PFID: Pittsburgh fast-food image dataset

We introduce the first visual dataset of fast foods with a total of 4,545 still images, 606 stereo pairs, 303 360 0 videos for structure from motion, and 27 privacy-preserving videos of eating events of volunteers. This work was motivated by research on fast food recognition for dietary assessment....

Full description

Saved in:
Bibliographic Details
Published in2009 16th IEEE International Conference on Image Processing (ICIP) pp. 289 - 292
Main Authors Mei Chen, Dhingra, K., Wen Wu, Lei Yang, Sukthankar, R., Jie Yang
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.11.2009
Subjects
Face recognition
Food image dataset
Food technology
Histograms
Image recognition
Large-scale systems
Object detection
Object recognition
Testing
Videos
Online AccessGet full text
ISBN9781424456536
1424456533
ISSN1522-4880
DOI10.1109/ICIP.2009.5413511

Cover

More Information
Summary:We introduce the first visual dataset of fast foods with a total of 4,545 still images, 606 stereo pairs, 303 360 0 videos for structure from motion, and 27 privacy-preserving videos of eating events of volunteers. This work was motivated by research on fast food recognition for dietary assessment. The data was collected by obtaining three instances of 101 foods from 11 popular fast food chains, and capturing images and videos in both restaurant conditions and a controlled lab setting. We benchmark the dataset using two standard approaches, color histogram and bag of SIFT features in conjunction with a discriminative classifier. Our dataset and the benchmarks are designed to stimulate research in this area and will be released freely to the research community.
ISBN:9781424456536
1424456533
ISSN:1522-4880
DOI:10.1109/ICIP.2009.5413511