What's Not on the Plate? Rethinking Food Computing through Indigenous Indian Datasets

2025-09-19

AI Summary

Food computing suffers from insufficient cultural inclusivity, modality limitations, and lack of community engagement—particularly acute for Indigenous cuisines and their ecological-cultural contexts in rural India, where data on endangered culinary traditions remains scarce. Method: We introduce the first multimodal dataset dedicated to Indigenous Indian diets, comprising 1,000 locally documented recipes from six states and ten linguistically endangered communities, collected via a customized mobile application by rural “first-time digital workers.” Data include text, high-resolution images, and audio recordings. Contribution/Results: This work establishes a “culturally inclusive, community co-created” paradigm for food computing, addressing the contextual poverty and modality homogeneity prevalent in mainstream food AI. The dataset enables food recognition, intangible culinary heritage preservation, and interdisciplinary research at the intersection of cultural AI, public health, and sustainable agri-food systems. It provides a reproducible technical pipeline and an ethics-grounded AI framework tailored to marginalized food systems.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Humans and AI: Crowd Sourcing and Human Computation

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance

Abstract

This paper presents a multimodal dataset of 1,000 indigenous recipes from remote regions of India, collected through a participatory model involving first-time digital workers from rural areas. The project covers ten endangered language communities in six states. Documented using a dedicated mobile app, the data set includes text, images, and audio, capturing traditional food practices along with their ecological and cultural contexts. This initiative addresses gaps in food computing, such as the lack of culturally inclusive, multimodal, and community-authored data. By documenting food as it is practiced rather than prescribed, this work advances inclusive, ethical, and scalable approaches to AI-driven food systems and opens new directions in cultural AI, public health, and sustainable agriculture.

Problem

• Addresses gaps in culturally inclusive multimodal food datasets
• Documents endangered indigenous food practices and ecological contexts
• Advances ethical AI approaches for sustainable food systems

Innovation

• Multimodal dataset from participatory rural data collection
• Documenting traditional food practices with ecological contexts
• Advancing inclusive ethical AI for food systems