charlienash / nricp

Matlab implementation of non-rigid iterative closest point
MIT License
205 stars 47 forks source link

nricp - Non-rigid iterative closest point

https://github.com/charlienash/nricp

![face demo] (https://github.com/charlienash/nricp/blob/master/demos/faceDemo.jpg)

nricp is a MATLAB implementation of a non-rigid variant of the iterative closest point algorithm. It can be used to register 3D surfaces or point-clouds. The method is described in the following paper:

'Optimal Step Nonrigid ICP Algorithms for Surface Registration', Amberg, Romandhani and Vetter, CVPR, 2007.

Features:

![missing data demo] (https://github.com/charlienash/nricp/blob/master/demos/faceDemoMissing.jpg)

Dependencies

Requires:

Installation

Download the nricp directory and the dependencies and add them to your MATLAB path.

Attribution

If you use this implementation in your academic projects, please cite the paper by Amberg et al:

@inproceedings{amberg2007optimal,
  title={Optimal step nonrigid icp algorithms for surface registration},
  author={Amberg, Brian and Romdhani, Sami and Vetter, Thomas},
  booktitle={Computer Vision and Pattern Recognition, 2007. CVPR'07. IEEE Conference on},
  pages={1--8},
  year={2007},
  organization={IEEE}
}

Contact

charlie.tc.nash@gmail.com