README.md

---
pretty_name: SoccerHigh
license: cc-by-nc-sa-4.0
task_categories:
- video-classification
- feature-extraction
tags:
- soccer
- sports
- video
- video-summarization
- highlight-detection
homepage: https://ipcv.github.io/SoccerHigh/
repository: https://github.com/IPCV/SoccerHigh
---

# ⚽ SoccerHigh

This model card provides the checkpoints for the **Baseline Model** introduced in:

**SoccerHigh: A Benchmark Dataset for Automatic Soccer Video Summarization**  
[![arXiv](https://img.shields.io/badge/arXiv-2501.01234-b31b1b.svg)](https://arxiv.org/abs/2509.01439)
[![ACM DL](https://img.shields.io/badge/ACM-DL-blue)](https://dl.acm.org/doi/pdf/10.1145/3728423.3759410)    
[Artur Díaz-Juan](https://scholar.google.com/citations?user=WlPmWzwAAAAJ&hl=ca), [Coloma Ballester](https://scholar.google.com/citations?user=fLNi-SoAAAAJ&hl=ca), [Gloria Haro](https://scholar.google.com/citations?user=edEh3UMAAAAJ&hl=ca)  
[ACM MMSports 2025](http://mmsports.multimedia-computing.de/mmsports2025/cfp.html)

## 📂 Model Checkpoints

| Name | Description |
|------|-------------|
| Baseline_VideoMAEv2-Giant | Baseline model trained with a VideoMAEv2 encoder for feature extraction using [*vit_g_hybrid_pt_1200e_k710_ft*](https://huggingface.co/OpenGVLab/VideoMAE2/resolve/main/distill/vit_s_k710_dl_from_giant.pth) weights |
| Baseline_VideoMAEv2-Small | Baseline model trained with a VideoMAEv2 encoder for feature extraction using [*vit_g_hybrid_pt_1200e_k710_ft*](https://github.com/OpenGVLab/VideoMAEv2/blob/master/docs/MODEL_ZOO.md) weights |
| Baseline_CLIP | Baseline model trained with a CLIP encoder for feature extraction using [*CLIP ViT-B-32 256×256 trained on DataComp-1B*](https://huggingface.co/laion/CLIP-ViT-B-32-256x256-DataComp-s34B-b86K) weights |
| Baseline_ResNet | Baseline model trained with a ResNet-152 encoder for feature extraction using [*IMAGENET1K_V2*](https://docs.pytorch.org/vision/main/models/generated/torchvision.models.resnet152.html#torchvision.models.ResNet152_Weights) weights |

---

## ⚙️ Installation

```bash
git clone https://github.com/IPCV/SoccerHigh.git
cd SoccerHigh/code
```

For more details, please visit the original repository: 
[https://github.com/IPCV/SoccerHigh](https://github.com/IPCV/SoccerHigh)

---

## 📖 Citation

If you use this code for a scientific publication, please reference the original paper:

```bibtex
@inproceedings{10.1145/3728423.3759410,
  author = {D\'{\i}az-Juan, Artur and Ballester, Coloma and Haro, Gloria},
  title = {SoccerHigh: A Benchmark Dataset for Automatic Soccer Video Summarization},
  year = {2025},
  isbn = {9798400711985},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3728423.3759410},
  doi = {10.1145/3728423.3759410},
  booktitle = {Proceedings of the 8th International ACM Workshop on Multimedia Content Analysis in Sports},
  pages = {121–130},
  numpages = {10},
  location = {Dublin, Ireland},
  series = {MMSports '25}
}
```
---

## 🛡️ License

This code is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license.