Real-Time Static Hand Gesture Recognition Using MediaPipe and OpenCV for Human-Computer Interaction

Hand gesture recognition plays a crucial role in Human-Computer Interaction (HCI), enabling intuitive and natural communication between humans and machines. This paper presents a real-time static hand gesture recognition system developed in Python, utilizing MediaPipe for precise hand landmark detection and OpenCV for video capture, visualization, and user feedback. The system is capable of recognizing four distinct gestures: open palm, fist, peace sign, and thumbs up, each mapped using a rule-based classification logic that analyzes the relative positions of finger landmarks. Unlike deep learning approaches, this method requires no extensive dataset or model training, offering lightweight, fast, and accurate performance on standard computing hardware. Real-time demonstrations validate the robustness of the system across different hand orientations and lighting conditions. This approach has potential applications in assistive technologies, gaming, sign language interfaces, and gesture-controlled devices, making it a practical solution for interactive environments. Future work may extend the system to recognize dynamic gestures, multiple hands, and a larger gesture vocabulary for more complex HCI tasks.