FoodPlaces: Learning deep features for food related scene understanding

Md Mostafa Kamal Sarker; Maria Leyva; Adel Saleh; Vivek Kumar Singh; Farhan Akram; Petia Radeva; Domenec Puig

doi:10.3233/978-1-61499-806-8-156

FoodPlaces: Learning deep features for food related scene understanding

Md Mostafa Kamal Sarker^*, Maria Leyva, Adel Saleh, Vivek Kumar Singh, Farhan Akram, Petia Radeva, Domenec Puig

^*Corresponding author for this work

Research output: Chapter/Conference proceeding › Conference proceeding › Academic › peer-review

2 Citations (Scopus)

Abstract

In human smart nutrition systems, environment based food classification has become popular to help analyzing the food intake based on the nutrition related activity. In this paper, we address the problem of food related environments, which refer to different eating places such as, bars, restaurants, coffee shops, etc. using state-of-the-art convolutional neural networks (CNNs). We collected a new dataset on different food related environments by integrating three publicly available datasets: Places365, ImageNet and SUN397. We have named it “FoodPlaces” and it contains 35 different types of classes. In order to achieve satisfactory results on the food related environment recognition, we fine-tuned several state-of-the-art CNNs, such as VGG16, RsNet50 and InceptionV3 using different transfer learning approaches. The results show that the fully fine-tunned InceptionV3 yields 75.22% classification accuracy among the discussed state-of-the-art CNNs.

Original language	English
Title of host publication	Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017
Editors	Isabel Aguilo, Cecilio Angulo, Rene Alquezar, Alberto Ortiz, Joan Torrens
Publisher	IOS Press BV
Pages	156-165
Number of pages	10
ISBN (Electronic)	9781614998051
DOIs	https://doi.org/10.3233/978-1-61499-806-8-156
Publication status	Published - 2017
Externally published	Yes
Event	20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017 - Deltebre, Terres de l'Ebre, Spain Duration: 25 Oct 2017 → 27 Oct 2017

Publication series

Series	Frontiers in Artificial Intelligence and Applications
Volume	300
ISSN	0922-6389

Conference

Conference	20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017
Country/Territory	Spain
City	Deltebre, Terres de l'Ebre
Period	25/10/17 → 27/10/17

Bibliographical note

Funding Information:
This research is funded by the program Marti Franques under the agreement between Universitat Rovira Virgili and Fundaci Catalunya La Pedrera.

Publisher Copyright:
© 2017 The authors and IOS Press. All rights reserved.

Access to Document

10.3233/978-1-61499-806-8-156

Cite this

Sarker, M. M. K., Leyva, M., Saleh, A., Singh, V. K., Akram, F., Radeva, P., & Puig, D. (2017). FoodPlaces: Learning deep features for food related scene understanding. In I. Aguilo, C. Angulo, R. Alquezar, A. Ortiz, & J. Torrens (Eds.), Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017 (pp. 156-165). IOS Press BV. https://doi.org/10.3233/978-1-61499-806-8-156

Sarker, Md Mostafa Kamal ; Leyva, Maria ; Saleh, Adel et al. / FoodPlaces : Learning deep features for food related scene understanding. Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017. editor / Isabel Aguilo ; Cecilio Angulo ; Rene Alquezar ; Alberto Ortiz ; Joan Torrens. IOS Press BV, 2017. pp. 156-165 (Frontiers in Artificial Intelligence and Applications, Vol. 300).

@inproceedings{53dc88d03dbe481db7e758c6097bc28b,

title = "FoodPlaces: Learning deep features for food related scene understanding",

abstract = "In human smart nutrition systems, environment based food classification has become popular to help analyzing the food intake based on the nutrition related activity. In this paper, we address the problem of food related environments, which refer to different eating places such as, bars, restaurants, coffee shops, etc. using state-of-the-art convolutional neural networks (CNNs). We collected a new dataset on different food related environments by integrating three publicly available datasets: Places365, ImageNet and SUN397. We have named it “FoodPlaces” and it contains 35 different types of classes. In order to achieve satisfactory results on the food related environment recognition, we fine-tuned several state-of-the-art CNNs, such as VGG16, RsNet50 and InceptionV3 using different transfer learning approaches. The results show that the fully fine-tunned InceptionV3 yields 75.22% classification accuracy among the discussed state-of-the-art CNNs.",

author = "Sarker, {Md Mostafa Kamal} and Maria Leyva and Adel Saleh and Singh, {Vivek Kumar} and Farhan Akram and Petia Radeva and Domenec Puig",

note = "Funding Information: This research is funded by the program Marti Franques under the agreement between Universitat Rovira Virgili and Fundaci Catalunya La Pedrera. Publisher Copyright: {\textcopyright} 2017 The authors and IOS Press. All rights reserved.; 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017 ; Conference date: 25-10-2017 Through 27-10-2017",

