[](https://github.com/open-mmlab/mmaction2/issues)
[](https://github.com/open-mmlab/mmaction2/issues)
-**Modular design**: We decompose a video understanding framework into different components. One can easily construct a customized video understanding framework by combining different modules.
根据模型情况填写表格：
| 卡数 | 性能 | Top5_acc |
| :------: | :------: |:------: |
| 1 | 74.77 samples/s(bs=16) | 1.0000 |
-**Support four major video understanding tasks**: MMAction2 implements various algorithms for multiple video understanding tasks, including action recognition, action localization, spatio-temporal action detection, and skeleton-based action detection. We support **27** different algorithms and **20** different datasets for the four major tasks.
### 参考
-**Well tested and documented**: We provide detailed documentation and API reference, as well as unit tests.
https://github.com/open-mmlab/mmdetection
## What's New
- (2022-03-04) We support **Multigrid** on Kinetics400, achieve 76.07% Top-1 accuracy and accelerate training speed.
- (2021-11-24) We support **2s-AGCN** on NTU60 XSub, achieve 86.06% Top-1 accuracy on joint stream and 86.89% Top-1 accuracy on bone stream respectively.
- (2021-10-29) We provide a demo for skeleton-based and rgb-based spatio-temporal detection and action recognition (demo/demo_video_structuralize.py).
- (2021-10-26) We train and test **ST-GCN** on NTU60 with 3D keypoint annotations, achieve 84.61% Top-1 accuracy (higher than 81.5% in the [paper](https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/viewPaper/17135)).
- (2021-10-25) We provide a script(tools/data/skeleton/gen_ntu_rgbd_raw.py) to convert the NTU60 and NTU120 3D raw skeleton data to our format.
- (2021-10-25) We provide a [guide](https://github.com/open-mmlab/mmaction2/blob/master/configs/skeleton/posec3d/custom_dataset_training.md) on how to train PoseC3D with custom datasets, [bit-scientist](https://github.com/bit-scientist) authored this PR!
- (2021-10-16) We support **PoseC3D** on UCF101 and HMDB51, achieves 87.0% and 69.3% Top-1 accuracy with 2D skeletons only. Pre-extracted 2D skeletons are also available.
**Release**: v0.24.0 was released in 05/05/2022. Please refer to [changelog.md](docs/changelog.md) for details and release history.
## Installation
MMAction2 depends on [PyTorch](https://pytorch.org/), [MMCV](https://github.com/open-mmlab/mmcv), [MMDetection](https://github.com/open-mmlab/mmdetection)(optional), and [MMPose](https://github.com/open-mmlab/mmdetection)(optional).
Below are quick steps for installation.
Please refer to [install.md](docs/install.md) for more detailed instruction.
A Colab tutorial is also provided. You may preview the notebook [here](demo/mmaction2_tutorial.ipynb) or directly [run](https://colab.research.google.com/github/open-mmlab/mmaction2/blob/master/demo/mmaction2_tutorial.ipynb) on Colab.
<td><ahref="https://github.com/open-mmlab/mmaction2/blob/master/tools/data/mit/README.md">Moments in Time</a> (<ahref="http://moments.csail.mit.edu/">Homepage</a>) (TPAMI'2019)</td>
<td><ahref="https://github.com/open-mmlab/mmaction2/blob/master/tools/data/mmit/README.md">Multi-Moments in Time</a> (<ahref="http://moments.csail.mit.edu/challenge_iccv_2019.html">Homepage</a>) (ArXiv'2019)</td>
Datasets marked with * are not fully supported yet, but related dataset preparation steps are provided. A summary can be found on the [**Supported Datasets**](https://mmaction2.readthedocs.io/en/latest/supported_datasets.html) page.
## Benchmark
To demonstrate the efficacy and efficiency of our framework, we compare MMAction2 with some other popular frameworks and official releases in terms of speed. Details can be found in [benchmark](docs/benchmark.md).
## Data Preparation
Please refer to [data_preparation.md](docs/data_preparation.md) for a general knowledge of data preparation.
The supported datasets are listed in [supported_datasets.md](docs/supported_datasets.md)
## FAQ
Please refer to [FAQ](docs/faq.md) for frequently asked questions.
## Projects built on MMAction2
Currently, there are many research works and projects built on MMAction2 by users from community, such as:
- Video Swin Transformer. [\[paper\]](https://arxiv.org/abs/2106.13230)[\[github\]](https://github.com/SwinTransformer/Video-Swin-Transformer)
- Evidential Deep Learning for Open Set Action Recognition, ICCV 2021 **Oral**. [\[paper\]](https://arxiv.org/abs/2107.10161)[\[github\]](https://github.com/Cogito2012/DEAR)
- Rethinking Self-supervised Correspondence Learning: A Video Frame-level Similarity Perspective, ICCV 2021 **Oral**. [\[paper\]](https://arxiv.org/abs/2103.17263)[\[github\]](https://github.com/xvjiarui/VFS)
etc., check [projects.md](docs/projects.md) to see all related projects.
## Contributing
We appreciate all contributions to improve MMAction2. Please refer to [CONTRIBUTING.md](https://github.com/open-mmlab/mmcv/blob/master/CONTRIBUTING.md) in MMCV for more details about the contributing guideline.
## Acknowledgement
MMAction2 is an open-source project that is contributed by researchers and engineers from various colleges and companies.
We appreciate all the contributors who implement their methods or add new features and users who give valuable feedback.
We wish that the toolbox and benchmark could serve the growing research community by providing a flexible toolkit to reimplement existing methods and develop their new models.
