Face and Food Recognition Using Convolutional Neural Networks

Abstract

This paper proposes a system with models which will recognize face and food from the given input images using CNN. The result will be saved and displayed. Shallow neural network is built over pre trained models for better results. The system can be used by schools which provide lunch to their students for keeping a proper record. The paper focuses on the models used for face and food recognition. The dataset for faces is built with around 400 images divided into 31 classes. Food dataset is built with around 500 images also with 31 classes. The validation accuracies of the models is 91% and 88% for face and food models respectively