 MB)
[data]
(tar.gz, 87 MB)
100 frames
Short sequence of a slightly closing hand with dense sampling in geometry and time.

used by [Mitra2007,Chang2008]
[sample]
(obj, 1.6 MB)
[data]
(tar.gz, 1.1 GB)
2200 frames
Turntable sequence of a soft bee with dense sampling in geometry and time.

used by [Mitra2007]
[sample]
(obj, 1.9 MB)
[data]
(tar.gz, 1.3 GB)
2200 frames
Turntable sequence of a soft coati with dense sampling in geometry and time.

used by [Mitra2007]
[sample]
(obj, 2.0 MB)
[data]
(tar.gz, 1.8 GB)
2200 frames
Turntable sequence of a teapot with dense sampling in geometry and time.

used by [Mitra2007]
References:
[Mitra2007] Niloy J. Mitra, Simon Floery, Maks Ovsjanikov, Natasha Gelfand, Leonidas Guibas and Helmut Pottmann. Dynamic Geometry Registration. In SGP '07: Proceedings of the fifth Eurographics symposium on Geometry processing, pages 173 - 182, 2007
[Chang2008] Will Chang and Matthias Zwicker. Automatic Registration for Articulated Shapes. In Computer Graphics Forum 27(5) (Symposium on Geometry Processing), pages 1459--1468, 2008