Skip to content
No description, website, or topics provided.
C++ Other
  1. C++ 98.8%
  2. Other 1.2%
Branch: master
Clone or download
Latest commit 746b8c2 Oct 3, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data clean up Sep 10, 2019
vipss disable multicomponent Oct 3, 2019
LICENSE.txt clean up Sep 10, 2019
ReadMe.md readmr Sep 10, 2019
env.sh environment & timing Sep 9, 2019

ReadMe.md

VIPSS - Zhiyang Huang (2019)


This code implements the algorithm described in

Variational Implicit Point Set Surface
Zhiyang Huang, Nathan Carr, Tao Ju.
ACM Transactions on Graphics (Proc. ACM Siggraph 2019)

The primary function of this program is to predict the normal as well as the underlying surface of a given set of unoriented points.

Currently, the code is only tested on Mac OS X 10.13 or above, it should work at other platforms with minor modification.

BUILDING

The code has only two dependencies: 1)Armadillo, 2)NLOPT

  1. http://arma.sourceforge.net/

  2. https://nlopt.readthedocs.io/en/latest/

You can download & install them by yourself, or run the env.sh script which will install homebrew first, and then install the two dependencies via homebrew.

$source env.sh

Then go to the vipss folder, build the Cmake file and make:
$cd vipss
$cmake .
$make

In the vipss directory, there should be an executable called "vipss" (or "vipss.exe" on Windows if it is successfully built).

RUNNING

To run the code from the command line, type:

$./vipss -i input_file_name [-l user_lambda] [-s number_voxel_per_line] [-o output_file_path]

where:

  1. -i: followed by the path of the input file. input_file_name is a path to the input file. currently, support file format includes ".xyz". The format of .xyz is:
    x1, y1, z1
    x2, y2, z2
    .....
    xn, yn, zn

  2. -l: optional argument. Followed by a float number indicating the lambda which balances the energy (see the paper for details). Default 0 (exact interpolation), you should set and tune this number according to your inputs.

  3. -s: optional argument. Followed by a unsigned integer number indicating the number of voxels in each dimension for the implicit surfacing. Only If -s is included in the command line, the program would output the surface ([input file name]_surface.ply). We recomment using 100 for a default value, and you should set this according to your inputs and the precision of the output. Notices that the surfacing algorithm takes quite a long time for surfacing the zero-level set, and it depends on the resolution and the shape of the zero-level set.

  4. -o: optional argument. Followed by the path of the output path. output_file_path is a path to the folder for generating output files. Default the folder of the input file.

  5. -t: optional argument. when it is activated, the program will create a txt file ([input file name]_time.txt) which records the timing information in this run.

Some examples have been placed at data folder for testing:

  1. $./vipss -i ../data/hand_ok/input.xyz -l 0 -s 200
  2. $./vipss -i ../data/walrus/input.xyz -l 0.003 -s 100

The program will generate the predicted normal in [input file name]_normal.ply. If -s is included in the command line, the program will generate the surface as the zero-level set of the solved implicit function ([input file name]_surface.ply).

🔔 To generate all the example in the paper, please run the makefigure.sh script in the vipss folder:
$source makefigure.sh
The result will be generated into the data folder respectively.

🔔 Notice: The current surface tracker does NOT producing multi-component surface. We refer the user to use CGAL implicit mesher if needed. We will update the program to it soon.

📣 For further questions about the code and the paper, please contact Zhiyang Huang at adshhzy@gmail.com or zhiyang.huang@wustl.edu (might be invalid after he graduated). You can also contact Prof. Tao Ju at taoju@wustl.edu.

You can’t perform that action at this time.