VidMark

Digital video watermarking tool using DWT-DCT with QIM method.

---

📋 Table of Contents

• About
• Features
• Screenshots
• Requirements
• Installation
• Quick Start
• Algorithm
• Project Structure
• Configuration
• Testing
• Build EXE
• Cleaning Scripts
• FAQ
• Contributing
• Acknowledgements

📖 About

VidMark is a desktop application for embedding digital watermarks into video files. It utilizes a combination of Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT) with Quantization Index Modulation (QIM) to hide digital signatures that are robust against compression, noise, and filtering.

This project was created as part of a university elective course.

✨ Features

🔒 Embedding

• DWT-DCT-QIM — Embedding in the frequency domain within the LL subband DCT coefficients.
• Adjustable Intensity — Presets: Invisible / Balanced / Robust, or manual delta input.
• Unique Seed — XOR of base seed + UUID + timestamp for enhanced security.
• Automatic Key Generation — A JSON key file is mandatory for extraction.

🔍 Extraction and Verification

• Majority Voting — Extraction from N uniformly selected frames.
• Reed-Solomon Codes — Data recovery from corrupted bits.
• CRC-16 Verification — Integrity validation after extraction.
• SYNC Marker — Confirmation of watermark presence before parsing.

🧪 Attack Resistance Testing

• 30+ Attack Types — Compression, noise, filtering, geometry, color transformations.
• H.264/H.265 Simulation — Real encoding/decoding via FFmpeg.
• NC/BER/PSNR Metrics — Calculated for every attack.
• CSV Export — Save results for analysis.

📊 Quality Metrics

• PSNR — Peak Signal-to-Noise Ratio.
• SSIM — Structural Similarity Index.
• NC — Normalized Correlation.
• BER — Bit Error Rate.

🎨 Interface

• macOS-style Light Theme — Clean, modern QSS styling.
• Real-time Preview — Side-by-side comparison of original and watermarked frames.
• Detailed Logging — Operation log with timestamps and export capability.
• Settings Dialog — Wavelet/level/block/delta/CRF settings with compatibility tooltips.

📸 Screenshots

Click to expand gallery

📥 Embedding	⚙️ Settings

✅ Verification	📤 Extraction

🧪 Attack Test	📋 Log

📋 Requirements

Component	Version	Purpose
Python	3.9+	Runtime environment.
FFmpeg	4.4+	Video encoding/decoding, audio copying, A/V tag normalization, compression attack simulation (H.264/H.265).
ffprobe	(w/ FFmpeg)	Extracting video metadata (audio streams, format, duration).
ffplay	(w/ FFmpeg)	Video preview via ffplay (optional, right-click on preview).

⚠️ FFmpeg is mandatory. VidMark requires ffmpeg, ffprobe, and ffplay in the system PATH. The application will show an error and exit if FFmpeg is not found.

Installing FFmpeg:

Platform	Command
Windows	Download from ffmpeg.org, add to PATH.
macOS	brew install ffmpeg
Ubuntu	sudo apt install ffmpeg
Arch	sudo pacman -S ffmpeg

🚀 Installation

git clone https://github.com/qexela/VidMark.git
cd VidMark

python -m venv venv
source venv/bin/activate      # Linux/macOS
venv\Scripts\activate         # Windows

pip install -r requirements.txt

⚡ Quick Start

1. Launch:

   ffmpeg -version # check ffmpeg installation
   python main.py

2. Embedding: Go to "Embedding" tab → Select Video → Enter Text → "Start" → Choose save path. Wait for processing — metrics will appear on the preview panel.
3. Key: ⚠️ Save the .json key file! Extraction is impossible without it.
4. Extraction: Go to "Extraction" tab → Select Watermarked Video → Select Key File → "Extract & Verify". Result: ✅ CONFIRMED / ⚠️ MISMATCH / ❌ NOT FOUND.

🔬 Algorithm

graph LR
    Input(Input Frame) --> DWT[DWT Haar]
    DWT --> LL[LL Subband]
    LL --> DCT[DCT 8x8]
    DCT --> QIM{Embed QIM}
    QIM --> IDWT[IDWT]
    IDWT --> Output(Output Frame)

Loading

The key file stores exact embedding parameters, including the scrambling seed. Without it, recovering the watermark is impossible.

