Estimate Height Weight and Body Mass Index From Face Image Using Machine Learning

Abstract

Human body photographs contain encodings for gender, weight, and other biometric data in addition to pupil color. When considering all of this information, body weight is a reliable indicator of health problems. Body measurements including height, weight, and body mass index (BMI) are important because they can be used for a variety of purposes like surveillance, re-identification, picture retrieval systems, and healthcare. The majority of earlier research on automated height, weight, and BMI measurement has made use of 3D whole body videos and 2D and full body 2D pictures. A person's weight in proportion to their height is gauged by their BMI. It serves more as a gauge for calculating total body fat. Because it is frequently used to evaluate health issues, BMI is important. Our chances of living a longer, healthier life are said to be increased by having a healthy body mass index (BMI).The BMI range may be identified and classified using picture analysis, which can help individuals forecast credibility, manage their BMI, and lead healthier lives. There hasn't been much discussion on face estimation. Being overweight has been linked to obesity, diabetes, and cardiovascular disease. Weight is also an important health indicator. Using the methods Linear Regression, Ridge Linear Regression, Random Forest Regressor, and Kernel Ridge Regression, CNN algorithm, and Resnet-50 Architecture, we test the feasibility of estimating height, weight, and BMI from single-shot face photos. In order to estimate height, weight, and BMI, we will evaluate these regression models and select the one with the highest test score.