NutritionVerse-Real: An Open Access Manually Collected 2D Food Scene Dataset for Dietary Intake Estimation
889 annotated food scene images with nutritional data for 251 dishes
- Source
- Kaggle
- Published
- 2023
- Region
- North America
- Format
- csv
- Access
- Account needed
- Themes
- Public Health, Agriculture & Farming, Machine Learning & AI
About this dataset
This dataset comprises 889 images of 251 distinct dishes and 45 unique food types, each annotated with segmentation masks and detailed nutritional information. It aims to facilitate research in dietary intake estimation and nutrition analysis.
How the data was collected
Images were manually collected in real-life settings. Each dish's ingredients were weighed, and their nutritional content was calculated using information from food packaging and the Canada Nutrient File. Segmentation masks were generated through human labeling.