Bangladeshi Street Food Calorie Estimation Using Improved YOLOv8 and Regression Model ∗ These authors contributed equally to this work
Aparup Dhar 1∗ , MD Tamim Hossain 1∗ , Pritom Barua 1 · 2025
Read Bangladeshi Street Food Calorie Estimation Using Improved YOLOv8 and Regression Model ∗ These authors contributed equally to this work by Aparup Dhar 1∗ , MD Tamim Hossain 1∗ , Pritom Barua 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-01415.txt.
Description
Estimating calories from a photo of food can aid diet tracking, but most vision systems were trained on Western dishes and ignore portion size. Aparup Dhar, Antu Das, Pritom Barua, and MD Tamim Hossain build a method tuned to Bangladeshi street food, which is widely eaten and culturally central. Their data set holds three thousand eight hundred eighty-five photos across six classes, from singara to peaju, plus a coin as a size reference.
Five detection and segmentation models were compared for finding and outlining the food. A Bangladeshi five taka coin in each photo set the scale, so measured sizes became real dimensions. Geometric features then fed regression models, including random forest and boosting methods, to predict calories. One detector, YOLO11n, led the comparison. The work aims at dietary tools that fit local food, not just imported menus.
Catalog record
| CREATOR | Aparup Dhar 1∗ , MD Tamim Hossain 1∗ , Pritom Barua 1 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 23 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-01415