Face Tracking System

Developed using a Raspberry Pi 5, this motion-tracking system automatically detects and follows a face in real time. I built it from the ground up using Python, integrating a high-definition webcam, SG90 servos, and a PCA9685 controller. The system runs a Mediapipe-based detection algorithm that calculates the position of the face, then moves the camera using two micro servos to keep the face centered.

To keep the movements smooth and avoid jitter, I added a moving average algorithm and implemented thresholds so the servos only move when necessary. Everything is structured into a modular Python program, with separate scripts for face detection and servo control, making the code easy to manage and extend.

This project serves as a demonstration of real-world embedded systems, combining computer vision, I2C communication, and mechanical control into a responsive and reliable solution. It has applications in robotics, automation, and surveillance, and could easily be extended to track multiple objects, operate remotely, or handle outdoor environments.