maipbui / pcc_assessment

A benchmark platform to evaluate the performance of 3D PCC algorithms.

PCC Assessment: Point Cloud Compression Assessment

Tested Environment

Ubuntu 20.04.3 LTS

Requirements

Installation

0. Install basic prerequisites:

sudo apt install git gcc g++ cmake subversion xvfb libblas-dev libatlas-base-dev nvidia-cuda-toolkit -y

1. Install Anaconda: https://www.anaconda.com/products/individual#linux
2. Install traditional PCC Algorithms: Draco, GPCC, VPCC

cd [root]
mkdir build
cd build
cmake ..
make

3. Install separate Conda environment for neural network-based PCC algorithms where {PCC_name} is the name of PCC algorithm. The env config files can be found under [root]/env folder:

cd [root]
conda env create -f env/{PCC_name}.yml

4. To activate installed Conda environment for neural network-based PCC algorithms:

conda activate {PCC_name}

Acknowledgement

Thank you and credits to xtorker for part of the easy-to-use Anaconda environment configs env