Limit Shapes – A Tool for Understanding Shape Differences and Variability in 3D Model Collections

This is our implementation of Limit Shape framework. The code was written by Ruqi Huang.

Setup

In the demo, we compared two sets of human shapes in different poses (rest vs. bending knee) from FAUST dataset, whose raw mesh representation are stored in '/Data'. Here we assume that the ground-truth correspondence among these shapes are known.

Simply running 'demo.m' in the main folder will produce the following results, which highlights the distinctive regions with respect to the most significant area distortion and extrinsic distortion.

Citation

If you use our work, please cite our paper.

@article{huang2019limitshape,
  title={Limit Shapes – A Tool for Understanding Shape Differences and Variability in 3D Model Collections},
  author={Huang, Ruqi and Achlioptas, Panos and Guibas, Leonidas and Ovsjanikov, Maks},
  journal={SGP},
  year={2019}
}

Contact

If you have any problem about this implementation, please feel free to contact via:

rqhuang88 AT gmail DOT com