Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams

Mostafa Kamal Sarker; Hatem A Rashwan; Estefanía Talavera; Syeda Furruka Banu; Petia Radeva; Domenec Puig

doi:10.1007/978-3-030-11021-5_26

Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams

Mostafa Kamal Sarker, Hatem A Rashwan, Estefanía Talavera, Syeda Furruka Banu, Petia Radeva, Domenec Puig

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

5 Citations (Scopus)

64 Downloads (Pure)

Abstract

First-person (wearable) camera continually captures unscripted interactions of the camera user with objects, people, and scenes reflecting his personal and relational tendencies. One of the preferences of people is their interaction with food events. The regulation of food intake and its duration has a great importance to protect against diseases. Consequently, this work aims to develop a smart model that is able to determine the recurrences of a person on food places during a day. This model is based on a deep end-to-end model for automatic food places recognition by analyzing egocentric photo-streams. In this paper, we apply multi-scale Atrous convolution networks to extract the key features related to food places of the input images. The proposed model is evaluated on an in-house private dataset called “EgoFoodPlaces”. Experimental results shows promising results of food places classification in egocentric photo-streams.

Original language	English
Title of host publication	Computer Vision - ECCV 2018 Workshops
Subtitle of host publication	Munich, Germany, September 8-14, 2018, Proceedings, Part V
Editors	Laura Leal-Taixé, Stefan Roth
Place of Publication	Cham, Switzerland
Publisher	Springer
Pages	423-433
Number of pages	11
ISBN (Electronic)	978-3-030-11021-5
ISBN (Print)	978-3-030-11020-8
DOIs	https://doi.org/10.1007/978-3-030-11021-5_26
Publication status	Published - 23 Jan 2019
Event	15th European Conference on Computer Vision, ECCV 2018 - Munich, Germany Duration: 8 Sept 2018 → 14 Sept 2018 Conference number: 15

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	11133
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	15th European Conference on Computer Vision, ECCV 2018
Abbreviated title	ECCV 2018
Country/Territory	Germany
City	Munich
Period	8/09/18 → 14/09/18

Keywords

2022 OA procedure

Access to Document

10.1007/978-3-030-11021-5_26

Sarker2019_Chapter_MACNetMulti-scaleAtrousConvoluFinal published version, 1.67 MBLicence: Taverne

Cite this

Sarker, M. K., Rashwan, H. A., Talavera, E., Banu, S. F., Radeva, P., & Puig, D. (2019). Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams. In L. Leal-Taixé, & S. Roth (Eds.), Computer Vision - ECCV 2018 Workshops: Munich, Germany, September 8-14, 2018, Proceedings, Part V (pp. 423-433). (Lecture Notes in Computer Science; Vol. 11133). Springer. https://doi.org/10.1007/978-3-030-11021-5_26

Sarker, Mostafa Kamal ; Rashwan, Hatem A ; Talavera, Estefanía et al. / Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams. Computer Vision - ECCV 2018 Workshops: Munich, Germany, September 8-14, 2018, Proceedings, Part V. editor / Laura Leal-Taixé ; Stefan Roth. Cham, Switzerland : Springer, 2019. pp. 423-433 (Lecture Notes in Computer Science).

@inproceedings{30958ac9c2404a07a80f4b6006eeb9c5,

title = "Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams",

abstract = "First-person (wearable) camera continually captures unscripted interactions of the camera user with objects, people, and scenes reflecting his personal and relational tendencies. One of the preferences of people is their interaction with food events. The regulation of food intake and its duration has a great importance to protect against diseases. Consequently, this work aims to develop a smart model that is able to determine the recurrences of a person on food places during a day. This model is based on a deep end-to-end model for automatic food places recognition by analyzing egocentric photo-streams. In this paper, we apply multi-scale Atrous convolution networks to extract the key features related to food places of the input images. The proposed model is evaluated on an in-house private dataset called “EgoFoodPlaces”. Experimental results shows promising results of food places classification in egocentric photo-streams.",

keywords = "2022 OA procedure",

author = "Sarker, \{Mostafa Kamal\} and Rashwan, \{Hatem A\} and Estefan{\'i}a Talavera and Banu, \{Syeda Furruka\} and Petia Radeva and Domenec Puig",

year = "2019",

month = jan,

day = "23",

doi = "10.1007/978-3-030-11021-5\_26",

language = "English",

isbn = "978-3-030-11020-8",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "423--433",

editor = "Laura Leal-Taix{\'e} and Stefan Roth",

booktitle = "Computer Vision - ECCV 2018 Workshops",

address = "Germany",

note = "15th European Conference on Computer Vision, ECCV 2018, ECCV 2018 ; Conference date: 08-09-2018 Through 14-09-2018",

