SE-ORNet
说明： PyTorch在我们的CVPR 2023论文SE ORNet中的实现：无监督点云形状中心的自集成方向感知网络...，
(PyTorch implementation for our CVPR 2023 paper SE-ORNet: Self-Ensembling Orientation-aware Network for Unsupervised Point Cloud Shape Correspondence)
(PyTorch implementation for our CVPR 2023 paper SE-ORNet: Self-Ensembling Orientation-aware Network for Unsupervised Point Cloud Shape Correspondence)
文件列表:
ChamferDistancePytorch/ (0, 2023-09-06)
LICENSE (1068, 2023-09-06)
__init__.py (0, 2023-09-06)
data/ (0, 2023-09-06)
data/__init__.py (0, 2023-09-06)
data/datasets/ (0, 2023-09-06)
data/datasets/surreal/ (0, 2023-09-06)
data/datasets/surreal/surreal_test.pth (16536350, 2023-09-06)
data/generate_smal.md (619, 2023-09-06)
data/point_cloud_db/ (0, 2023-09-06)
data/point_cloud_db/download_surreal.sh (633, 2023-09-06)
data/point_cloud_db/faust.py (3767, 2023-09-06)
data/point_cloud_db/point_cloud_dataset.py (7029, 2023-09-06)
data/point_cloud_db/shrec.py (5642, 2023-09-06)
data/point_cloud_db/smal.py (3769, 2023-09-06)
data/point_cloud_db/surreal.py (4010, 2023-09-06)
data/point_cloud_db/tosca.py (5055, 2023-09-06)
models/ (0, 2023-09-06)
models/correspondence_utils.py (1819, 2023-09-06)
models/data_augment_utils.py (9722, 2023-09-06)
models/metrics/ (0, 2023-09-06)
models/metrics/aux_for_loss.py (389, 2023-09-06)
models/metrics/metrics.py (2704, 2023-09-06)
models/modules/ (0, 2023-09-06)
models/modules/dgcnn.py (3399, 2023-09-06)
models/modules/dgcnn_modular.py (4456, 2023-09-06)
models/modules/dgcnn_with_angle.py (9605, 2023-09-06)
models/modules/orient_module.py (9884, 2023-09-06)
... ...
LICENSE (1068, 2023-09-06)
__init__.py (0, 2023-09-06)
data/ (0, 2023-09-06)
data/__init__.py (0, 2023-09-06)
data/datasets/ (0, 2023-09-06)
data/datasets/surreal/ (0, 2023-09-06)
data/datasets/surreal/surreal_test.pth (16536350, 2023-09-06)
data/generate_smal.md (619, 2023-09-06)
data/point_cloud_db/ (0, 2023-09-06)
data/point_cloud_db/download_surreal.sh (633, 2023-09-06)
data/point_cloud_db/faust.py (3767, 2023-09-06)
data/point_cloud_db/point_cloud_dataset.py (7029, 2023-09-06)
data/point_cloud_db/shrec.py (5642, 2023-09-06)
data/point_cloud_db/smal.py (3769, 2023-09-06)
data/point_cloud_db/surreal.py (4010, 2023-09-06)
data/point_cloud_db/tosca.py (5055, 2023-09-06)
models/ (0, 2023-09-06)
models/correspondence_utils.py (1819, 2023-09-06)
models/data_augment_utils.py (9722, 2023-09-06)
models/metrics/ (0, 2023-09-06)
models/metrics/aux_for_loss.py (389, 2023-09-06)
models/metrics/metrics.py (2704, 2023-09-06)
models/modules/ (0, 2023-09-06)
models/modules/dgcnn.py (3399, 2023-09-06)
models/modules/dgcnn_modular.py (4456, 2023-09-06)
models/modules/dgcnn_with_angle.py (9605, 2023-09-06)
models/modules/orient_module.py (9884, 2023-09-06)
... ...
## SE-ORNet: Self-Ensembling Orientation-aware Network for Unsupervised Point Cloud Shape Correspondence
PyTorch implementation for our CVPR 2023 paper SE-ORNet.
[](https://paperswithcode.com/sota/3d-dense-shape-correspondence-on-shrec-19?p=se-ornet-self-ensembling-orientation-aware)
[[Project Webpage](https://chuxwa.github.io/SE-ORNet/)]
[[Paper](https://arxiv.org/abs/2304.05395)]
## News
* **28. February 2023**: SE-ORNet is accepted at CVPR 2023. :fire:
* **10. April 2023**: [SE-ORNet preprint](https://arxiv.org/abs/2304.05395) released on arXiv.
* **Coming Soon**: Code will be released soon.