## Citation
If you find this project useful in your research, please consider cite:
```BibTeX
@misc{2020mmaction2,
title={OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark},
[](https://github.com/open-mmlab/mmaction2/issues)
[](https://github.com/open-mmlab/mmaction2/issues)
<pstyle="font-size:1.5vw;">Spatio-Temporal Action Detection Results on AVA-2.1</p>
</div>
## Major Features
-**Modular design**: We decompose a video understanding framework into different components. One can easily construct a customized video understanding framework by combining different modules.
-**Support four major video understanding tasks**: MMAction2 implements various algorithms for multiple video understanding tasks, including action recognition, action localization, spatio-temporal action detection, and skeleton-based action detection. We support **27** different algorithms and **20** different datasets for the four major tasks.
-**Well tested and documented**: We provide detailed documentation and API reference, as well as unit tests.
## What's New
- (2022-03-04) We support **Multigrid** on Kinetics400, achieve 76.07% Top-1 accuracy and accelerate training speed.
- (2021-11-24) We support **2s-AGCN** on NTU60 XSub, achieve 86.06% Top-1 accuracy on joint stream and 86.89% Top-1 accuracy on bone stream respectively.
- (2021-10-29) We provide a demo for skeleton-based and rgb-based spatio-temporal detection and action recognition (demo/demo_video_structuralize.py).
- (2021-10-26) We train and test **ST-GCN** on NTU60 with 3D keypoint annotations, achieve 84.61% Top-1 accuracy (higher than 81.5% in the [paper](https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/viewPaper/17135)).
- (2021-10-25) We provide a script(tools/data/skeleton/gen_ntu_rgbd_raw.py) to convert the NTU60 and NTU120 3D raw skeleton data to our format.
- (2021-10-25) We provide a [guide](https://github.com/open-mmlab/mmaction2/blob/master/configs/skeleton/posec3d/custom_dataset_training.md) on how to train PoseC3D with custom datasets, [bit-scientist](https://github.com/bit-scientist) authored this PR!
- (2021-10-16) We support **PoseC3D** on UCF101 and HMDB51, achieves 87.0% and 69.3% Top-1 accuracy with 2D skeletons only. Pre-extracted 2D skeletons are also available.
**Release**: v0.24.0 was released in 05/05/2022. Please refer to [changelog.md](docs/changelog.md) for details and release history.
## Installation
MMAction2 depends on [PyTorch](https://pytorch.org/), [MMCV](https://github.com/open-mmlab/mmcv), [MMDetection](https://github.com/open-mmlab/mmdetection)(optional), and [MMPose](https://github.com/open-mmlab/mmdetection)(optional).
Below are quick steps for installation.
Please refer to [install.md](docs/install.md) for more detailed instruction.
A Colab tutorial is also provided. You may preview the notebook [here](demo/mmaction2_tutorial.ipynb) or directly [run](https://colab.research.google.com/github/open-mmlab/mmaction2/blob/master/demo/mmaction2_tutorial.ipynb) on Colab.
<td><ahref="https://github.com/open-mmlab/mmaction2/blob/master/tools/data/mit/README.md">Moments in Time</a> (<ahref="http://moments.csail.mit.edu/">Homepage</a>) (TPAMI'2019)</td>
<td><ahref="https://github.com/open-mmlab/mmaction2/blob/master/tools/data/mmit/README.md">Multi-Moments in Time</a> (<ahref="http://moments.csail.mit.edu/challenge_iccv_2019.html">Homepage</a>) (ArXiv'2019)</td>
Datasets marked with * are not fully supported yet, but related dataset preparation steps are provided. A summary can be found on the [**Supported Datasets**](https://mmaction2.readthedocs.io/en/latest/supported_datasets.html) page.
## Benchmark
To demonstrate the efficacy and efficiency of our framework, we compare MMAction2 with some other popular frameworks and official releases in terms of speed. Details can be found in [benchmark](docs/benchmark.md).
## Data Preparation
Please refer to [data_preparation.md](docs/data_preparation.md) for a general knowledge of data preparation.
The supported datasets are listed in [supported_datasets.md](docs/supported_datasets.md)
## FAQ
Please refer to [FAQ](docs/faq.md) for frequently asked questions.
## Projects built on MMAction2
Currently, there are many research works and projects built on MMAction2 by users from community, such as:
- Video Swin Transformer. [\[paper\]](https://arxiv.org/abs/2106.13230)[\[github\]](https://github.com/SwinTransformer/Video-Swin-Transformer)
- Evidential Deep Learning for Open Set Action Recognition, ICCV 2021 **Oral**. [\[paper\]](https://arxiv.org/abs/2107.10161)[\[github\]](https://github.com/Cogito2012/DEAR)
- Rethinking Self-supervised Correspondence Learning: A Video Frame-level Similarity Perspective, ICCV 2021 **Oral**. [\[paper\]](https://arxiv.org/abs/2103.17263)[\[github\]](https://github.com/xvjiarui/VFS)
etc., check [projects.md](docs/projects.md) to see all related projects.
## Contributing
We appreciate all contributions to improve MMAction2. Please refer to [CONTRIBUTING.md](https://github.com/open-mmlab/mmcv/blob/master/CONTRIBUTING.md) in MMCV for more details about the contributing guideline.
## Acknowledgement
MMAction2 is an open-source project that is contributed by researchers and engineers from various colleges and companies.
We appreciate all the contributors who implement their methods or add new features and users who give valuable feedback.
We wish that the toolbox and benchmark could serve the growing research community by providing a flexible toolkit to reimplement existing methods and develop their new models.
## Citation
If you find this project useful in your research, please consider cite:
```BibTeX
@misc{2020mmaction2,
title={OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark},
