I created a gesture-based air drawing system that enables drawing in the air using hand movements ✋ ✊ 👆 Developed using Python, OpenCV, and a pre trained MediaPipe model for real time hand tracking and gesture recognition. I also implemented a two-hand control mechanism where one hand acts as a “permission switch” to reduce accidental inputs and improve control. It started as a simple idea, but quickly turned into a fun challenge, especially when the system decided to draw unexpected things 😁 Just a beginning, with plenty of improvements and features to explore next ⚡ Github repository: https://lnkd.in/g2ysHBai Project Report: https://lnkd.in/gxAn47ZS #ComputerVision #Python #OpenCV #MediaPipe
This is great!. You could integrate this into virtual calls as a mechanism to draw diagrams.
That's Awesome
❤️
Thanoj Buddhima, this is really cool. It is impressive to see how smoothly your air drawing system coordinates with hand gestures. Could develop into a tool where people with speech difficulties, yet are very good at hand gestures to perform a task or series of tasks. Keep it up!
Impressive work, Thanoj! Integrating a two-hand control mechanism as a 'permission switch' is a very smart way to solve the accidental input issue in gesture recognition. Great use of MediaPipe and OpenCV. Keep up the great work!
Very cool project and excellent idea to use hand to change color, the changement is basic but that open perspective to define parameter according to the movement of the hand. Congratulations !
Great
Great work Thanoj Buddhima 🎉
This is really cool 👏 I like how you didn’t just stop at basic hand tracking but actually thought about usability with the two-hand “permission switch” — that’s a smart touch and shows real problem-solving. Also, that moment when the system starts drawing “unexpected things” honestly, that’s where the real learning (and fun) happens. Feels like a solid foundation, you could take this into gesture-based UI, virtual whiteboards, or even AR-style interactions. Excited to see where you take it next 🚀