Resulting in a 90% reduction in use of mouse. Further improvements can enhance accessibility for physically abled people.
While building this project, one major challenge was ensuring the accuracy and responsiveness of hand gesture recognition. Initial versions struggled with correctly identifying gestures in varying lighting conditions and against different backgrounds. To overcome this, I implemented advanced pre-processing techniques and fine-tuned the MediaPipe model parameters. Additionally, integrating the hand gesture recognition with OpenCV to achieve smooth cursor control required iterative testing and optimization to reduce latency and improve user experience.
Technologies used
Discussion