Repository for the Project on: Harris 3D: A robust extension of the Harris operator for interest point detection on 3D meshes

Students: Lucas Elbert Bjoern Michele

Explanations for the code:

Base.py: Includes the basic class which can calculate the Harris response values and selects interest points (for all local neighborhood methods)
test_experiments.py: Includes the class Experiment and Experiment move. These two classes provide everything to run the experiments.
Experiments-Hyperparameter.ipynb: Jupyter notebook which includes the calculation, saving and visualization of the values for the Hyperparameter selection.
Experiments-Transformations.ipynb: Jupyter notebook which includes the calculation, saving and visualization of the values for the transformations experiments.
Basic shapes.ipynb: Jupyter notebook which includes experiments with the basic forms (pyramid, cube), and the visualization of the intermeidate steps of the interest points selection.
ply.py: Basic functions to read and write ply files (code taken from the course TPs)
bunny.ply : Bunny point cloud (taken from the course TPs)

The bunny and the human mesh files can be downloaded here https://drive.google.com/drive/folders/1a2Fd6Vw1SJIwW_M1iUP6pm88QLg7xybT?usp=sharing. The files should be kept in the Datasets/tosca structure parallel to the 3D Harris folder (not in it).

BjoernMichele/3DHarris