ABOUT FACEBMI

Understanding AI-Powered BMI Category Estimation

FaceBMI is an intelligent web-based application that estimates a user's Body Mass Index (BMI) category through facial image analysis using Artificial Intelligence (AI). By combining computer vision and machine learning technologies, the system analyzes facial features captured from a webcam and predicts the user's BMI category without requiring traditional height and weight measurements. The application is designed to demonstrate how AI can be applied to healthcare screening in a simple, interactive, and accessible way.

Project Background

Body Mass Index (BMI) is widely used as an indicator for assessing whether an individual's body weight falls within a healthy range. Conventional BMI calculation requires users to manually provide their height and weight, which may not always be convenient or readily available. Recent advancements in computer vision and deep learning have shown that certain facial characteristics are correlated with BMI categories, opening new possibilities for non-contact health assessment. FaceBMI explores this concept by leveraging machine learning to estimate BMI categories using only facial images captured through a webcam.

Project Objective

The primary objective of FaceBMI is to develop an AI-powered web application capable of estimating a user's BMI category through facial image analysis. The system aims to provide users with an intuitive scanning experience while demonstrating the practical application of computer vision and machine learning in healthcare. In addition, the project seeks to increase awareness of personal health by providing users with basic health recommendations based on the predicted BMI category.

How FaceBMI Works

The application begins by requesting permission to access the user's webcam. Once permission is granted, OpenCV detects the user's face before MediaPipe FaceMesh identifies detailed facial landmarks. Relevant facial measurements are then extracted and processed by a TensorFlow-based machine learning model to estimate the user's BMI category. Finally, the predicted BMI category together with general health recommendations is displayed to the user.

Technology Behind FaceBMI

FaceBMI combines modern web development technologies with computer vision and machine learning. The frontend is built using HTML5, Bootstrap, CSS3, and JavaScript to provide a responsive and interactive user experience. The backend is powered by Python and Flask, while OpenCV performs face detection and MediaPipe FaceMesh extracts facial landmarks. These landmark measurements are processed by a TensorFlow-based machine learning model to estimate the user's BMI category. Together, these technologies create an intelligent AI-powered health assessment system.

Privacy and User Data

User privacy is an important consideration throughout the development of FaceBMI. Facial images are processed only for the purpose of generating BMI predictions and are not permanently stored by the system. Images exist only temporarily during processing and are discarded immediately after a prediction has been generated. This approach helps protect user privacy while maintaining the functionality of the application.

Project Scope

FaceBMI is intended as a BMI category estimation system rather than a medical diagnostic tool. The prediction results are generated using a trained machine learning model and should be regarded as general health information instead of professional medical advice. Factors such as lighting conditions, camera quality, facial positioning, and dataset limitations may influence prediction accuracy. Therefore, users are encouraged to consult qualified healthcare professionals for comprehensive health assessments.

Looking Ahead

FaceBMI demonstrates the growing potential of artificial intelligence in transforming healthcare through accessible digital solutions. As machine learning models continue to improve and larger datasets become available, future versions of the system may achieve higher prediction accuracy, support additional health indicators, and provide more personalized health recommendations. The project serves as a foundation for exploring innovative AI-driven healthcare applications that are both practical and user-friendly.