## Installation
1. Create a virtual environment via `conda`.
```shell
conda create -n se_ornet python=3.10 -y
conda activate se_ornet
```
2. Install `torch` and `torchvision`.
```shell
conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.3 -c pytorch -y
```
3. Install environments
```shell
sh setup.sh
```
## Code structure
```
├── SE-ORNet
│ ├── __init__.py
│ ├── train.py <- the main file
│ ├── models
│ │ ├── metrics
│ │ ├── modules
│ │ ├── runners
│ │ ├── correspondence_utils.py
│ │ ├── data_augment_utils.py
│ │ └── shape_corr_trainer.py
│ ├── utils
│ │ ├── __init__.py
│ │ ├── argparse_init.py
│ │ ├── cyclic_scheduler.py
│ │ ├── model_checkpoint_utils.py
│ │ ├── pytorch_lightning_utils.py
│ │ ├── switch_functions.py
│ │ ├── tensor_utils.py
│ │ └── warmup.py
│ ├── visualization
│ │ ├── __init__.py
│ │ ├── mesh_container.py
│ │ ├── mesh_visualization_utils.py
│ │ ├── mesh_visualizer.py
│ │ ├── orca_xvfb.bash
│ │ └── visualize_api.py
│ └── ChamferDistancePytorch
├── data
│ ├── point_cloud_db
│ ├── __init__.py
│ └── generate_smal.md
├── .gitignore
├── .gitmodules
├── README.md
└── LICENSE
```
## Dependencies
The main dependencies of the project are the following:
```yaml
python: 3.10
cuda: 11.3
pytorch: 1.12.1
```
## Datasets
The method was evaluated on:
* SURREAL
* 230k shapes (DPC uses the first 2k).
* [Dataset website](https://www.di.ens.fr/willow/research/surreal/data/)
* This code downloads and preprocesses SURREAL automatically.
* SHREC’19
* 44 Human scans.
* [Dataset website](http://3dor2019.ge.imati.cnr.it/shrec-2019/)
* This code downloads and preprocesses SURREAL automatically.
* SMAL
* 10000 animal models (2000 models per animal, 5 animals).
* [Dataset website](https://smal.is.tue.mpg.de/)
* Due to licencing concerns, you should register to [SMAL](https://smal.is.tue.mpg.de/) and download the dataset.
* You should follow data/generate_smal.md after downloading the dataset.
* To ease the usage of this benchmark, the processed dataset can be downloaded from [here](https://mailtauacil-my.sharepoint.com/:f:/g/personal/dvirginzburg_mail_tau_ac_il/Ekm37j0fi71Fn305v9nmXHABCSc1mWFa17uAc2jOngcyTQ?e=Ns2InB). Please extract and put under `data/datasets/smal`
* TOSCA
* 41 Animal figures.
* [Dataset website](http://tosca.cs.technion.ac.il/book/resources_data.html)
* This code downloads and preprocesses TOSCA automatically.
* To ease the usage of this benchmark, the processed dataset can be downloaded from [here](https://mailtauacil-my.sharepoint.com/:f:/g/personal/dvirginzburg_mail_tau_ac_il/EoMgplq-XqlGpl6K6lW6C8gBCxfq2gWXQ4f94xchF3dc9g?e=USid0X). Please extract and put under `data/datasets/tosca`
## Models
The metrics are obtained in 5 training runs followed by 5 test runs. We report both the best and the average values (the latter are given in round brackets).
**Human Datasets**
| Dataset | mAP@0.25 | mAP@0.5 | Download |
|:-------:|:--------:|:-------:|:--------:|
| SHREC’19 | 17.5 (16.8) | 5.1 (5.6) | [model](https://drive.google.com/drive/folders/1YG342B5f4Yhb7Z9OMo3KdHvOmvHJxzQm?usp=sharing) |
| SURREAL | 22.3 (21.3) | 4.5 (4.8) | [model](https://drive.google.com/drive/folders/1NiL2JF5Rd_xmbJkOqiJOP1b_nCmSY7-Y?usp=sharing) |
**Animal Datasets**
| Dataset | mAP@0.25 | mAP@0.5 | Download |
|:-------:|:--------:|:-------:|:--------:|
| TOSCA | 40.8 (38.1) | 2.7 (2.8) | [model](https://drive.google.com/drive/folders/19dAHpJe1o7KMSTQBDyQ0xMu2lTveg_AT?usp=sharing) |
| SMAL | 38.3 (36.2) | 3.3 (3.8) | [model](https://drive.google.com/drive/folders/1yYV_civ6j9-9Jmn1KYp8LybU6MEo57Ch?usp=sharing) |
## Training & inference
For training run
```
python train.py --dataset_name
```
The code is based on [PyTorch-Lightning](https://pytorch-lightning.readthedocs.io/en/latest/), all PL [hyperparameters](https://pytorch-lightning.readthedocs.io/en/latest/common/trainer.html) are supported.
For testing, simply add `--do_train false` flag, followed by `--resume_from_checkpoint` with the relevant checkpoint.
```
python train.py --do_train false --resume_from_checkpoint
```
Test phase visualizes each sample, for faster inference pass `--show_vis false`.
We provide a trained checkpoint repreducing the results provided in the paper, to test and visualize the model run
```
python train.py --show_vis --do_train false --resume_from_checkpoint data/ckpts/surreal_ckpt.ckpt
```
## BibTeX
If you like our work and use the codebase or models for your research, please cite our work as follows.
```bibtex
@inproceedings{
Deng2023seornet,
title={{SE}-{ORN}et: Self-Ensembling Orientation-aware Network for Unsupervised Point Cloud Shape Correspondence},
author={Jiacheng Deng, ChuXin Wang, Jiahao Lu, Jianfeng He, Tianzhu Zhang, Jiyang Yu, Zhe Zhang},
booktitle={Conference on Computer Vision and Pattern Recognition 2023},
year={2023},
url={https://openreview.net/forum?id=DS6AyDWnAv}
}
```
近期下载者：
相关文件：
收藏者：