Deep Point Cloud Benchmark

This is a framework for running common depp learning models for point cloud analysis tasks against classic benchmark. It heavily relies on pytorch geometric and hydra core.

Getting started

You will first need to install poetry in order to setup a virtual environments and install the relevant packages, then run

poetry install

This will install all required dependencies in a new virtual environment.

Benchmark pointnet++ on Segmentation task for shapenet

poetry run python train.py tested_model.name=pointnet2

Contributers