Introducing PleaseBlink!—a lightweight application crafted to gently remind you to blink your eyes during prolonged screen exposure.

• Blink Detection: Harnessing the power of dlib for accurate blink recognition. Support both local and remote detection.
• Interactive Notifications: Utilizing rumps to provide user-friendly notifications.
• Personalized Settings: Easily adjustable detection and notification parameters to cater to your preferences.

1. Use prebuilt .app Just download the latest release from here and double-click to run it.

2. Use source code Clone this repo and install the dependencies as instructions on Develop. Then change to root directory of this repo and run

   python ./PleaseBlink/main.py

   Use local blink detection by default. If you want to use remote detection, please run

   python ./PleaseBlink/main.py --detect_mode remote --remote_host <ip>:<port> # on the client side
   python ./PleaseBlink/server.py --port <port> # on the server side

Once running, PleaseBlink! operates quietly in the background monitoring your blink frequency.

Locate the icon on the status bar, click it to unveil the menu, and tweak the parameters as needed. The status bar icon will morph to RED indicating a prolonged period of no blinking. Upon detecting a blink, it shifts to GREEN.

PleaseBlink! will count the number of blinks over a specified duration. Should the blink count fall below the set threshold, a friendly reminder pops up prompting you to blink.

Note❗️: For the notification feature to function correctly, please ensure that Python has the necessary permissions to send notifications. Navigate to System Preferences → Notifications, scroll down to PleaseBlink, and activate the notification option.

PleaseBlink! is tailored for MacOS environments.

• Execute pip install -r requirements.txt to install the necessary dependencies.
• Execute pip install -r requirements_server.txt to install dependencies on the server if you want to use remote detection.

Note❗: It's noted that rumps and dlib might encounter issues with Anaconda's Python version, especially on M1 chips. It's recommended to use Brew's Python version and initiate a virtual environment.