Face-Authenticated Touchless Interface for Gesture-Based Interaction and Personalized File Access using Raspberry pi

Abstract

This research introduces a multi-modal, touchless interface for the Raspberry Pi 4, designed to enhance Human-Computer Interaction (HCI) through the fusion of facial recognition and hand gesture control. By utilizing lightweight deep learning frameworks, including MediaPipe for gesture tracking and TensorFlow Lite for CNN-based facial embeddings, the system provides secure, identity-based file access with 95% accuracy and sub-100ms latency. Robustness in diverse lighting is achieved through CLAHE preprocessing, while an ergonomic gesture set integrated with pyautogui enables seamless cursor navigation and scrolling. Security is reinforced through identity-specific file management using os/shutil and VeraCrypt-encrypted volumes, which unlock only upon a verified facial match. Experimental results involving 50 users demonstrate a 98% gesture precision rate and a consistent 30fps processing speed with zero false positives in authentication process and This innovative framework offers a high-performance, low-latency solution for post-pandemic computing, shared kiosks, and edge-based security applications for effective performing in traditional single purpose authentication systems through its will integration of contactless navigation and automated the data privacy.