DefaceIT

<p align="center"> <a href="#defaceit-فارسی">راهنمای فارسی</a> | <a href="#defaceit-فارسی">Persian Guide</a> </p>

DefaceIT is a cross-platform application for blurring faces and license plates in videos using YOLOv11. The app supports both English and Persian languages and is available for desktop (macOS, Linux, Windows) and Android.

Features

Easy to use graphical interface
Fast processing with GPU acceleration support (CUDA, MPS, CPU)
Accurate detection using YOLOv11-based face and license plate detection
Audio preservation with automatic audio merging
Audio pitch shifting with preview functionality
Cross-platform support (macOS, Linux, Windows, Android)
Bilingual interface (English and Persian)
Customizable settings (blur strength, confidence, blur type)
Native Android app with modern Material Design UI

Requirements

Python 3.8 or higher
ffmpeg (for audio preservation)
- macOS: brew install ffmpeg
- Linux: sudo apt-get install ffmpeg (Ubuntu/Debian) or sudo yum install ffmpeg (RHEL/CentOS)
- Windows: Download from ffmpeg.org

Installation

macOS / Linux

Open Terminal
Navigate to the faceblur_app directory:
```
cd faceblur_app
```
Run the setup script:
```
chmod +x setup.sh
./setup.sh
```
Run the application:
```
python run.py
```

The run.py script will automatically:

Detect your operating system
Use the virtual environment if available
Fall back to system Python if needed
Handle tkinter detection and errors

Windows

Open Command Prompt or PowerShell
Navigate to the faceblur_app directory:
```
cd faceblur_app
```
Run the setup script:
```
setup.bat
```
Run the application:
```
python run.py
```

The run.py script will automatically:

Detect your operating system
Use the virtual environment if available
Fall back to system Python if needed
Handle tkinter detection and errors

Docker (Web Interface - Recommended)

The easiest way to run DefaceIT is using Docker with a simple web interface. No need to install Python, dependencies, or deal with GUI complications.

Prerequisites

Docker and Docker Compose installed (Get Docker)

Quick Start

Clone the repository:

git clone https://github.com/pourmand1376/DefaceIT.git
cd DefaceIT

Start the application:
```
docker-compose up
```
Open your browser and go to:
```
http://localhost:8080
```
Upload your video, adjust settings, and process!

The web interface provides:

Drag-and-drop video upload
All processing settings (blur strength, confidence, device, etc.)
Real-time progress tracking
Direct download of processed videos

Using GPU (NVIDIA)

To enable GPU acceleration for faster processing:

Install NVIDIA Container Toolkit

Uncomment the GPU section in docker-compose.yml:

deploy:
  resources:
    reservations:
      devices:
        - driver: nvidia
          count: all
          capabilities: [gpu]

Restart the application:
```
docker-compose down
docker-compose up
```

Stopping the Application

docker-compose down

Usage

Launch the application using the instructions above
Click "Browse..." next to "Input Video" to select your video file
Optionally set the output location (auto-generated if not set)
Adjust settings:
- Blur Strength: Higher values = more blur (21-101)
- Confidence: Lower values = more detections (0.05-0.5)
- Blur Type: Gaussian or Pixelate
- Detect: Choose faces and/or license plates
- Device: Auto-detect, CPU, or GPU
- Audio Pitch Shift: Adjust pitch in semitones (-12 to +12)
Click "Preview Audio" to test the pitch shift (optional)
Click "Start Processing"
Wait for processing to complete
Your blurred video will be saved with audio preserved

Android App

DefaceIT is also available as a native Android application with a modern Material Design interface.

Download

Download the latest release APK from the Releases page or build it yourself.

Requirements

Android 7.0 (API level 24) or higher
Camera permission (optional, for future camera features)
Storage permission (for reading and saving videos)

Installation

Download the DefaceIT-release.apk from the releases page
Enable "Install from Unknown Sources" in your Android settings
Open the downloaded APK file
Follow the installation prompts

Features

Modern Material Design 3 UI built with Jetpack Compose
Face detection using Google ML Kit
Real-time video processing
Audio pitch shifting support
Bilingual interface (English and Persian)
Same powerful blurring capabilities as the desktop version

Building from Source

To build the Android app from source:

Open the project in Android Studio
Sync Gradle dependencies
Build the release APK:
```
cd DefaceIT
./gradlew assembleRelease
```
The APK will be located at DefaceIT/app/build/outputs/apk/release/

Building Standalone Executable

Using PyInstaller

pip install pyinstaller
pyinstaller --onefile --windowed --name DefaceIT defaceit_gui.py

The executable will be in the dist folder.

