# OSL-ActionSpotting: A Unified Library for Action Spotting in Sports Videos
[](https://arxiv.org/abs/xxx.xxx)
[](https://github.com/SoccerNet/sn-spotting-pip/blob/main/LICENSE)
OSL-ActionSpotting is a plug-and-play library that unifies action
spotting algorithms.
## 🥳 What's New
- A technical report of this library will be provided soon.
## 📖 Major Features
- **Support SoTA TAD methods with modular design.** We decompose the TAD pipeline into different components, and implement them in a modular way. This design makes it easy to implement new methods and reproduce existing methods.
- **Support multiple datasets.** We support new datasets by giving a intermediate JSON format.
- **Support feature-based training and end-to-end training.** The feature-based training can easily be extended to end-to-end training with raw video input, and the video backbone can be easily replaced.
## 🌟 Model Zoo
| Feature based | End to end |
|:-------------:|:----------:|
| [AvgPool](https://arxiv.org/pdf/1804.04527.pdf) | [E2E-Spot](https://arxiv.org/pdf/2207.10213.pdf) |
| [MaxPool](https://arxiv.org/pdf/1804.04527.pdf) | |
| [NetVLAD](https://arxiv.org/pdf/1804.04527.pdf) | |
| [NetRVLAD](https://arxiv.org/pdf/1804.04527.pdf) | |
| [CALF](https://arxiv.org/pdf/1912.01326.pdf) | |
| [AvgPool++](https://arxiv.org/pdf/2104.06779.pdf) | |
| [MaxPool++](https://arxiv.org/pdf/2104.06779.pdf) | |
| [NetVLAD++](https://arxiv.org/pdf/2104.06779.pdf) | |
| [NetRVLAD++](https://arxiv.org/pdf/2104.06779.pdf)| |
## 🛠️ Installation
Please refer to [install.md](docs/install.md) for installation and data preparation.
## 🚀 Usage
Please refer to [usage.md](docs/usage.md) for details of training and evaluation scripts.
## 🤝 Roadmap
All the things that need to be done in the future is in [roadmap.md](docs/en/roadmap.md).
## 🖊️ Citation
If you think this repo is helpful, please cite us:
```bibtex
@misc{name,
title={},
author={},
howpublished = {\url{https://github.com/SoccerNet/sn-spotting-pip}},
year={2024}
}
```
If you have any questions, please contact: `yassine.benzakour@student.uliege.be`.
Raw data
{
"_id": null,
"home_page": "https://github.com/OpenSportsLab/OSL-ActionSpotting",
"name": "oslactionspotting",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": "SoccerNet, SDK, Spotting, Football, Soccer, Video",
"author": "['Silvio Giancola', 'Yassine Benzakour']",
"author_email": "['silvio.giancola@kaust.edu.sa', 'yassine.benzakour@student.uliege.be']",
"download_url": "https://files.pythonhosted.org/packages/a4/cf/83c95aad2a391ecd4bf744f916c251a580998568a44948eea17ff94c460a/oslactionspotting-0.1.2.tar.gz",
"platform": null,
"description": "# OSL-ActionSpotting: A Unified Library for Action Spotting in Sports Videos\n\n[](https://arxiv.org/abs/xxx.xxx)\n[](https://github.com/SoccerNet/sn-spotting-pip/blob/main/LICENSE)\n\nOSL-ActionSpotting is a plug-and-play library that unifies action\nspotting algorithms.\n\n## \ud83e\udd73 What's New\n\n- A technical report of this library will be provided soon.\n\n## \ud83d\udcd6 Major Features\n\n- **Support SoTA TAD methods with modular design.** We decompose the TAD pipeline into different components, and implement them in a modular way. This design makes it easy to implement new methods and reproduce existing methods.\n- **Support multiple datasets.** We support new datasets by giving a intermediate JSON format.