Gesture Controlled Video Player using Raspberry Pi and MediaPipe

The Gesture Controlled Video Player system operates by combining computer vision and edge machine learning directly on a Raspberry Pi 4 Model B. At the core of the setup, the Pi Camera Module captures a continuous real-time video stream of the user's hand movements. This raw frame data is processed using Python 3 alongside OpenCV, which prepares the image arrays for inference. The MediaPipe framework then executes a pre-trained machine learning model optimized for hand landmark detection, identifying key skeletal coordinates of the hand in the video frames. By analyzing the spatial positions and relative distances of these tracked landmarks—such as the thumb and index finger—the software program detects specific predefined hand gestures. Once a gesture is recognized, the script translates it into a standard operating system command by utilizing the PyAutoGUI library. PyAutoGUI simulates keyboard or mouse inputs locally on the Raspberry Pi, effectively triggering playback controls like play, pause, or volume adjustment within the active video player application. This entire pipeline runs locally on the edge, enabling seamless touch-free interaction without relying on cloud-based processing.