Troubleshooting

Tkinter Not Found Error

If you see: ModuleNotFoundError: No module named '_tkinter'

macOS Solutions:

Option 1: Use System Python (Recommended)
```
/usr/bin/python3 defaceit_gui.py
```
Option 2: Install python-tk for Homebrew Python
```
brew install python-tk
```
Option 3: Use the macOS launcher script
```
./run_macos.sh
```

Linux Solutions:

# Ubuntu/Debian
sudo apt-get install python3-tk

# RHEL/CentOS/Fedora
sudo yum install python3-tk
# or
sudo dnf install python3-tk

Windows Solutions: Tkinter should be included with Python. If not:

Reinstall Python from python.org
Make sure "tcl/tk and IDLE" is checked during installation

Virtual Environment Issues

If tkinter doesn't work in your venv:

Option 1: Use system Python directly

python3 defaceit_gui.py

Option 2: Install dependencies globally (not recommended)

pip3 install -r requirements.txt
python3 defaceit_gui.py

Other Common Issues

No audio in output: Make sure ffmpeg is installed and in your PATH
Slow processing: Try using GPU acceleration or lower video resolution
Missing faces: Lower the confidence threshold (try 0.1)
Too much blur: Reduce blur strength
App runs but is slow: Make sure GPU acceleration is enabled (select "Auto" or "GPU" in device settings)
Faces not being detected: Lower the confidence threshold, increase blur strength for better coverage, make sure "Detect Faces" is checked
librosa not installed: Run pip install librosa soundfile for audio pitch shifting features

Notes

First run will download YOLOv11n model (~5.4MB)
Processing speed depends on your hardware (GPU recommended)
Audio preservation requires ffmpeg to be installed
Large videos may take some time to process

Credits

Developer: Shin

X (Twitter): @hey_itsmyturn
Website: https://sh1n.org
Telegram: https://t.me/itsthealephyouknowfromtwitter

Support the Developer

Donate (Crypto): https://nowpayments.io/donation/shin
Donate (Card): https://buymeacoffee.com/hey_itsmyturn

Note: Translation and Readme was generated by Cursor AI

DefaceIT (فارسی)

DefaceIT یک برنامه چند پلتفرمی برای تار کردن چهره‌ها و پلاک‌ها در ویدیوها با استفاده از YOLOv11 است. این برنامه از زبان‌های انگلیسی و فارسی پشتیبانی می‌کند و برای دسکتاپ (macOS, Linux, Windows) و اندروید در دسترس است.

ویژگی‌ها

رابط گرافیکی ساده
پردازش سریع با پشتیبانی از شتاب GPU (CUDA, MPS, CPU)
تشخیص دقیق با استفاده از YOLOv11
حفظ صدا با ادغام خودکار صدا
تغییر زیر و بم صدا با قابلیت پیش‌نمایش
پشتیبانی از چند پلتفرم (macOS, Linux, Windows, Android)
رابط دو زبانه (انگلیسی و فارسی)
تنظیمات قابل تنظیم
برنامه اندروید بومی با رابط کاربری Material Design مدرن

نیازمندی‌ها

Python 3.8 یا بالاتر
ffmpeg (برای حفظ صدا)
- macOS: brew install ffmpeg
- Linux: sudo apt-get install ffmpeg (Ubuntu/Debian) یا sudo yum install ffmpeg (RHEL/CentOS)
- Windows: از ffmpeg.org دانلود کنید

نصب

macOS / Linux

Terminal را باز کنید
به پوشه faceblur_app بروید:
```
cd faceblur_app
```
اسکریپت نصب را اجرا کنید:
```
chmod +x setup.sh
./setup.sh
```
برنامه را اجرا کنید:
```
python run.py
```

اسکریپت run.py به صورت خودکار:

سیستم عامل شما را تشخیص می‌دهد
در صورت وجود از محیط مجازی استفاده می‌کند
در صورت نیاز به Python سیستم بازمی‌گردد
تشخیص tkinter و خطاها را مدیریت می‌کند

Windows

Command Prompt یا PowerShell را باز کنید
به پوشه faceblur_app بروید:
```
cd faceblur_app
```
اسکریپت نصب را اجرا کنید:
```
setup.bat
```
برنامه را اجرا کنید:
```
python run.py
```

اسکریپت run.py به صورت خودکار:

سیستم عامل شما را تشخیص می‌دهد
در صورت وجود از محیط مجازی استفاده می‌کند
در صورت نیاز به Python سیستم بازمی‌گردد
تشخیص tkinter و خطاها را مدیریت می‌کند

Docker (رابط وب - توصیه می‌شود)

ساده‌ترین راه برای اجرای DefaceIT استفاده از Docker با یک رابط وب ساده است. نیازی به نصب Python، وابستگی‌ها یا مشکلات رابط گرافیکی نیست.

DefaceIT

Install / Use

README

DefaceIT

Features

Requirements

Installation

macOS / Linux

Windows

Docker (Web Interface - Recommended)

Prerequisites

Quick Start

Using GPU (NVIDIA)

Stopping the Application

Usage

Android App

Download

Requirements

Installation

Features

Building from Source

Building Standalone Executable

Using PyInstaller

Troubleshooting

Tkinter Not Found Error

Virtual Environment Issues

Other Common Issues

Notes

Credits

Support the Developer

DefaceIT (فارسی)

ویژگی‌ها

نیازمندی‌ها

نصب

macOS / Linux

Windows

Docker (رابط وب - توصیه می‌شود)