🚛 HaulVision

AI-Powered Truck Detection • Real-Time • Highway & Logistics Optimized

A production-grade Streamlit application built around a fine-tuned YOLO object detection model optimized for real-time truck identification, achieving high precision across dense highway and logistics environments.

---

✨ Features

Feature	Description
📷 Image Detection	Upload images (JPG, PNG, BMP, WEBP) for instant truck detection
🎬 Video Detection	Process videos (MP4, AVI, MOV, MKV, WMV) frame-by-frame with progress tracking
⚡ Real-Time Inference	Fast YOLO-based detection with GPU acceleration support
🎯 Adjustable Thresholds	Fine-tune confidence & NMS IoU via interactive sliders
📊 Rich Analytics	Detection counts, confidence distribution, class breakdown, inference timing
🎨 Custom Bounding Boxes	Color-coded boxes with corner accents, semi-transparent fills, confidence labels
⬇️ Export Results	Download annotated images (PNG) and videos (MP4)
🤗 HF Model Hosting	Model auto-downloaded from Hugging Face Hub with caching
🌙 Premium Dark UI	Gradient headers, glassmorphism cards, hover animations

---

🚀 Quick Start

Prerequisites

• Python 3.10 or higher
• pip (Python package manager)
• GPU recommended (CUDA-compatible) for faster inference

Installation

# Clone the repository
git clone https://github.com/hamidraiz786/HaulVision.git
cd HaulVision

# Install dependencies
pip install -r requirements.txt

Run the App

streamlit run app.py

The app will open at http://localhost:8501 in your default browser.

Note: On first launch, the model (~51MB) will be automatically downloaded from Hugging Face and cached locally.

---

📁 Project Structure

HaulVision/
├── app.py                 # Main Streamlit application (single-file)
├── kvision5.pt            # Fine-tuned YOLO model weights
├── requirements.txt       # Python dependencies
├── README.md              # This file
├── assets/
│   └── banner.jpg         # Repository banner image
└── .streamlit/
    └── config.toml        # Streamlit theme configuration

---

🖥️ Usage

Image Detection

1. Select 📷 Image mode in the sidebar
2. Upload an image (JPG, PNG, BMP, or WEBP)
3. Adjust Confidence and NMS IoU thresholds as needed
4. View results in the Detection Result, Original Image, and Details tabs
5. Click ⬇️ Download Annotated Image to save results

Video Detection

1. Select 🎬 Video mode in the sidebar
2. Upload a video (MP4, AVI, MOV, MKV, or WMV)
3. Adjust Frame Skip to control processing speed (higher = faster)
4. Click 🚀 Run Detection to begin processing
5. Monitor real-time progress and frame previews
6. Download the annotated video when processing completes

---

⚙️ Configuration

Sidebar Controls

Control	Range	Default	Description
Confidence Threshold	0.05 – 1.0	0.25	Minimum detection confidence score
NMS IoU Threshold	0.05 – 1.0	0.45	Non-Maximum Suppression overlap threshold
Frame Skip (Video)	1 – 10	2	Process every N-th frame for speed

Theme Customization

Edit .streamlit/config.toml to customize the app's appearance:

[theme]
primaryColor = "#00D4FF"
backgroundColor = "#0E1117"
secondaryBackgroundColor = "#161B22"
textColor = "#E6EDF3"

---

🧠 Model Details

Property	Value
Architecture	YOLO (Fine-tuned)
Task	Object Detection
Specialization	Truck identification in highway & logistics
Model Size	~51 MB
Input Resolution	640×640
Hosted On	🤗 Hugging Face

---

📦 Dependencies

streamlit>=1.38.0
ultralytics>=8.2.0
opencv-python-headless>=4.9.0
Pillow>=10.0.0
numpy>=1.24.0
torch>=2.0.0
torchvision>=0.15.0
huggingface-hub>=0.23.0

---

🐳 Docker Deployment (Optional)

FROM python:3.11-slim

WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

COPY . .
EXPOSE 8501

HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health
ENTRYPOINT ["streamlit", "run", "app.py", "--server.port=8501", "--server.address=0.0.0.0"]

docker build -t haulvision .
docker run -p 8501:8501 haulvision

---

🤝 Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

---

📄 License

This project is licensed under the MIT License — see the LICENSE file for details.

---

Built with ❤️ using Streamlit & Ultralytics YOLO
Model hosted on 🤗 Hugging Face