Skip to content

Files

Failed to load latest commit information.

Latest commit

 Cannot retrieve latest commit at this time.

History

History

ARNet_shape_reconstruction

README.md

ARNet_shape_reconstruction

Installation

The code has been tested on Ubuntu 20.04 with CUDA 12.0.

virtualenv -p /usr/bin/python venv_recon
source venv_recon/bin/activate
cd ARNet_shape_reconstruction
pip install -r requirements.txt

Chamfer Distance Loss

Please find the chamferdist/chamfer.py file under venv_recon/lib, and replace the chamfer.py in this python library to our chamfer.py file under the folder.

Data Preparation

Download the dataset from the Google Drive and unzip under the folder.

About Configs

  • exp_name - used to specify data split
  • ckpt_path - used to specify saved model for testing

Training

Run the following command:

CUDA_VISIBLE_DEVICES=6 python main.py data_split_<1-3>

Evaluation

Run the following command for evaluation of the trained models:

CUDA_VISIBLE_DEVICES=6 python test.py data_split_<1-3>

The visual reconstruction results are saved under image folder.The numerical reconstruction results will be saved in .npy file under reconstruction_results folder.