Awesome
GARNet
This repository contains the source code for the paper GARNet: Global-Aware Multi-View 3D Reconstruction Network and the Cost-Performance Tradeoff.
Cite this work
@article{zhu2023garnet,
title={GARNet: Global-Aware Multi-View 3D Reconstruction Network and the Cost-Performance Tradeoff},
author={Zhu, Zhenwei and Yang, Liying and Lin, Xuxin and Yang, Lin and Liang, Yanyan},
journal={Pattern Recognition},
pages={109674},
year={2023},
publisher={Elsevier}
}
Datasets
We use the ShapeNet in our experiments, which are available below:
- ShapeNet rendering images: http://cvgl.stanford.edu/data2/ShapeNetRendering.tgz
- ShapeNet voxelized models: http://cvgl.stanford.edu/data2/ShapeNetVox32.tgz
Pretrained Models
The pretrained models on ShapeNet are available as follows:
Prerequisites
Clone the Code Repository
git clone https://github.com/GaryZhu1996/GARNet.git
Install Python Denpendencies
cd GARNet
conda env create -f environment.yml
3D Reconstruction Model
For training, please use the following command:
python runner.py
For testing, please follow the steps below:
- Update the setting of
__C.CONST.WEIGHTS
inconfig.py
as the path of the reconstruction model; - Run the following command:
python runner.py --test --score_only
View-Reduction Approach
For training IMB, please follow the steps below:
- Update the setting of
inference_model_path
incore/train_imb.py
as the path of the reconstruction model; - Run the following command:
python runner_train_imb.py
For testing with view-reduction approach, please follow the steps below:
- Update the setting of
__C.CONST.WEIGHTS
inconfig.py
as the path of the reconstruction model; - Update the setting of
__C.CONST.IMB_WEIGHTS
inconfig.py
as the path of the IMB model; - Run the following command:
python runner.py --test --score_only
License
This project is open sourced under MIT license.