Research Article
Gesture-Controlled Home Automation: AI-Driven Interaction for Smart Living
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357917, author={Gayathri Devi. K and Arulkarthick. C and Priyadharshini. S and Sandhiya Sree. K and Dhanasekar. A}, title={Gesture-Controlled Home Automation: AI-Driven Interaction for Smart Living}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={smart home automation gesture recognition computer vision iot machine learning arduino mediapipe deep learning touchless control}, doi={10.4108/eai.28-4-2025.2357917} }- Gayathri Devi. K
Arulkarthick. C
Priyadharshini. S
Sandhiya Sree. K
Dhanasekar. A
Year: 2025
Gesture-Controlled Home Automation: AI-Driven Interaction for Smart Living
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357917
Abstract
Smart home automation technology has been greatly advanced to provide users with convenient and smart control of home appliances. Touchless (e.g., gesture-based) home automation applications can deliver seamless user experiences allowing hands-off control, for when interacting directly with switches is inconvenient, or voice commands cannot be used. This paper demonstrates AI based gesture control automation using computer vision, machine learning and IOTs, to give convenient usage of the home devices to the user using hand gestures. A Python-based camera driven gesture recognition system that recognizes different hand gestures and sends corresponding commands to an Arduino Uno, which is connected to other smart devices like a multi-color LED lights, fan, etc. The system is intended to be used as an online gesture recognition system with low delay and high rate of recognition. By hand tracking with Media Pipe, gesture classification with deep learning models, and serial communication protocols for fractioning the device in tools, we make very easy the integration of the device in tools and giving users the possibility to be adapted to these systems. Evaluation was conducted with different lighting and gesture complexities, to assess the reliability and robustness. It has security concerns: Gesture-based password can help preclude unauthorized access to control of smart home etc.
