Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/chrischoy/spatiotemporalsegmentation
4D Spatio-Temporal Semantic Segmentation on a 3D video (a sequence of 3D scans)
https://github.com/chrischoy/spatiotemporalsegmentation
3d 3d-vision convolutional-neural-network cvpr minkowski-engine scannet semantic-segmentation spatio-temporal-analysis
Last synced: about 12 hours ago
JSON representation
4D Spatio-Temporal Semantic Segmentation on a 3D video (a sequence of 3D scans)
- Host: GitHub
- URL: https://github.com/chrischoy/spatiotemporalsegmentation
- Owner: chrischoy
- License: mit
- Created: 2019-09-28T11:45:58.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2022-12-05T06:34:28.000Z (about 2 years ago)
- Last Synced: 2025-01-22T13:07:18.918Z (8 days ago)
- Topics: 3d, 3d-vision, convolutional-neural-network, cvpr, minkowski-engine, scannet, semantic-segmentation, spatio-temporal-analysis
- Language: Python
- Homepage:
- Size: 1.21 MB
- Stars: 294
- Watchers: 12
- Forks: 52
- Open Issues: 31
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
[pypi-url]: https://pypi.org/project/MinkowskiEngine/
# Spatio-Temporal Segmentation
This repository contains the accompanying code for [4D-SpatioTemporal ConvNets: Minkowski Convolutional Neural Networks, CVPR'19](https://arxiv.org/abs/1904.08755).
## Change Log
- 2020-05-19 The latest Minkowski Engine since the commit [be5c3](https://github.com/StanfordVL/MinkowskiEngine/commit/be5c3c18b26d6a62380d613533b7a939a5458705), does not require explicit cache clear and can use the memory more efficiently.
- 2020-05-04: As pointed out by Thomas Chaton on [Issue#30](https://github.com/chrischoy/SpatioTemporalSegmentation/issues/30), I also found out that the training script contains bugs that models cannot reach the target performance described in the Model Zoo with the latest MinkowskiEngine. I am in the process of debugging the bugs, but I am having some difficulty finding the bugs. So, I created another git repo [SpatioTemporalSegmentation-ScanNet](https://github.com/chrischoy/SpatioTemporalSegmentation-ScanNet) from my other private repo that reaches the target performance. Please refer to the [SpatioTemporalSegmentation-ScanNet](https://github.com/chrischoy/SpatioTemporalSegmentation-ScanNet) for the ScanNet training. I'll update this repo once I find the bugs and merge SpatioTemporalSegmentation-ScanNet with this repo. Sorry for the trouble.
## Requirements
- Ubuntu 14.04 or higher
- CUDA 10.1 or higher
- pytorch 1.3 or higher
- python 3.6 or higher
- GCC 6 or higher
## Installation
You need to install `pytorch` and [`Minkowski Engine`][pypi-url] either with `pip` or with anaconda.
### Pip
The MinkowskiEngine is distributed via [PyPI MinkowskiEngine][pypi-url] which can be installed simply with `pip`.
First, install pytorch following the [instruction](https://pytorch.org). Next, install `openblas`.
```
sudo apt install libopenblas-dev
pip install torch torchvision
pip install -U git+https://github.com/StanfordVL/MinkowskiEngine
```
Next, clone the repository and install the rest of the requirements
```
git clone https://github.com/chrischoy/SpatioTemporalSegmentation/
cd SpatioTemporalSegmentation
pip install -r requirements.txt
```
### Troubleshooting
Please visit the MinkowskiEngine [issue pages](https://github.com/StanfordVL/MinkowskiEngine/) if you have difficulties installing Minkowski Engine.
## ScanNet Training
1. Download the ScanNet dataset from [the official website](http://kaldir.vc.in.tum.de/scannet_benchmark/documentation). You need to sign the terms of use.
2. Next, preprocess all scannet raw point cloud with the following command after you set the path correctly.
```
python -m lib.datasets.preprocessing.scannet
```
3. Train the network with
```
export BATCH_SIZE=N;
./scripts/train_scannet.sh 0 \
-default \
"--scannet_path /path/to/preprocessed/scannet"
```
Modify the `BATCH_SIZE` accordingly.
The first argument is the GPU id and the second argument is the path postfix
and the last argument is the miscellaneous arguments.
### mIoU vs. Overall Accuracy
The official evaluation metric for ScanNet is mIoU.
OA, Overal Accuracy is not the official metric since it is not discriminative. This is the convention from the 2D semantic segmentation as the pixelwise overall accuracy does not capture the fidelity of the semantic segmentation.
On 3D ScanNet semantic segmentation, OA: 89.087 -> mIOU 71.496 mAP 76.127 mAcc 79.660 on the ScanNet validation set v2.
