Revamping Cross-Modal Recipe Retrieval with Hierarchical Transformers and Self-supervised Learning

TL;DR

This paper introduces a hierarchical Transformer model with self-supervised learning for cross-modal recipe retrieval, achieving state-of-the-art results by effectively encoding recipe components and leveraging semantic relationships.

Contribution

The paper presents a simplified end-to-end hierarchical Transformer model with a novel self-supervised loss for improved cross-modal recipe retrieval.

Findings

Achieves state-of-the-art performance on Recipe1M dataset.

Effective encoding of recipe components improves retrieval accuracy.

Self-supervised loss leverages semantic relationships within recipes.

Abstract

Cross-modal recipe retrieval has recently gained substantial attention due to the importance of food in people's lives, as well as the availability of vast amounts of digital cooking recipes and food images to train machine learning models. In this work, we revisit existing approaches for cross-modal recipe retrieval and propose a simplified end-to-end model based on well established and high performing encoders for text and images. We introduce a hierarchical recipe Transformer which attentively encodes individual recipe components (titles, ingredients and instructions). Further, we propose a self-supervised loss function computed on top of pairs of individual recipe components, which is able to leverage semantic relationships within recipes, and enables training using both image-recipe and recipe-only samples. We conduct a thorough analysis and ablation studies to validate our design choices. As a result, our proposed method achieves state-of-the-art performance in the cross-modal recipe retrieval task on the Recipe1M dataset. We make code and models publicly available.