\n- **Support feature-based training and end-to-end training.** The feature-based training can easily be extended to end-to-end training with raw video input, and the video backbone can be easily replaced.\n\n## \ud83c\udf1f Model Zoo\n\n| Feature based | End to end |\n|:-------------:|:----------:|\n| [AvgPool](https://arxiv.org/pdf/1804.04527.pdf) | [E2E-Spot](https://arxiv.org/pdf/2207.10213.pdf) |\n| [MaxPool](https://arxiv.org/pdf/1804.04527.pdf) | |\n| [NetVLAD](https://arxiv.org/pdf/1804.04527.pdf) | |\n| [NetRVLAD](https://arxiv.org/pdf/1804.04527.pdf) | |\n| [CALF](https://arxiv.org/pdf/1912.01326.pdf) | |\n| [AvgPool++](https://arxiv.org/pdf/2104.06779.pdf) | |\n| [MaxPool++](https://arxiv.org/pdf/2104.06779.pdf) | |\n| [NetVLAD++](https://arxiv.org/pdf/2104.06779.pdf) | |\n| [NetRVLAD++](https://arxiv.org/pdf/2104.06779.pdf)| |\n\n## \ud83d\udee0\ufe0f Installation\n\nPlease refer to [install.md](docs/install.md) for installation and data preparation.\n\n## \ud83d\ude80 Usage\n\nPlease refer to [usage.md](docs/usage.md) for details of training and evaluation scripts.\n\n## \ud83e\udd1d Roadmap\n\nAll the things that need to be done in the future is in [roadmap.md](docs/en/roadmap.md).\n\n## \ud83d\udd8a\ufe0f Citation\n\nIf you think this repo is helpful, please cite us:\n\n```bibtex\n@misc{name,\n title={},\n author={},\n howpublished = {\\url{https://github.com/SoccerNet/sn-spotting-pip}},\n year={2024}\n}\n```\n\nIf you have any questions, please contact: `yassine.benzakour@student.uliege.be`.\n\n\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Action Spotting SDK",
"version": "0.1.2",
"project_urls": {
"Homepage": "https://github.com/OpenSportsLab/OSL-ActionSpotting"
},
"split_keywords": [
"soccernet",
" sdk",
" spotting",
" football",
" soccer",
" video"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "73b1f3c3814ad5731012a3f6e093777cdd001cd093952e0ecdd6aa31c4b46283",
"md5": "b7075210e22c41121b87a57c303b091a",
"sha256": "55b52de700a96e39b249c3a88f71620ed09f528e2fd5e1338083d4d4419ff4b3"
},
"downloads": -1,
"filename": "oslactionspotting-0.1.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "b7075210e22c41121b87a57c303b091a",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 15186,
"upload_time": "2024-05-06T17:53:30",
"upload_time_iso_8601": "2024-05-06T17:53:30.692324Z",
"url": "https://files.pythonhosted.org/packages/73/b1/f3c3814ad5731012a3f6e093777cdd001cd093952e0ecdd6aa31c4b46283/oslactionspotting-0.1.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "a4cf83c95aad2a391ecd4bf744f916c251a580998568a44948eea17ff94c460a",
"md5": "ed727ee9af8e621a1f76c5324a7690c5",
"sha256": "19fabf02dcf49eb37f8a2b7fd63b9738739e0bb929b30373a676ac10cc732816"
},
"downloads": -1,
"filename": "oslactionspotting-0.1.2.tar.gz",
"has_sig": false,
"md5_digest": "ed727ee9af8e621a1f76c5324a7690c5",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 15768,
"upload_time": "2024-05-06T17:53:31",
"upload_time_iso_8601": "2024-05-06T17:53:31.797605Z",
"url": "https://files.pythonhosted.org/packages/a4/cf/83c95aad2a391ecd4bf744f916c251a580998568a44948eea17ff94c460a/oslactionspotting-0.1.2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-05-06 17:53:31",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "OpenSportsLab",
"github_project": "OSL-ActionSpotting",
"github_not_found": true,
"lcname": "oslactionspotting"
}
JSON
