This project is for implementing shape reconstruction program via level set method. It is based on a paper written by H. Zhao(2000).
Goal: Shape Reconstruction from an unorganized data set.
key Idea: Want to find a function on an image whose zero level set is nearly the same as what we want to reconstruct.
To do this, we define a functional that measures how well a function reconstruct a shape. Let
To minimize this functional, we use Level Set Method. If you want more detalis, see the paper.
There is one main file named 'shape_reconstruction' and 4 sample data for input. You can change the input in the main file. You also can change a hyper-parameter p.
Here are some sample results.
You can see the red curve converges to the data points.
This project is collaborated with Kwanghyun Choi(Seoul National University).
[1] H. Zhao, et al. Implicit and Nonparametric Shape Reconstruction from Unorganized Data Using a Variational Level Set Method. Computer Vision and Image Understanding 80, 295–314 (2000).