Sora2-Watermark-Remover

KO | EN

gitlite — search

#python #ai #typescript #machine-learning #deep-learning #docker #macos #awesome #kubernetes #vue #awesome-list #pytorch #ios #mcp

Sora2-Watermark-Remover

(★ 23)

Sora2 and any video Watermark-Remover

Sora2-Watermark-Remover 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Video Inpainting Tool

An open-source tool for removing watermarks and unwanted objects from videos using AI-powered inpainting.

> 🚀 **Don't want to set it up locally?** Use our hosted service at [watermark.center](https://watermark.center) - No technical knowledge required!

## ⚡ Quick Start Options

### Option 1: Use Our Hosted Service (Recommended for Non-Technical Users)
Visit [watermark.center](https://watermark.center) and start removing watermarks immediately. No installation, no setup, just upload and process!

### Option 2: Self-Host (Free & Open Source)
Follow the installation instructions below to run it on your own machine.

---

## Features

- 🎥 Upload and process videos
- 🎯 Select rectangular areas for inpainting
- 🤖 Automatic segment detection
- 🎨 AI-powered video inpainting using Replicate (SAM 2 + ProPainter)
- 📦 Video gallery to view and download processed videos
- 🔊 Audio preservation with FFmpeg
- 💾 Simple file-based storage (no database required!)

## Technology Stack

**Backend:**
- FastAPI (Python)
- JSON file storage (no database!)
- Replicate API (SAM 2 Video + ProPainter)
- OpenCV
- FFmpeg

**Frontend:**
- React + TypeScript
- Vite
- Axios

## Prerequisites

- Python 3.9+
- Node.js 16+
- FFmpeg
- Replicate API Token ([Get one free](https://replicate.com/account/api-tokens))

## Installation

### Backend Setup

1. Navigate to backend directory:
```bash
cd backend
```

2. Create virtual environment:
```bash
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
```

3. Install dependencies:
```bash
pip install -r requirements.txt
```

4. Configure environment variables:
```bash
cp .env.example .env
```

Edit `.env` with your configuration:
- `REPLICATE_API_TOKEN`: Your Replicate API token ([Get it here](https://replicate.com/account/api-tokens))
- `NGROK_URL`: Your backend URL (default: `http://localhost:8080`)

5. Run the backend:
```bash
python main.py
```

The API will be available at `http://localhost:8080`

### Frontend Setup

1. Navigate to frontend directory:
```bash
cd frontend
```

2. Install dependencies:
```bash
npm install
```

3. Configure API URL (optional):
```bash
cp .env.example .env
```

The frontend is already configured to connect to `http://localhost:8080`

4. Run the development server:
```bash
npm run dev
```

The frontend will be available at `http://localhost:3000`

## Usage

1. **Upload Video**: Click the upload area and select a video file
2. **Draw Rectangles**: Click on the video canvas to draw rectangular areas you want to inpaint
3. **Process Video**: Click "Process Video" to start the inpainting process
4. **Download**: Once processing is complete, download your processed video
5. **View Gallery**: All processed videos appear in the gallery on the main page

## Data Storage

All data is stored locally on your disk:
- **Videos**: Stored in `backend/outputs/`
- **Metadata**: Stored in `backend/outputs/videos.json`
- **Uploads**: Stored in `backend/uploads/`

No database required! Everything is stored in simple JSON files.

## API Endpoints

- `POST /api/detect-segments` - Detect segments in video
- `POST /api/process-video` - Process video with inpainting
- `GET /api/videos` - Get all processed videos
- `DELETE /api/videos/{filename}` - Delete a processed video
- `GET /health` - Health check

## How It Works

1. **Segment Detection**: Analyzes the video to detect when selected rectangular areas are visible
2. **Mask Generation**: Creates masks for the selected areas
3. **Inpainting**: Applies ProPainter to remove the selected areas and fill them intelligently
4. **Audio Merging**: Preserves and merges the original audio using FFmpeg
5. **Storage**: Saves the processed video and metadata to disk

## Requirements

### Replicate API
This project uses Replicate's AI models. You'll need:
- A Replicate account (free to start)
- API token from https://replicate.com/account/api-tokens
- Credits for API usage (pay-as-you-go)

**Note**: If you don't want to manage Replicate credits yourself, use our hosted service at [watermark.center](https://watermark.center)

### FFmpeg
Required for audio processing. Install it:
- **macOS**: `brew install ffmpeg`
- **Ubuntu**: `sudo apt install ffmpeg`
- **Windows**: Download from [ffmpeg.org](https://ffmpeg.org/download.html)

## Project Structure

```
SoraMarkRemover/
├── backend/
│   ├── main.py              # FastAPI application
│   ├── storage.py           # File-based storage system
│   ├── video_processor.py   # Video processing logic
│   ├── segment_detector.py  # Segment detection
│   ├── models.py            # Data models
│   ├── requirements.txt     # Python dependencies
│   ├── uploads/             # Uploaded videos (temporary)
│   └── outputs/             # Processed videos + metadata
│       └── videos.json      # Video metadata
├── frontend/
│   ├── src/
│   │   ├── components/      # React components
│   │   ├── pages/           # Pages
│   │   └── ...
│   └── package.json
└── README.md
```

## Why Open Source?

We believe in transparency and community collaboration. By open-sourcing this tool, we enable:
- 🔍 **Transparency**: See exactly how your videos are processed
- 🛠️ **Customization**: Modify the code to fit your specific needs
- 🌍 **Community**: Contribute improvements and features
- 💰 **Cost Control**: Run it yourself if you prefer

**But we also offer a hosted service** at [watermark.center](https://watermark.center) for those who:
- Don't have technical expertise
- Want a hassle-free experience
- Prefer not to manage API credits and infrastructure

Both options use the same technology - you choose what works best for you!

## Contributing

Contributions are welcome! Please see [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.

## License

MIT License - feel free to use this project for any purpose. See [LICENSE](LICENSE) for details.

## Support

- **Self-hosted issues**: Open an issue on GitHub
- **Hosted service**: Contact support at [watermark.center](https://watermark.center)

## Acknowledgments

- [SAM 2 Video](https://replicate.com/zsxkib/sam-2-video) for segmentation
- [ProPainter](https://replicate.com/zsxkib/propainter) for video inpainting
- Built with FastAPI and React
- Hosted service available at [watermark.center](https://watermark.center)

---

**Remember**: You can use this code for free, or skip the setup and use [watermark.center](https://watermark.center) directly! 🚀

Sora2-Watermark-Remover

Sora2-Watermark-Remover 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Sora2-Watermark-Remover

Sora2-Watermark-Remover 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README