year = "2017",

doi = "10.3233/978-1-61499-806-8-156",

language = "English",

series = "Frontiers in Artificial Intelligence and Applications",

publisher = "IOS Press BV",

pages = "156--165",

editor = "Isabel Aguilo and Cecilio Angulo and Rene Alquezar and Alberto Ortiz and Joan Torrens",

booktitle = "Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017",

address = "Netherlands",

}

Sarker, MMK, Leyva, M, Saleh, A, Singh, VK, Akram, F, Radeva, P & Puig, D 2017, FoodPlaces: Learning deep features for food related scene understanding. in I Aguilo, C Angulo, R Alquezar, A Ortiz & J Torrens (eds), Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017. IOS Press BV, Frontiers in Artificial Intelligence and Applications, vol. 300, pp. 156-165, 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017, Deltebre, Terres de l'Ebre, Spain, 25/10/17. https://doi.org/10.3233/978-1-61499-806-8-156

FoodPlaces: Learning deep features for food related scene understanding. / Sarker, Md Mostafa Kamal; Leyva, Maria; Saleh, Adel et al.
Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017. ed. / Isabel Aguilo; Cecilio Angulo; Rene Alquezar; Alberto Ortiz; Joan Torrens. IOS Press BV, 2017. p. 156-165 (Frontiers in Artificial Intelligence and Applications, Vol. 300).

Research output: Chapter/Conference proceeding › Conference proceeding › Academic › peer-review

TY - GEN

T1 - FoodPlaces

T2 - 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017

AU - Sarker, Md Mostafa Kamal

AU - Leyva, Maria

AU - Saleh, Adel

AU - Singh, Vivek Kumar

AU - Akram, Farhan

AU - Radeva, Petia

AU - Puig, Domenec

N1 - Funding Information: This research is funded by the program Marti Franques under the agreement between Universitat Rovira Virgili and Fundaci Catalunya La Pedrera. Publisher Copyright: © 2017 The authors and IOS Press. All rights reserved.

PY - 2017

Y1 - 2017

N2 - In human smart nutrition systems, environment based food classification has become popular to help analyzing the food intake based on the nutrition related activity. In this paper, we address the problem of food related environments, which refer to different eating places such as, bars, restaurants, coffee shops, etc. using state-of-the-art convolutional neural networks (CNNs). We collected a new dataset on different food related environments by integrating three publicly available datasets: Places365, ImageNet and SUN397. We have named it “FoodPlaces” and it contains 35 different types of classes. In order to achieve satisfactory results on the food related environment recognition, we fine-tuned several state-of-the-art CNNs, such as VGG16, RsNet50 and InceptionV3 using different transfer learning approaches. The results show that the fully fine-tunned InceptionV3 yields 75.22% classification accuracy among the discussed state-of-the-art CNNs.

AB - In human smart nutrition systems, environment based food classification has become popular to help analyzing the food intake based on the nutrition related activity. In this paper, we address the problem of food related environments, which refer to different eating places such as, bars, restaurants, coffee shops, etc. using state-of-the-art convolutional neural networks (CNNs). We collected a new dataset on different food related environments by integrating three publicly available datasets: Places365, ImageNet and SUN397. We have named it “FoodPlaces” and it contains 35 different types of classes. In order to achieve satisfactory results on the food related environment recognition, we fine-tuned several state-of-the-art CNNs, such as VGG16, RsNet50 and InceptionV3 using different transfer learning approaches. The results show that the fully fine-tunned InceptionV3 yields 75.22% classification accuracy among the discussed state-of-the-art CNNs.

UR - http://www.scopus.com/inward/record.url?scp=85034212940&partnerID=8YFLogxK

U2 - 10.3233/978-1-61499-806-8-156

DO - 10.3233/978-1-61499-806-8-156

M3 - Conference proceeding

AN - SCOPUS:85034212940

T3 - Frontiers in Artificial Intelligence and Applications

SP - 156

EP - 165

BT - Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017

A2 - Aguilo, Isabel

A2 - Angulo, Cecilio

A2 - Alquezar, Rene

A2 - Ortiz, Alberto

A2 - Torrens, Joan

PB - IOS Press BV

Y2 - 25 October 2017 through 27 October 2017

ER -

Sarker MMK, Leyva M, Saleh A, Singh VK, Akram F, Radeva P et al. FoodPlaces: Learning deep features for food related scene understanding. In Aguilo I, Angulo C, Alquezar R, Ortiz A, Torrens J, editors, Recent Advances in Artificial Intelligence Research and Development - Proceedings of the 20th International Conference of the Catalan Association for Artificial Intelligence, CCIA 2017. IOS Press BV. 2017. p. 156-165. (Frontiers in Artificial Intelligence and Applications, Vol. 300). doi: 10.3233/978-1-61499-806-8-156

FoodPlaces: Learning deep features for food related scene understanding

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this