A Smile Detection for Hands-Free Selfie Capture Using Machine Learning
Keywords:
Smile Detection, Hands-Free Selfie, Convolutional Neural Networks, OpenCV, Real-Time Image ProcessingAbstract
This paper presents a novel, real-time smile detection system designed to enable hands-free selfie capture using machine learning. The system leverages computer vision techniques and deep learning models to accurately detect smiles in live camera feeds, triggering automatic photo capture without user intervention. Built on a modular architecture utilizing OpenCV for face detection and a convolutional neural network (CNN) for smile classification, the application ensures low-latency performance suitable for mobile and embedded platforms. The system is evaluated on public datasets such as GENKI-4K and CelebA, achieving an average accuracy of 94.2% in real-world lighting and expression conditions. A lightweight, Flask-based web interface offers live preview, detection feedback, and photo gallery integration. Experimental results show that the system operates at over 15 FPS on mid-range hardware, confirming its applicability for edge devices. Future extensions include emotion-based gesture capture, multilingual voice commands, and AR filter integration. The system demonstrates the potential of machine learning to create intuitive, user-friendly photo applications with minimal manual input.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Journal of Data Science

This work is licensed under a Creative Commons Attribution 4.0 International License.