Stick Figure Generator from Human Photos

This project uses Python and MediaPipe to detect human poses in photos and render them as simple stick figures.

Features

• Uses MediaPipe pose estimation to detect human landmarks (joints)
• Draws stick figures based on the detected keypoints
• Optionally overlays the stick figure on the original image, or outputs a standalone figure on a white background

Requirements

• Python 3.7 or higher
• mediapipe
• opencv-python
• numpy

You can install the required packages with:

pip install -r requirements.txt

Directory Structure

.
├── inputs/
│   └── input.jpg
├── outputs/
│   └── output.jpg
├── scripts/
│   └── stick_figure.py
├── .gitignore
├── LICENSE
├── README_ja.md
├── README.md (you are here)
└── requirements.txt

Usage

Run the following command from the project root directory:

python scripts/stick_figure.py --input inputs/input.jpg --output outputs/output.jpg --overlay True

Setting the --overlay option to True will overlay the stick figure on the original image. Setting it to False will draw the stick figure on a white background instead.

Notes

• Input images should clearly contain a person. To ensure accurate pose detection, it is recommended to use images where the full body and face are visible.
• The output image will be saved to outputs/output.jpg.

License

MIT License

---

Special thanks to MediaPipe and OpenCV ❤️