HabitKicker

A computer vision application for Windows that automatically detects unwanted habits like nail-biting, hair-pulling, and slouching using real-time landmark analysis.

Features

• Real-time habit detection using computer vision
• Posture calibration for personalized monitoring
• Customizable detection settings for:
  • Nail biting distance (distance from finger to mouth)
  • Hair pulling distance (distance from finger to hair)
• User-friendly GUI with dark mode
• Configurable notifications and alerts
• Live camera feed with visual feedback
  • Modifiable FPS (⭐2-5 FPS is recommended for most systems to conserve resources)

Requirements

• Python 3.12.10
• Dependencies:
  • OpenCV (opencv-python) == 4.11.0.86
  • MediaPipe == 0.10.14
  • PyQt6 == 6.8.1
  • Pygame == 2.6.1
• Hardware:
  • Minimum 480p webcam

Installation

Run the application using HabitKicker/main.exe in Releases

Or from source:

1. Clone the repository:

git clone https://github.com/ExxML/HabitKicker.git
cd HabitKicker

2. Create and activate a local environment:

python -m venv .venv
.venv\Scripts\activate

3. Install the required packages:

pip install -r requirements.txt

4. Run in terminal:

python habitkicker\main.py

Compile to .exe Using Nuitka

1. Install Nuitka:

pip install nuitka

2. Generate standalone .exe:

python -m nuitka --standalone --windows-icon-from-ico="data/HabitKicker.ico" --windows-console-mode=disable --enable-plugin=pyqt6 --enable-plugin=tk-inter --include-data-dir=".venv\Lib\site-packages\mediapipe\modules=mediapipe/modules" --include-data-dir="data=data" --include-data-dir="sounds=sounds" habitkicker\main.py

Note: Win32 calls are used in this program to ensure the notification overlay stays on top. Windows Defender may flag this as a virus. Add an exclusion in Settings > Privacy & Security > Windows Security > Virus & Threat Protection > Manage settings > Exclusions > Add an exclusion.

3. [Optional] Auto-run on startup:

• Create a shortcut for main.exe
• Move the shortcut to the Startup folder (Press Win + R, then type shell:startup)

Getting Started

1. Launch the application
2. Use the "Calibrate Posture" button to set your baseline posture
3. Adjust the detection settings according to your preferences
4. Enable/disable notifications and alerts as needed
5. The application will monitor your habits and provide real-time feedback

Shortcuts

• Click tray icon to toggle window visibility
• Ctrl + C to calibrate posture
• Ctrl + T to toggle camera feed
• Ctrl + X to hide window

License

This project is licensed under the MIT License.

Version

Current version: 1.0