Then why is the overall accuracy least discriminative metric? This is due to the fact that most of the scenes consist of large structures
such as walls, floors, or background and scores on these will dominate the statistics if you use Overall Accuracy.
## Synthia 4D Experiment
1. Download the dataset from [download](http://cvgl.stanford.edu/data2/Synthia4D.tar)
2. Extract
```
cd /path/to/extract/synthia4d
wget http://cvgl.stanford.edu/data2/Synthia4D.tar
tar -xf Synthia4D.tar
tar -xvjf *.tar.bz2
```
3. Training
```
export BATCH_SIZE=N; \
./scripts/train_synthia4d.sh 0 \
"-default" \
"--synthia_path /path/to/extract/synthia4d"
```
The above script trains a network. You have to change the arguments accordingly. The first argument to the script is the GPU id. Second argument is the log directory postfix; change to mark your experimental setup. The final argument is a series of the miscellaneous aruments. You have to specify the synthia directory here. Also, you have to wrap all arguments with " ".
## Stanford 3D Dataset
1. Download the stanford 3d dataset from [the website](http://buildingparser.stanford.edu/dataset.html)
2. Preprocess
Modify the input and output directory accordingly in
`lib/datasets/preprocessing/stanford.py`
And run
```
python -m lib.datasets.preprocessing.stanford
```
3. Train
```
./scripts/train_stanford.sh 0 \
"-default" \
"--stanford3d_path /PATH/TO/PREPROCESSED/STANFORD"
```
## Model Zoo
| Model | Dataset | Voxel Size | Conv1 Kernel Size | Performance | Link |
|:----------------:|:-------------------:|:----------:|:-----------------:|:-------------------------------------------------------------------------------:|:------:|
| Mink16UNet34C | ScanNet train + val | 2cm | 3 | Test set 73.6% mIoU, no sliding window | [download](https://node1.chrischoy.org/data/publications/minknet/Mink16UNet34C_ScanNet.pth) |
| Mink16UNet34C | ScanNet train | 2cm | 5 | Val 72.219% mIoU, no rotation average, no sliding window [per class performance](https://github.com/chrischoy/SpatioTemporalSegmentation/issues/13) | [download](https://node1.chrischoy.org/data/publications/minknet/MinkUNet34C-train-conv1-5.pth) |
| Mink16UNet18 | Stanford Area5 train | 5cm | 5 | Area 5 test 65.828% mIoU, no rotation average, no sliding window [per class performance](https://pastebin.com/Gj3PrPFr) | [download](https://node1.chrischoy.org/data/publications/minknet/Mink16UNet18-stanford-conv1-5.pth) |
| Mink16UNet34 | Stanford Area5 train | 5cm | 5 | Area 5 test 66.348% mIoU, no rotation average, no sliding window [per class performance](https://pastebin.com/WzhfGMQG) | [download](https://node1.chrischoy.org/data/publications/minknet/Mink16UNet34-stanford-conv1-5.pth) |
| 3D Mink16UNet14A | Synthia CVPR19 train | 15cm | 3 | CVPR19 test 81.903% mIoU, no rotation average, no sliding window [per class performance](https://pastebin.com/hN88iQqa) | [download](https://node1.chrischoy.org/data/publications/minknet/Mink16UNet14A-synthia-conv1-3.pth) |
| 3D Mink16UNet18 | Synthia CVPR19 train | 15cm | 3 | CVPR19 test 82.762% mIoU, no rotation average, no sliding window [per class performance](https://pastebin.com/zvt0yLFV) | [download](https://node1.chrischoy.org/data/publications/minknet/Mink16UNet18-synthia-conv1-3.pth) |
Note that sliding window style evaluation (cropping and stitching results) used in many related works effectively works as an ensemble (rotation averaging) which boosts the performance.
## Demo
The demo code will download weights and an example scene first and then visualize prediction results.
| Dataset | Scannet | Stanford |
|:--------:|:------------------------:|:-------------------------:|
| Command | `python -m demo.scannet` | `python -m demo.stanford` |
| Result |  |  |
## Citing this work
If you use the Minkowski Engine, please cite:
- [4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks, CVPR'19](https://arxiv.org/abs/1904.08755), [[pdf]](https://arxiv.org/pdf/1904.08755.pdf)
```
@inproceedings{choy20194d,
title={4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks},
author={Choy, Christopher and Gwak, JunYoung and Savarese, Silvio},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
pages={3075--3084},
year={2019}
}
```
## Related projects
- [Minkowski Engine, a neural network library for sparse tensors](https://github.com/StanfordVL/MinkowskiEngine)
- [Fully Convolutional Geometric Features, ICCV'19, fast and accurate 3D features](https://github.com/chrischoy/FCGF)