Multi-modal Cooking Workflow Construction for Food Recipes

Abstract

Understanding food recipe requires anticipating the implicit causal effects of cooking actions, such that the recipe can be converted into a graph describing the temporal workflow of the recipe. This is a non-trivial task that involves common-sense reasoning. However, existing efforts rely on hand-crafted features to extract the workflow graph from recipes due to the lack of large-scale labeled datasets. Moreover, they fail to utilize the cooking images, which constitute an important part of food recipes. In this paper, we build MM-ReS, the first large-scale dataset for cooking workflow construction, consisting of 9,850 recipes with human-labeled workflow graphs. Cooking steps are multi-modal, featuring both text instructions and cooking images. We then propose a neural encoder-decoder model that utilizes both visual and textual information to construct the cooking workflow, which achieved over 20% performance gain over existing hand-crafted baselines. CCS CONCEPTS • Information systems → Multimedia and multimodal retrieval.

Structure-Aware Procedural Text Generation From an Image Sequence

“…Although the above work has focused on parsing text-only recipes, little work has not addressed cross-modal analysis due to limited available datasets. Pan et al [11] recently created a novel cross-modal dataset, MM-ReS dataset. This dataset consists of recipes, image sequences, and annotated tree structures, allowing us to analyze the cause-and-effect relations between step texts and images in the recipe and image sequence.…”