CC-FOOD-100

a new food dataset(rgb+depth) CC Food-100 for food detection and nutrient calculation. The dataset has been collected in real college canteen in China. Depth images are raw data that can be used for food volume calculation or other downstream tasks.

Name	Address
CC-FOOD-100	https://pan.baidu.com/s/1Cb4AULANrZSCKgsg10XFaQ password:9ucz

Dataset structure

|-- <CC-FOOD-100>
    |-- <json> //Store label files
        0.json
        1.json
        ...
    |-- <depth> //Store depth images
        0.npy
        1.npy
        ...
    |-- <rgb> //Store color images
        0.png
        1.png
        ...
    |-- <xml> //Store voc xml
        0.xml
        1.xml
        ...
        

@article{GAO2024141739,
title = {A vision-based dietary survey and assessment system for college students in China},
journal = {Food Chemistry},
pages = {141739},
year = {2024},
issn = {0308-8146},
doi = {https://doi.org/10.1016/j.foodchem.2024.141739},
url = {https://www.sciencedirect.com/science/article/pii/S0308814624033892},
author = {Zicheng Gao and Xufeng Yuan and Jie Lei and Hao Guo and Francesco Marinello and Lorenzo Guerrini and Alberto Carraro},
keywords = {Nutritional intake analysis, RGB-D image, College students' dietary habits, Automated food recognition, Volume estimation},
}