FoodBD: a polygon-annotated meal image dataset of Bangladeshi cuisines with visual and nutritional labels
FoodBD comprises 3,523 smartphone-captured meal images representing authentic Bangladeshi meals, with minimal preprocessing to preserve real-world complexity. Each image is annotated with polygon-based segmentation across 67 food categories. Additionally, among them 1,837 images include expert-estimated nutritional information (carbohydrate, protein, fat, fiber, calorie, and glycemic load). The dataset is split into training, validation, and test subsets, facilitating reproducibility in machine learning pipelines.
Authors
- Mohammed Eunus Ali (ORCID: https://orcid.org/0000-0002-0384-7616)
- Mahmuda Naznin (ORCID: https://orcid.org/0000-0001-8753-4619)
- Mohammad Mehedy Masud (ORCID: https://orcid.org/0000-0002-5274-5982)
- B. Ahmed (ORCID: https://orcid.org/0000-0001-9973-6621)
- Md. Enamul Haque (ORCID: https://orcid.org/0000-0002-0702-9688)
- A. K. Obidul Huq (ORCID: https://orcid.org/0000-0002-1679-6520)
Institutions
- Bangladesh University of Engineering and Technology (BD)
- United Arab Emirates University (AE)
- Mawlana Bhashani Science and Technology University (BD)
- Monash University (AU)
- United International University (BD)
Publication Details
- Journal
- BMC Research Notes
- Published
- 2025-12-09
- DOI
- https://doi.org/10.1186/s13104-025-07583-8
- Primary Topic
- Nutritional Studies and Diet
- Type
- article
- Field-Weighted Citation Impact
- 0.00