The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Vues en texte intégral
81
Ces dernières années, l’essor d’une alimentation saine a conduit à diverses applications de gestion alimentaire dotées d’une fonction de reconnaissance d’image pour enregistrer automatiquement les repas quotidiens. Cependant, la plupart des fonctions de reconnaissance d’images des applications existantes ne sont pas directement utiles pour les photos de plats composés de plusieurs plats et ne peuvent pas estimer automatiquement les calories des aliments. Pendant ce temps, les méthodologies de reconnaissance d’images ont considérablement progressé grâce à l’avènement du réseau neuronal convolutif (CNN). CNN a amélioré la précision de divers types de tâches de reconnaissance d'images telles que la classification et la détection d'objets. Par conséquent, nous proposons une estimation des calories alimentaires basée sur CNN pour les photos de plats contenant plusieurs plats. Notre méthode estime simultanément l'emplacement des plats et les calories des aliments par un apprentissage multitâche de la détection des plats de nourriture et de l'estimation des calories des aliments avec un seul CNN. Il est prévu d'atteindre une vitesse élevée et une petite taille de réseau grâce à une estimation simultanée dans un seul réseau. Étant donné qu'il n'existe actuellement aucun ensemble de données de photos d'aliments contenant plusieurs plats annotées à la fois avec des cadres de délimitation et des calories alimentaires, dans ce travail, nous utilisons alternativement deux types d'ensembles de données pour former un seul CNN. Pour les deux types d'ensembles de données, nous utilisons des photos d'aliments comportant plusieurs plats annotées avec des cadres de délimitation et des photos d'aliments contenant un seul plat avec les calories alimentaires. Nos résultats ont montré que notre méthode multitâche permettait d'obtenir une précision plus élevée, une vitesse plus élevée et une taille de réseau plus petite qu'un modèle séquentiel de détection d'aliments et d'estimation des calories alimentaires.
Takumi EGE
The University of Electro-Communications
Keiji YANAI
The University of Electro-Communications
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Takumi EGE, Keiji YANAI, "Simultaneous Estimation of Dish Locations and Calories with Multi-Task Learning" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 7, pp. 1240-1246, July 2019, doi: 10.1587/transinf.2018CEP0004.
Abstract: In recent years, a rise in healthy eating has led to various food management applications which have image recognition function to record everyday meals automatically. However, most of the image recognition functions in the existing applications are not directly useful for multiple-dish food photos and cannot automatically estimate food calories. Meanwhile, methodologies on image recognition have advanced greatly because of the advent of Convolutional Neural Network (CNN). CNN has improved accuracies of various kinds of image recognition tasks such as classification and object detection. Therefore, we propose CNN-based food calorie estimation for multiple-dish food photos. Our method estimates dish locations and food calories simultaneously by multi-task learning of food dish detection and food calorie estimation with a single CNN. It is expected to achieve high speed and small network size by simultaneous estimation in a single network. Because currently there is no dataset of multiple-dish food photos annotated with both bounding boxes and food calories, in this work we use two types of datasets alternately for training a single CNN. For the two types of datasets, we use multiple-dish food photos annotated with bounding boxes and single-dish food photos with food calories. Our results showed that our multi-task method achieved higher accuracy, higher speed and smaller network size than a sequential model of food detection and food calorie estimation.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018CEP0004/_p
Copier
@ARTICLE{e102-d_7_1240,
author={Takumi EGE, Keiji YANAI, },
journal={IEICE TRANSACTIONS on Information},
title={Simultaneous Estimation of Dish Locations and Calories with Multi-Task Learning},
year={2019},
volume={E102-D},
number={7},
pages={1240-1246},
abstract={In recent years, a rise in healthy eating has led to various food management applications which have image recognition function to record everyday meals automatically. However, most of the image recognition functions in the existing applications are not directly useful for multiple-dish food photos and cannot automatically estimate food calories. Meanwhile, methodologies on image recognition have advanced greatly because of the advent of Convolutional Neural Network (CNN). CNN has improved accuracies of various kinds of image recognition tasks such as classification and object detection. Therefore, we propose CNN-based food calorie estimation for multiple-dish food photos. Our method estimates dish locations and food calories simultaneously by multi-task learning of food dish detection and food calorie estimation with a single CNN. It is expected to achieve high speed and small network size by simultaneous estimation in a single network. Because currently there is no dataset of multiple-dish food photos annotated with both bounding boxes and food calories, in this work we use two types of datasets alternately for training a single CNN. For the two types of datasets, we use multiple-dish food photos annotated with bounding boxes and single-dish food photos with food calories. Our results showed that our multi-task method achieved higher accuracy, higher speed and smaller network size than a sequential model of food detection and food calorie estimation.},
keywords={},
doi={10.1587/transinf.2018CEP0004},
ISSN={1745-1361},
month={July},}
Copier
TY - JOUR
TI - Simultaneous Estimation of Dish Locations and Calories with Multi-Task Learning
T2 - IEICE TRANSACTIONS on Information
SP - 1240
EP - 1246
AU - Takumi EGE
AU - Keiji YANAI
PY - 2019
DO - 10.1587/transinf.2018CEP0004
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2019
AB - In recent years, a rise in healthy eating has led to various food management applications which have image recognition function to record everyday meals automatically. However, most of the image recognition functions in the existing applications are not directly useful for multiple-dish food photos and cannot automatically estimate food calories. Meanwhile, methodologies on image recognition have advanced greatly because of the advent of Convolutional Neural Network (CNN). CNN has improved accuracies of various kinds of image recognition tasks such as classification and object detection. Therefore, we propose CNN-based food calorie estimation for multiple-dish food photos. Our method estimates dish locations and food calories simultaneously by multi-task learning of food dish detection and food calorie estimation with a single CNN. It is expected to achieve high speed and small network size by simultaneous estimation in a single network. Because currently there is no dataset of multiple-dish food photos annotated with both bounding boxes and food calories, in this work we use two types of datasets alternately for training a single CNN. For the two types of datasets, we use multiple-dish food photos annotated with bounding boxes and single-dish food photos with food calories. Our results showed that our multi-task method achieved higher accuracy, higher speed and smaller network size than a sequential model of food detection and food calorie estimation.
ER -