CALORIE ESTIMATION OF FOOD AND BEVERAGES USING DEEP LEARNING

CALORIE ESTIMATION OF FOOD AND BEVERAGES USING DEEP LEARNING

SIDDESH K T 1,   KOTRU SWAMY 2, NISHCHITHA S G 3
[1] Assistant Professor Department of MCA, BIET, Davangere

[2] Assistant Professor  Department of MCA, BIET, Davangere

[3] Student, Department of MCA, BIET, Davangere

ABSTRACT

This study deals with the important need for precise calorie estimations in today's fast-paced, health-conscious world, where existing measurement methods frequently fall short due to human error and approximation. In response, we offer a novel solution that uses automated deep learning algorithms to calculate calories from food and beverage pictures. Our method uses convolutional neural networks (CNNs) to recognize food items and portions from photos, which are supplemented by recurrent neural networks (RNNs) to handle different food compositions. Our approach, which trains on a huge dataset of annotated photos, promises to produce exact and trustworthy calorie predictions, providing a robust option for those looking to better track their caloric consumption. This study advances nutritional science by utilizing cutting-edge technology to solve practical health challenges, which aligns with the increased emphasis on informed food choices and individualized health management in modern lifestyles.

Keyword: Calorie Estimation, Object Detection, Food Classification, Deep Learning,CNN, RNN.