We all enjoy watching movies from a comfy spot with pillows, snacks and blankets. But...those ads at the at the climax of an epic scene? Not so enjoyable. You had to get up, stumble through your snack setup, pillows, get to your laptop and skip the ad. Enter Wizard.ly, a desktop application that allows us to execute commands with nothing but a few gestures of your palm - almost like it's straight out of a Harry Potter movie.
Do you want to:
- play/pause a YouTube video,
- scroll pages,
- adjust the volume,
- move the mouse and click,
- or even send a message using your voice?
The machine should be running Python 3.8 or above.
- The system should have OpenCV and MediaPipe installed. If not
pip install opencv-python mediapipe
- Then, in your command prompt, run
python wizardly.py. - Enjoy!
The core of the project is built using:
- Python
- OpenCV
- MediaPipe
- Numpy
Other libraries used to enhance functionality include:
I'm glad to have used my OpenCV knowledge to build a tool that I can see myself using regularly. If you do check out my code, you will find that my method for scrolling pages a bit cheeky 😁 - I was very excited when that worked out. Also, having developed a module for hand tracking using OOP principles, it will enable me to work on future projects more efficiently. Lastly, I find it amazing that we can achieve wizardly powers with the power of Computer Vision, and it compells me to dive deeper!
Credits to Murtaza's Workshop for helping me learn OpenCV.