Wan 2.0 Video Generation Project

Generate stunning AI-powered videos from text prompts or images using the Wan 2.0 model by Alibaba.

✨ Features

• 🎬 Text-to-Video Generation - Create videos from descriptive text prompts
• 🖼️ Image-to-Video Animation - Bring static images to life with AI
• 🌐 Modern Web Interface - Beautiful, responsive UI with glassmorphism design
• ⚡ Real-time Processing - Fast video generation with progress tracking
• 📱 Mobile Responsive - Works seamlessly on all devices
• 🎨 Customizable Settings - Control resolution, FPS, and duration

🚀 Quick Start

Prerequisites

• Python 3.8 or higher
• CUDA-compatible GPU (recommended) or CPU
• 8GB+ RAM (16GB+ recommended)

Installation

1. Clone or navigate to the project directory:

cd d:\Freelancing\ImagetoVideo

2. Create a virtual environment:

python -m venv venv

3. Activate the virtual environment:

Windows:

venv\Scripts\activate

Linux/Mac:

source venv/bin/activate

4. Install dependencies:

pip install -r requirements.txt

5. Set up environment variables:

copy .env.example .env

Edit .env file to configure your settings (optional).

Running the Application

1. Start the Flask server:

python app.py

2. Open your browser and navigate to:

http://localhost:5000

3. Start creating videos!

📖 Usage

Text-to-Video

1. Select the Text to Video tab
2. Enter a descriptive prompt (e.g., "A majestic dragon flying over a medieval castle at sunset")
3. Optionally add a negative prompt to avoid unwanted elements
4. Configure settings (duration, FPS, resolution)
5. Click Generate Video
6. Wait for processing and download your video!

Image-to-Video

1. Select the Image to Video tab
2. Upload an image (PNG, JPG, JPEG, or WEBP)
3. Optionally add a motion prompt (e.g., "gentle camera zoom")
4. Configure settings
5. Click Generate Video
6. Download your animated video!

🎨 Example Prompts

Text-to-Video Examples:

• "A serene lake at sunrise with mist rising from the water, cinematic 4k"
• "Futuristic city with flying cars and neon lights, cyberpunk style"
• "Ocean waves crashing on a rocky shore, slow motion"
• "Northern lights dancing in the night sky over snowy mountains"

Image-to-Video Motion Prompts:

• "Slow zoom in with gentle camera movement"
• "Pan from left to right smoothly"
• "Character walking forward naturally"
• "Leaves rustling in the wind"

⚙️ Configuration

Edit the .env file to customize:

# Model Configuration
MODEL_NAME=alibaba-pai/wan-2.0-5b
DEVICE=cuda  # Options: cuda, cpu, mps
USE_FP16=True

# Generation Settings
DEFAULT_RESOLUTION=720
DEFAULT_FPS=24
DEFAULT_DURATION=5
MAX_VIDEO_LENGTH=10

# Server Settings
HOST=0.0.0.0
PORT=5000

🏗️ Project Structure

ImagetoVideo/
├── app.py                 # Flask backend server
├── model_handler.py       # Wan 2.0 model integration
├── requirements.txt       # Python dependencies
├── .env.example          # Environment configuration template
├── .gitignore            # Git ignore rules
├── static/               # Frontend files
│   ├── index.html        # Main web interface
│   ├── css/
│   │   └── style.css     # Styling with glassmorphism
│   └── js/
│       └── app.js        # Client-side JavaScript
├── uploads/              # Temporary image uploads
└── outputs/              # Generated videos

🔧 API Endpoints

Health Check

GET /api/health

Text-to-Video

POST /api/text-to-video
Content-Type: application/json

{
  "prompt": "Your prompt here",
  "negative_prompt": "Optional",
  "duration": 5,
  "fps": 24,
  "resolution": 720
}

Image-to-Video

POST /api/image-to-video
Content-Type: multipart/form-data

image: <file>
prompt: "Optional motion description"
duration: 5
fps: 24
resolution: 720

Download Video

GET /api/download/<filename>

🎯 System Requirements

Minimum Requirements:

• CPU: 4+ cores
• RAM: 8GB
• Storage: 10GB free space
• GPU: Not required (CPU mode available)

Recommended Requirements:

• CPU: 8+ cores
• RAM: 16GB+
• Storage: 20GB+ free space
• GPU: NVIDIA GPU with 8GB+ VRAM (RTX 3060 or better)

🐛 Troubleshooting

Model Loading Issues

If the model fails to load, the application will run in mock mode for demonstration purposes. To use the actual Wan 2.0 model:

1. Ensure you have sufficient GPU memory
2. Try setting USE_FP16=True in .env to reduce memory usage
3. Consider using a smaller model variant
4. Check your internet connection for model downloads

CUDA Out of Memory

• Reduce resolution in settings
• Decrease video duration
• Set USE_FP16=True
• Close other GPU-intensive applications

Slow Generation

• Use GPU instead of CPU (set DEVICE=cuda)
• Reduce resolution and FPS
• Enable half-precision (USE_FP16=True)

📝 Notes

• First run will download the Wan 2.0 model (~10GB), which may take time
• Video generation time depends on duration, resolution, and hardware
• Generated videos are saved in the outputs/ directory
• Uploaded images are temporarily stored and automatically cleaned up

🌟 Technologies Used

• Backend: Flask, Python
• AI/ML: PyTorch, Hugging Face Diffusers, Wan 2.0
• Frontend: HTML5, CSS3, Vanilla JavaScript
• Video Processing: OpenCV, imageio

📄 License

This project is for educational and demonstration purposes. Please refer to the Wan 2.0 model license for commercial usage restrictions.

🤝 Contributing

Contributions are welcome! Feel free to submit issues or pull requests.

🔗 Resources

• Wan 2.0 Model on Hugging Face
• Wan AI Official Website
• Flask Documentation
• PyTorch Documentation

---

Made with ❤️ using Wan 2.0 AI