This is an example code for the paper "Geometric analysis of shape variability of lower jaws of prehistoric humans" by Jing Ren, Peter Wonka, Gowtham Harihara, and Maks Ovsjanikov. The paper can be found here.
The dataset consists of 10 jaws of prehistoric humans from the anthropologists and we visualize them below. We provide the original dataset and the remodelled dataset where we clean the dataset a bit to remove outliers, fix holes, etc. The analysis is done on both datasets.
We use functional map pipeline with BCICP refinement method to compute dense
correspondences between every pair of jaws in the dataset. Below we visualize the map via color transfer (the corresponding regions have the same color).
The computed maps are saved in the Dataset folder.
See run_compute_maps.m to see the paramters for map computation.
Alternatively, one can try the ZoomOut method to efficiently refine the computed maps.
We deform on shape into another and use the corresponding transformation to define the shape similarity between the shape pair for shape analysis.
See run_nonRigidMatching for more details.
Here is a video that visualizes the shape deformation over iterations.
- We also provide the code to reproduce the figures in the paper
- Please let us know ([email protected]) if you have any question regarding the algorithms/paper (ง ´͈౪`͈)ว or you find any bugs in the implementation (ꐦ°᷄д°᷅)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. For any commercial uses or derivatives, please contact us ([email protected], [email protected], [email protected]).
