Halloween Project @ Quincy College Computer Club by 🧟♀️ Thanh Van Nguyen and 🎃 Arjun Pramanik
Detect alien hands using 100% effective science! This project uses a webcam to analyze hand landmarks and classify hands as "Alien" or "Human" based on finger proportions. Capture and save detected alien hands, and play themed videos and sounds for each detection.
- Real-time hand landmark detection using MediaPipe
- Classifies hands as "Alien" or "Human" based on finger length difference
- Saves images of detected alien hands
- Plays themed video/audio on detection
- Palm line detection test script (see
palm.py) - Works with webcam or test images
- Python 3.12 or before
- Webcam
- Clone this repo
git clone <LINK_HERE> cd alien-detector
- Set up the virtual environment and activate it
pip install virtualenv virtualenv .venv .venv\Scripts\Activate
- Install dependencies:
pip install -r requirements.txt
Run the main detector (ensure the virtual environment is activated first):
python main.pyControls:
c: Capture and classify current handq: Quit
Warning
By using this script, you acknowledge that you may be assisting a top-secret government operation in the detection of extraterrestrial hand activity. The authors accept no responsibility for the results of this program. Use at your own risk, alien!
main.py: Main alien hand detectorpalm.py: Palm line detection (test script)hand_landmarker.task: Pre-trained machine learning model to identify hands and hand landmarks, by Google, for use with MediaPipealien_hands_detected/: Saved detection result imagesvids/: Saved video clipsdemo-images/: Demo images showcasing the model at workresult-vids/: Media files for detection eventsother/: Contains Blender animation by Thanh for other Halloween project work
This project was developed on Windows 11 and tested with PowerShell. WSL may have issues accessing cameras, so running on Linux could require additional setup.
- Thanh Van Nguyen
- Arjun Pramanik


