TEAM_INSTRUCTIONS.md

# Highway Vehicle Detection - Team Instructions

## Quick Setup

1. **Clone the repository:**
   ```bash
   git clone https://huggingface.co/bichuche0705/highway-vehicle-detection-code
   cd highway-vehicle-detection-code
   ```

2. **Install dependencies:**
   ```bash
   pip install ultralytics opencv-python numpy
   ```

3. **Run the application:**
   ```bash
   python main.py
   ```

## External Resources

### Test Video
Watch the model in action: [YouTube Demo](https://www.youtube.com/watch?v=wqctLW0Hb_0&list=PLJKyZ_NuOhJQzif2-6-Kq9OiOj_UjJWvi)

### Main Dataset
Download the complete training dataset: [Kaggle Dataset](https://www.kaggle.com/datasets/sakshamjn/vehicle-detection-8-classes-object-detection/data)

## What's Included

- **Trained Models**: Ready-to-use YOLOv8m models
- **Source Code**: Complete detection application
- **Fine-tuning Dataset**: 92 images used for model improvement
- **Training Logs**: Performance metrics and visualizations
- **Documentation**: Complete project documentation

## Model Files

- `models/yolov8m_stage2_improved_best.pt` - Final model (recommended)
- `models/yolov8m_stage1_smart_best.pt` - Stage 1 model

## Usage Examples

### Basic Detection
```python
from ultralytics import YOLO
model = YOLO('models/yolov8m_stage2_improved_best.pt')
results = model('image.jpg')
results[0].show()
```

### Video Processing
```python
results = model('video.mp4', save=True)
```

## Support

For questions or issues, contact the project author.