Copying code (either from other students or from external sources) is strictly prohibited! We will be using automatic anti-plagiarism tools, and any violation of this rule will lead to expulsion from the class. Late submissions will generally not be accepted. In case of serious illness or emergency, please notify Daniele and provide a relevant medical certificate.
For each assignment, you will use the geometry processing library libigl, which includes implementations of many of the algorithms presented in class. In particular you will be using the python bindings of igl.
The libigl
library includes a set of tutorials, an introduction to which can be found in the two previous links. You are advised to look over the relevant tutorials before starting the implementation for the assignments; you are also encouraged to examine the source code of all the library functions that you use in your code to see how they were
implemented.
No libraries apart from libigl
, numpy
, meshplot
, and scipy
are permitted unless permission is granted in advance.
For every assignment, you have to provide a video report of a maximal lenght of 3 minutes. The video should contain either captions explaining what you did and the results of your assignment or a voice commentary. To prepare the video, I recommend to use either the recording option in NYU Zoom (you can create a call with only you, share the window with your browser, and then press record) or OBS (https://obsproject.com/). The video should be in a format playable using the VLC player (https://www.videolan.org/vlc/).
Before we can begin, you must install igl
, meshplot
, and jupiter
trough conda-forge
conda install -c conda-forge igl
conda install -c conda-forge meshplot
conda install scipy
conda install jupyter
Launch the notebook
jupyter notebook
You need first install nodejs
with version >= 12.0.0, e.g.
conda install nodejs==14.7.0 -c conda-forge
Next install jupyter lab with version < 3.0, e.g.
conda install -c conda-forge jupyterlab==2.2.0
Then install two jupyter lab extensions
jupyter labextension install jupyter-threejs
jupyter labextension install @jupyter-widgets/jupyterlab-manager
Now you will be able to launch the jupyter lab and see meshplot displays.
jupyter lab
Or you can simply download the file environment.yml
uploaded to create conda env and all these should have already been set up. The first line of the yml
file sets the environment's name.
conda env create -f environment.yml
To install the package manager conda we refer to its website.