Detecting, classifying, and mapping retail storefronts using street-level imagery

Shahin Sharifi Noorian; Sihang Qiu; Achilleas Psyllidis; Alessandro Bozzon; Geert Jan Houben

doi:10.1145/3372278.3390706

Detecting, classifying, and mapping retail storefronts using street-level imagery

Shahin Sharifi Noorian, Sihang Qiu, Achilleas Psyllidis, Alessandro Bozzon, Geert Jan Houben

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

12 Citations (Scopus)

169 Downloads (Pure)

Abstract

Up-to-date listings of retail stores and related building functions are challenging and costly to maintain. We introduce a novel method for automatically detecting, geo-locating, and classifying retail stores and related commercial functions, on the basis of storefronts extracted from street-level imagery. Specifically, we present a deep learning approach that takes storefronts from street-level imagery as input, and directly provides the geo-location and type of commercial function as output. Our method showed a recall of 89.05% and a precision of 88.22% on a real-world dataset of street-level images, which experimentally demonstrated that our approach achieves human-level accuracy while having a remarkable run-time efficiency compared to methods such as Faster Region-Convolutional Neural Networks (Faster R-CNN) and Single Shot Detector (SSD).

Original language	English
Title of host publication	ICMR 2020 - Proceedings of the 2020 International Conference on Multimedia Retrieval
Publisher	Association for Computing Machinery (ACM)
Pages	495-501
Number of pages	7
ISBN (Electronic)	978-1-4503-7087-5
DOIs	https://doi.org/10.1145/3372278.3390706
Publication status	Published - 2020
Event	10th ACM International Conference on Multimedia Retrieval - Dublin, Ireland Duration: 26 Jun 2020 → 29 Jun 2020 http://www.icmr2020.org/

Conference

Conference	10th ACM International Conference on Multimedia Retrieval
Abbreviated title	ICMR 2020
Country/Territory	Ireland
City	Dublin
Period	26/06/20 → 29/06/20
Internet address	http://www.icmr2020.org/

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Convolutional neural networks
Object detection
Street-level imagery
Urban data extraction

Access to Document

10.1145/3372278.3390706

3372278.3390706Final published version, 2.32 MB

Cite this

@inproceedings{6e99ca5f41184d4086e126bf71b1a6eb,

title = "Detecting, classifying, and mapping retail storefronts using street-level imagery",

abstract = "Up-to-date listings of retail stores and related building functions are challenging and costly to maintain. We introduce a novel method for automatically detecting, geo-locating, and classifying retail stores and related commercial functions, on the basis of storefronts extracted from street-level imagery. Specifically, we present a deep learning approach that takes storefronts from street-level imagery as input, and directly provides the geo-location and type of commercial function as output. Our method showed a recall of 89.05% and a precision of 88.22% on a real-world dataset of street-level images, which experimentally demonstrated that our approach achieves human-level accuracy while having a remarkable run-time efficiency compared to methods such as Faster Region-Convolutional Neural Networks (Faster R-CNN) and Single Shot Detector (SSD).",

keywords = "Convolutional neural networks, Object detection, Street-level imagery, Urban data extraction",

author = "{Sharifi Noorian}, Shahin and Sihang Qiu and Achilleas Psyllidis and Alessandro Bozzon and Houben, {Geert Jan}",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 10th ACM International Conference on Multimedia Retrieval, ICMR 2020 ; Conference date: 26-06-2020 Through 29-06-2020",

year = "2020",

doi = "10.1145/3372278.3390706",

language = "English",

pages = "495--501",

booktitle = "ICMR 2020 - Proceedings of the 2020 International Conference on Multimedia Retrieval",

publisher = "Association for Computing Machinery (ACM)",

address = "United States",

url = "http://www.icmr2020.org/",

}

Sharifi Noorian, S, Qiu, S, Psyllidis, A , Bozzon, A & Houben, GJ 2020, Detecting, classifying, and mapping retail storefronts using street-level imagery. in ICMR 2020 - Proceedings of the 2020 International Conference on Multimedia Retrieval. Association for Computing Machinery (ACM), pp. 495-501, 10th ACM International Conference on Multimedia Retrieval, Dublin, Ireland, 26/06/20. https://doi.org/10.1145/3372278.3390706

Detecting, classifying, and mapping retail storefronts using street-level imagery. / Sharifi Noorian, Shahin; Qiu, Sihang; Psyllidis, Achilleas et al.
ICMR 2020 - Proceedings of the 2020 International Conference on Multimedia Retrieval. Association for Computing Machinery (ACM), 2020. p. 495-501.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Detecting, classifying, and mapping retail storefronts using street-level imagery

AU - Sharifi Noorian, Shahin

AU - Qiu, Sihang

AU - Psyllidis, Achilleas

AU - Bozzon, Alessandro

AU - Houben, Geert Jan

N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2020

Y1 - 2020

N2 - Up-to-date listings of retail stores and related building functions are challenging and costly to maintain. We introduce a novel method for automatically detecting, geo-locating, and classifying retail stores and related commercial functions, on the basis of storefronts extracted from street-level imagery. Specifically, we present a deep learning approach that takes storefronts from street-level imagery as input, and directly provides the geo-location and type of commercial function as output. Our method showed a recall of 89.05% and a precision of 88.22% on a real-world dataset of street-level images, which experimentally demonstrated that our approach achieves human-level accuracy while having a remarkable run-time efficiency compared to methods such as Faster Region-Convolutional Neural Networks (Faster R-CNN) and Single Shot Detector (SSD).

AB - Up-to-date listings of retail stores and related building functions are challenging and costly to maintain. We introduce a novel method for automatically detecting, geo-locating, and classifying retail stores and related commercial functions, on the basis of storefronts extracted from street-level imagery. Specifically, we present a deep learning approach that takes storefronts from street-level imagery as input, and directly provides the geo-location and type of commercial function as output. Our method showed a recall of 89.05% and a precision of 88.22% on a real-world dataset of street-level images, which experimentally demonstrated that our approach achieves human-level accuracy while having a remarkable run-time efficiency compared to methods such as Faster Region-Convolutional Neural Networks (Faster R-CNN) and Single Shot Detector (SSD).

KW - Convolutional neural networks

KW - Object detection

KW - Street-level imagery

KW - Urban data extraction

UR - http://www.scopus.com/inward/record.url?scp=85086894625&partnerID=8YFLogxK

U2 - 10.1145/3372278.3390706

DO - 10.1145/3372278.3390706

M3 - Conference contribution

AN - SCOPUS:85086894625

SP - 495

EP - 501

BT - ICMR 2020 - Proceedings of the 2020 International Conference on Multimedia Retrieval

PB - Association for Computing Machinery (ACM)

T2 - 10th ACM International Conference on Multimedia Retrieval

Y2 - 26 June 2020 through 29 June 2020

ER -

Detecting, classifying, and mapping retail storefronts using street-level imagery

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this