This project presents a robust gesture-based action recognition system that translates hand gestures into meaningful computer actions, featuring high-accuracy models like a Keras Linear Sequential model and YOLOv5. It identifies five distinct hand poses—open, close, ok, pointer, and brightness control(peace sign)—and maps them to actions such as brightness and volume control, tab navigation, and scrolling,automatic powerpoint presentation setup. The system combines deep learning with computer vision and human-computer interaction principles, offering an intuitive interface for diverse user needs.
