This work implements a pipeline to automatically register point clouds captured by depth sensors like the Microsoft Kinect. The method neither makes assumptions about the view order of the sensors, nor uses any kind of other task-dependent prior knowledge. All point clouds within the input set are aligned in a common, global coordinate system by a successive application of pairwise registration steps. The order of the individual transformations is automatically derived from a global point cloud graph, which uses the overlap of two individual point clouds to establish a weighted link between them. The experiments prove the generality of the proposed approach by applying it to data from a single but moving sensor, multiple Kinects that run simultaneously, as well as laser scanning data. The obtained accuracies in terms of the mean nearest point neighbour distance are below 0.01% of the maximum point distance of the reference data in all cases.


FPCF - Fast Point Cloud Fusion
The FPCF framework is a able to perform an automatic registration of an unorganised point cloud set.

The required external dependencies are located in the "dependencies" directory.
Additional documentation with compile instructions are located in the "doc" directory.
The framework was tested using Windows 8 64-Bit, Microsoft Visual Studio 2010, and libraries with the following version number.:
- CMake ( http://www.cmake.org/cmake/resources/software.html )
- Boost 1.53 ( http://www.boost.org/ )
- Eigen 3.0.5 ( http://eigen.tuxfamily.org/index.php )
- Flann 1.8.4 ( http://www.cs.ubc.ca/research/flann/#download )
- PCL 1.7 ( http://www.cs.ubc.ca/research/flann/ )
- Qhull 2012.1 (http://www.qhull.org/ )
- VTK 5.10.1 (http://www.vtk.org/ )
- Qt 4.8.0 ( http://qt-project.org/ )



If you use FPCF for your research please cite:

T. Weber, R. Hänsch, O. Hellwich, Automatic registration of unordered point clouds acquired by Kinect sensors using an overlap heuristic, ISPRS Journal of Photogrammetry and Remote Sensing, Volume 102, April 2015, Pages 96-109

@ARTICLE{ Haensch:2014:ISPRSJ,
  author = {Hänsch, Ronny and Weber, Thomas and Hellwich, Olaf},
  title  = {Automatic registration of unordered point clouds acquired by Kinect sensors using an overlap heuristic},
  journal= {ISPRS Journal of Photogrammetry and Remote Sensing},
  volume = {102},
  pages  = {96-109},
  year   = {2015},
  issn   = {0924-2716},
  doi    = {http://dx.doi.org/10.1016/j.isprsjprs.2014.12.014},
  url    = {http://www.sciencedirect.com/science/article/pii/S0924271614002895}


FPCF - Fast Point Cloud Fusion Copyright (c) 2014, Thomas Weber

FPCF is free software: you can redistribute it and/or modify it under the terms of the Modified BSD License.

FPCF is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the Modified BSD License for more details.


Comparison of 3D Interest Point Detectors and Descriptors for Point Cloud Fusion
Hänsch, R., Weber, T. & Hellwich, O., ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences 2014, pp. 57-64, Zurich, Switzerland, September 2014.