Safran/Shape-as-Point
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

📝 update README.md for better installation instructions

Browse source
This commit is contained in:
Laurent FAINSIN 2023-05-30 11:46:06 +02:00
parent 377132fdd5
commit 43164f892f
1 changed files with 16 additions and 15 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

31
README.md
View file

@ -6,9 +6,9 @@
This repository contains the implementation of the paper: This repository contains the implementation of the paper:
Shape As Points: A Differentiable Poisson Solver Shape As Points: A Differentiable Poisson Solver
[Songyou Peng](https://pengsongyou.github.io/), [Chiyu "Max" Jiang](https://www.maxjiang.ml/), [Yiyi Liao](https://yiyiliao.github.io/), [Michael Niemeyer](https://m-niemeyer.github.io/), [Marc Pollefeys](https://www.inf.ethz.ch/personal/pomarc/) and [Andreas Geiger](http://www.cvlibs.net/) [Songyou Peng](https://pengsongyou.github.io/), [Chiyu "Max" Jiang](https://www.maxjiang.ml/), [Yiyi Liao](https://yiyiliao.github.io/), [Michael Niemeyer](https://m-niemeyer.github.io/), [Marc Pollefeys](https://www.inf.ethz.ch/personal/pomarc/) and [Andreas Geiger](http://www.cvlibs.net/)
**NeurIPS 2021 (Oral)** **NeurIPS 2021 (Oral)**
If you find our code or paper useful, please consider citing If you find our code or paper useful, please consider citing
@ -23,7 +23,7 @@ If you find our code or paper useful, please consider citing
## Installation ## Installation
First you have to make sure that you have all dependencies in place. First you have to make sure that you have all dependencies in place.
The simplest way to do so, is to use [anaconda](https://www.anaconda.com/). The simplest way to do so, is to use [anaconda](https://www.anaconda.com/).
You can create an anaconda environment called `sap` using You can create an anaconda environment called `sap` using
``` ```
@ -31,14 +31,15 @@ conda env create -f environment.yaml
conda activate sap conda activate sap
``` ```
Next, you should install [PyTorch3D](https://pytorch3d.org/) (**>=0.5**) yourself from the [official instruction](https://github.com/facebookresearch/pytorch3d/blob/master/INSTALL.md#3-install-wheels-for-linux). Next, you should install [PyTorch3D](https://pytorch3d.org/) (**>=0.5**) yourself from the [official instruction](https://github.com/facebookresearch/pytorch3d/blob/master/INSTALL.md#3-install-wheels-for-linux).
And install [PyTorch Scatter](https://github.com/rusty1s/pytorch_scatter): ```bash
```sh git clone https://github.com/facebookresearch/pytorch3d.git
conda install pytorch-scatter -c pyg cd pytorch3d
module load compilers
pip install -e .
``` ```
## Demo - Quick Start ## Demo - Quick Start
First, run the script to get the demo data: First, run the script to get the demo data:
@ -82,7 +83,7 @@ You can find the reconstrution on `out/demo_shapenet_outlier/generation/vis`.
We have different dataset for our optimization-based and learning-based settings. We have different dataset for our optimization-based and learning-based settings.
### Dataset for Optimization-based Reconstruction ### Dataset for Optimization-based Reconstruction
Here we consider the following dataset: Here we consider the following dataset:
- [Thingi10K](https://arxiv.org/abs/1605.04797) (synthetic) - [Thingi10K](https://arxiv.org/abs/1605.04797) (synthetic)
- [Surface Reconstruction Benchmark (SRB)](https://github.com/fwilliams/deep-geometric-prior) (real scans) - [Surface Reconstruction Benchmark (SRB)](https://github.com/fwilliams/deep-geometric-prior) (real scans)
- [MPI Dynamic FAUST](https://dfaust.is.tue.mpg.de/) (real scans) - [MPI Dynamic FAUST](https://dfaust.is.tue.mpg.de/) (real scans)
@ -94,20 +95,20 @@ You can download the processed dataset (~200 MB) by running:
bash scripts/download_optim_data.sh bash scripts/download_optim_data.sh
``` ```
### Dataset for Learning-based Reconstruction ### Dataset for Learning-based Reconstruction
We train and evaluate on [ShapeNet](https://shapenet.org/). We train and evaluate on [ShapeNet](https://shapenet.org/).
You can download the processed dataset (~220 GB) by running: You can download the processed dataset (~220 GB) by running:
```bash ```bash
bash scripts/download_shapenet.sh bash scripts/download_shapenet.sh
``` ```
After, you should have the dataset in `data/shapenet_psr` folder. After, you should have the dataset in `data/shapenet_psr` folder.
Alternatively, you can also preprocess the dataset yourself. To this end, you can: Alternatively, you can also preprocess the dataset yourself. To this end, you can:
* first download the preprocessed dataset (73.4 GB) by running [the script](https://github.com/autonomousvision/occupancy_networks#preprocessed-data) from Occupancy Networks. * first download the preprocessed dataset (73.4 GB) by running [the script](https://github.com/autonomousvision/occupancy_networks#preprocessed-data) from Occupancy Networks.
* check [`scripts/process_shapenet.py`](https://github.com/autonomousvision/shape_as_points/tree/main/scripts/process_shapenet.py), modify the base path and run the code * check [`scripts/process_shapenet.py`](https://github.com/autonomousvision/shape_as_points/tree/main/scripts/process_shapenet.py), modify the base path and run the code
## Usage for Optimization-based 3D Reconstruction ## Usage for Optimization-based 3D Reconstruction
For our optimization-based setting, you can consider running with a coarse-to-fine strategy: For our optimization-based setting, you can consider running with a coarse-to-fine strategy:
```python ```python
@ -124,7 +125,7 @@ You might need to modify the `CONFIG.yaml` accordingly.
## Usage for Learning-based 3D Reconstruction ## Usage for Learning-based 3D Reconstruction
### Mesh Generation ### Mesh Generation
To generate meshes using a trained model, use To generate meshes using a trained model, use
```python ```python
python generate.py configs/learning_based/CONFIG.yaml python generate.py configs/learning_based/CONFIG.yaml