https://github.com/alihassanml/mma-fighting-object-detection-yolov8
https://github.com/alihassanml/mma-fighting-object-detection-yolov8
Last synced: about 2 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/alihassanml/mma-fighting-object-detection-yolov8
- Owner: alihassanml
- License: mit
- Created: 2025-03-07T21:12:43.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2025-03-07T21:22:38.000Z (7 months ago)
- Last Synced: 2025-07-28T02:37:46.834Z (2 months ago)
- Language: Python
- Size: 27.8 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# MMA Fighting Action Detection using YOLOv8
## 📌 Overview
This project utilizes **YOLOv8** for real-time **MMA fighting action detection**, classifying actions into:
- 🥋 **Grappling** (0)
- 🦵 **Kick** (1)
- ✊ **Punch** (2)
- 🧍 **Stand** (3)
It is trained on a custom dataset and can be used for **video analysis**, **sports analytics**, and **real-time fight detection**.
## 🚀 Features
- **Real-time action detection** from videos and images
- **Custom YOLOv8 model training** for MMA fighting
- **Fast and accurate predictions**
- Supports **live webcam inference**
## 📦 Installation
Clone the repository and install dependencies:
```bash
git clone https://github.com/alihassanml/MMA-Fighting-Object-Detection-YOLOv8.git
cd MMA-Fighting-Object-Detection-YOLOv8
pip install -r requirements.txt
```
## 🔧 Training the Model
To train YOLOv8 on your dataset, run:
```bash
yolo task=detect mode=train model=yolov8n.pt data=mma.yaml epochs=50 imgsz=640
```
Adjust parameters as needed for **epochs**, **batch size**, and **learning rate**.
## 🎯 Running Inference
To test on an image:
```bash
yolo task=detect mode=predict model=runs/detect/train/weights/best.pt source=your_image.jpg
```
For video:
```bash
yolo task=detect mode=predict model=runs/detect/train/weights/best.pt source=your_video.mp4
```
## 📊 Results & Performance
- Achieved **high accuracy** in action detection
- Works in **real-time on GPU**
- Supports **live video streaming**
## 🤖 Deployment
You can deploy this model using **FastAPI**, **Flask**, or integrate it into a **web application** for real-time fight analysis.
## 📌 Notebook
You can check the training and implementation details in this **[Kaggle Notebook](https://www.kaggle.com/code/alihassanml/mma-fighting-classfication-yolov8/notebook)**.
## 🙌 Contributing
Feel free to open issues or submit pull requests to improve this project!
## 📜 License
This project is **open-source** under the **MIT License**.
---
🔥 **Developed by [Ali Hassan](https://github.com/alihassanml)**
```