Learning to Generate 3D Shapes and Scenes | CVPR 2021 Workshop [Paper]
Deep Mesh Prior is an unsupervised mesh restoration method using graph convolutional networks, which takes a single incomplete mesh as input data and directly outputs the reconstructed mesh without being trained using large-scale datasets.
git clone https://github.com/astaka-pe/DeepMeshPrior.git
cd DeepMeshPrior
conda env create -f environment.yml
conda activate dmp
mkdir datasets/{c_output,d_output}
mkdir logs/
python denoise.py -i datasets/d_input/dragon
- To view the loss and MAD, run
tensorboard --logdir logs/denoise
in another terminal and click http://localhost:6006.
python completion.py -i datasets/c_input/dragon
- To view the loss, run
tensorboard --logdir logs/competion
in another terminal and click http://localhost:6006.