}

Sarker, MK, Rashwan, HA, Talavera, E, Banu, SF, Radeva, P & Puig, D 2019, Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams. in L Leal-Taixé & S Roth (eds), Computer Vision - ECCV 2018 Workshops: Munich, Germany, September 8-14, 2018, Proceedings, Part V. Lecture Notes in Computer Science, vol. 11133, Springer, Cham, Switzerland, pp. 423-433, 15th European Conference on Computer Vision, ECCV 2018, Munich, Germany, 8/09/18. https://doi.org/10.1007/978-3-030-11021-5_26

Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams. / Sarker, Mostafa Kamal; Rashwan, Hatem A; Talavera, Estefanía et al.
Computer Vision - ECCV 2018 Workshops: Munich, Germany, September 8-14, 2018, Proceedings, Part V. ed. / Laura Leal-Taixé; Stefan Roth. Cham, Switzerland: Springer, 2019. p. 423-433 (Lecture Notes in Computer Science; Vol. 11133).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams

AU - Sarker, Mostafa Kamal

AU - Rashwan, Hatem A

AU - Talavera, Estefanía

AU - Banu, Syeda Furruka

AU - Radeva, Petia

AU - Puig, Domenec

N1 - Conference code: 15

PY - 2019/1/23

Y1 - 2019/1/23

N2 - First-person (wearable) camera continually captures unscripted interactions of the camera user with objects, people, and scenes reflecting his personal and relational tendencies. One of the preferences of people is their interaction with food events. The regulation of food intake and its duration has a great importance to protect against diseases. Consequently, this work aims to develop a smart model that is able to determine the recurrences of a person on food places during a day. This model is based on a deep end-to-end model for automatic food places recognition by analyzing egocentric photo-streams. In this paper, we apply multi-scale Atrous convolution networks to extract the key features related to food places of the input images. The proposed model is evaluated on an in-house private dataset called “EgoFoodPlaces”. Experimental results shows promising results of food places classification in egocentric photo-streams.

AB - First-person (wearable) camera continually captures unscripted interactions of the camera user with objects, people, and scenes reflecting his personal and relational tendencies. One of the preferences of people is their interaction with food events. The regulation of food intake and its duration has a great importance to protect against diseases. Consequently, this work aims to develop a smart model that is able to determine the recurrences of a person on food places during a day. This model is based on a deep end-to-end model for automatic food places recognition by analyzing egocentric photo-streams. In this paper, we apply multi-scale Atrous convolution networks to extract the key features related to food places of the input images. The proposed model is evaluated on an in-house private dataset called “EgoFoodPlaces”. Experimental results shows promising results of food places classification in egocentric photo-streams.

KW - 2022 OA procedure

U2 - 10.1007/978-3-030-11021-5_26

DO - 10.1007/978-3-030-11021-5_26

M3 - Conference contribution

SN - 978-3-030-11020-8

T3 - Lecture Notes in Computer Science

SP - 423

EP - 433

BT - Computer Vision - ECCV 2018 Workshops

A2 - Leal-Taixé, Laura

A2 - Roth, Stefan

PB - Springer

CY - Cham, Switzerland

T2 - 15th European Conference on Computer Vision, ECCV 2018

Y2 - 8 September 2018 through 14 September 2018

ER -

Sarker MK, Rashwan HA, Talavera E, Banu SF, Radeva P, Puig D. Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams. In Leal-Taixé L, Roth S, editors, Computer Vision - ECCV 2018 Workshops: Munich, Germany, September 8-14, 2018, Proceedings, Part V. Cham, Switzerland: Springer. 2019. p. 423-433. (Lecture Notes in Computer Science). doi: 10.1007/978-3-030-11021-5_26

Macnet: Multi-scale atrous convolution networks for food places classification in egocentric photo-streams

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this