Deep Image-to-Recipe Translation

TL;DR

This paper introduces a deep learning approach combining computer vision and natural language processing to predict ingredients from food images and generate recipes, highlighting transfer learning's effectiveness.

Contribution

It develops a novel deep learning framework for image-to-recipe translation, integrating custom CNNs, transfer learning, and sequence models for recipe generation.

Findings

Transfer learning with ResNet-50 and GloVe embeddings significantly improves performance.

Metrics like IoU, F1, and perplexity are crucial for evaluating model accuracy.

Progress made in bridging food images and recipes, with future potential for model and dataset improvements.

Abstract

The modern saying, "You Are What You Eat" resonates on a profound level, reflecting the intricate connection between our identities and the food we consume. Our project, Deep Image-to-Recipe Translation, is an intersection of computer vision and natural language generation that aims to bridge the gap between cherished food memories and the art of culinary creation. Our primary objective involves predicting ingredients from a given food image. For this task, we first develop a custom convolutional network and then compare its performance to a model that leverages transfer learning. We pursue an additional goal of generating a comprehensive set of recipe steps from a list of ingredients. We frame this process as a sequence-to-sequence task and develop a recurrent neural network that utilizes pre-trained word embeddings. We address several challenges of deep learning including imbalanced datasets, data cleaning, overfitting, and hyperparameter selection. Our approach emphasizes the importance of metrics such as Intersection over Union (IoU) and F1 score in scenarios where accuracy alone might be misleading. For our recipe prediction model, we employ perplexity, a commonly used and important metric for language models. We find that transfer learning via pre-trained ResNet-50 weights and GloVe embeddings provide an exceptional boost to model performance, especially when considering training resource constraints. Although we have made progress on the image-to-recipe translation, there is an opportunity for future exploration with advancements in model architectures, dataset scalability, and enhanced user interaction.