AI-Enabled Multi-Modal Smart Door Access System using Face Voice and Gesture Recognition

The Multi-Mode Smart Door Access System planned to be designed will employ an arduino control system and ESP32-CAM to ensure that the authentication is as secure as possible without touching anything. It accomplishes this through voice checks and face checks. The mean response time was 1.8s and accuracy was 97%. False Acceptance rate (FAR) was 2 and the False Rejection rate (FRR) was 1.5. The experiment revealed that ArcFace (AUC = 0.95) was more effective than other models in observing things. The system performed good in varying degrees of light and noise, which demonstrated that it will be able to perform in intelligent homes and other settings in real time. This project is similar to SDG 9 (Industry, Innovation, and Infrastructure) and SDG 11 (Sustainable Cities and Communities).