Food Recognition using Fusion of Classifiers based on CNNs

TL;DR

This paper proposes a fusion of multiple CNN classifiers to improve food recognition accuracy, addressing overfitting issues in deep models, and demonstrates superior performance on public datasets.

Contribution

It introduces a novel classifier fusion approach combining different CNN models to enhance food recognition accuracy.

Findings

Outperforms individual CNN models on Food-101 and Food-11 datasets.

Fusion approach reduces overfitting and improves classification performance.

Effective for fine-grained and high-level food categorization.

Abstract

With the arrival of convolutional neural networks, the complex problem of food recognition has experienced an important improvement in recent years. The best results have been obtained using methods based on very deep convolutional neural networks, which show that the deeper the model,the better the classification accuracy will be obtain. However, very deep neural networks may suffer from the overfitting problem. In this paper, we propose a combination of multiple classifiers based on different convolutional models that complement each other and thus, achieve an improvement in performance. The evaluation of our approach is done on two public datasets: Food-101 as a dataset with a wide variety of fine-grained dishes, and Food-11 as a dataset of high-level food categories, where our approach outperforms the independent CNN models.