📂 Project Structure

📂 Expand file tree

VidMark/
├── 🚀 main.py                     # Entry point
├── ⚙️ config.py                   # Global configuration
├── 🌐 i18n.py                     # Internationalization (EN/RU)
├── 📋 requirements.txt
├── 📖 README.md
├── 📖 README.ru.md
├── 📜 LICENSE                     # GPLv3
├── 🙈 .gitignore
│
├── 📁 core/                       # Watermarking algorithms
│   ├── 🔧 embedder.py             # DWT-DCT-QIM Embedding
│   ├── 🔍 extractor.py            # DWT-DCT-QIM Extraction
│   ├── 🛡️ ecc.py                  # Reed-Solomon codes
│   ├── 🔀 scrambler.py            # Bit scrambling
│   ├── 📦 payload.py              # Payload formation and parsing
│   ├── 📊 metrics.py              # PSNR, SSIM, NC, BER
│   ├── 🧪 attacks.py              # Attack simulator
│   └── 🔑 keyfile.py              # Key file management
│
├── 📁 ui/                         # PyQt5 Interface
│   ├── 🏠 main_window.py
│   ├── 📥 embed_tab.py
│   ├── 📤 extract_tab.py
│   ├── 🧪 attack_tab.py
│   ├── 📋 log_tab.py
│   └── ⚙️ settings_dialog.py
│
├── 📁 workers/                    # Background threads
│   └── 🔄 video_worker.py
│
├── 📁 utils/                      # Utilities
│   ├── 🎬 video_utils.py          # Video I/O, FFmpeg
│   └── 🖼️ image_utils.py          # Image conversion
│
├── 📁 assets/
│   ├── 🎨 icon.ico
│   └── 🎨 style.qss
│
├── 📁 scripts/
│   ├── 🔨 builder.py
│   └── 🧹 clean.bat / clean.sh / clean.ps1
│
├── 📁 tests/
│   └── 🧪 test_*.py (10 modules, 30+ tests)
│
└── 📁 screenshots/

⚙️ Configuration

Intensity Presets

Preset	Delta	PSNR (Typical)	Robustness	Application
Invisible	20.0	> 48 dB	Lower	Proof of authorship
Balanced	35.0	~44-46 dB	Good	General use (Default)
Robust	55.0	~40-43 dB	Maximum	Aggressive conditions

Wavelet Compatibility

Wavelet	Level 1	Level 2	Min. Delta
haar	✅ Excellent	⚠️ Questionable	25+
db2	✅ Excellent	⚠️ Questionable	30+
db4	⚠️ Questionable	⚠️ Questionable	40+
bior4.4	⚠️ Questionable	✅ Excellent	30+
coif2	❌ Poor	❌ Poor	60+

🧪 Testing

pytest tests/ -v
python tests/test_cli.py

📦 Build EXE

python scripts/build.py              # Interactive menu
python scripts/build.py --build      # Direct build
python scripts/build.py --build --console  # With console window

🧹 Cleaning Scripts

Scripts to clean the repository from temporary files (caches, build artifacts, etc.).

./scripts/clean.sh          # Linux/macOS
scripts\clean.bat           # Windows CMD
scripts\clean.ps1           # Windows PowerShell

❓ FAQ

Q: Why can't I extract the watermark? A: Incorrect key file, excessive compression (CRF > 28), or geometric distortions (rotation > 2 deg).

Q: Can I embed a watermark into audio? A: No, VidMark works only with the video stream. Audio is copied unchanged.

Q: What happens if I lose the key file? A: The watermark cannot be recovered without the key.

🤝 Contributing

This project was created as a university assignment and is open for educational contributions.

1. Fork the repository
2. Create a branch: git checkout -b feature/improvement
3. Commit: git commit -m 'Add improvement'
4. Push: git push origin feature/improvement
5. Open a Pull Request

🙏 Acknowledgements

• PyWavelets — Wavelet transforms
• SciPy — DCT/IDCT
• OpenCV — Video processing
• FFmpeg — Video encoding
• reedsolo — Reed-Solomon codes
• PyQt5 — GUI framework (GPLv3 license)
• QtAwesome — Icon fonts
• Big Buck Bunny — Test video (CC BY 3.0)

---

Licensed under GPLv3 • © 2026 qexela