Multi-label building functions classification from ground pictures using convolutional neural networks

S. Srivastava; John E. Vargas Muñoz; David Swinkels; D. Tuia

doi:10.1145/3281548.3281559

Multi-label building functions classification from ground pictures using convolutional neural networks

S. Srivastava
, John E. Vargas Muñoz
, David Swinkels
, D. Tuia

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

29 Citations (Scopus)

Abstract

We approach the problem of multi building function classification for buildings from the city of Amsterdam using a collection of Google Street View (GSV) pictures acquired at multiple zoom levels (field of views, FoV) and the corresponding governmental census data per building. Since buildings can have multiple usages, we cast the problem as multilabel classification task. To do so, we trained a CNN model end-to-end with the task of predicting multiple co-occurring building function classes per building. We fuse the individual features of three FoVs by using volumetric stacking. Our proposed model outperforms baseline CNN models that use either single or multiple FoVs.

Original language	English
Title of host publication	Proceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery
Place of Publication	New York
Publisher	Association for Computing Machinery (ACM)
Pages	43-46
ISBN (Print)	9781450360364
DOIs	https://doi.org/10.1145/3281548.3281559
Publication status	Published - 6 Nov 2018
Event	2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery - Seattle, United States Duration: 6 Nov 2018 → 6 Nov 2018 https://udi.ornl.gov/geoai

Conference/symposium

Conference/symposium	2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery
Abbreviated title	GeoAI'18
Country/Territory	United States
City	Seattle
Period	6/11/18 → 6/11/18
Internet address	https://udi.ornl.gov/geoai

Keywords

Building Function
Convolutional Neural Network
Google Street View
Multi-Label Classification
Urban space

Access to Document

10.1145/3281548.3281559

Cite this

Srivastava, S., Vargas Muñoz, J. E., Swinkels, D., & Tuia, D. (2018). Multi-label building functions classification from ground pictures using convolutional neural networks. In Proceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery (pp. 43-46). Association for Computing Machinery (ACM). https://doi.org/10.1145/3281548.3281559

@inproceedings{ff0605f6e6c1486fbea45b78aa65bde7,

title = "Multi-label building functions classification from ground pictures using convolutional neural networks",

abstract = "We approach the problem of multi building function classification for buildings from the city of Amsterdam using a collection of Google Street View (GSV) pictures acquired at multiple zoom levels (field of views, FoV) and the corresponding governmental census data per building. Since buildings can have multiple usages, we cast the problem as multilabel classification task. To do so, we trained a CNN model end-to-end with the task of predicting multiple co-occurring building function classes per building. We fuse the individual features of three FoVs by using volumetric stacking. Our proposed model outperforms baseline CNN models that use either single or multiple FoVs.",

keywords = "Building Function, Convolutional Neural Network, Google Street View, Multi-Label Classification, Urban space",

author = "S. Srivastava and \{Vargas Mu{\~n}oz\}, \{John E.\} and David Swinkels and D. Tuia",

year = "2018",

month = nov,

day = "6",

doi = "10.1145/3281548.3281559",

language = "English",

isbn = "9781450360364",

pages = "43--46",

booktitle = "Proceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery",

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

address = "United States",

note = "2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery ; Conference date: 06-11-2018 Through 06-11-2018",

url = "https://udi.ornl.gov/geoai",

}

Srivastava, S, Vargas Muñoz, JE, Swinkels, D & Tuia, D 2018, Multi-label building functions classification from ground pictures using convolutional neural networks. in Proceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery. Association for Computing Machinery (ACM), New York, pp. 43-46, 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery, Seattle, Washington, United States, 6/11/18. https://doi.org/10.1145/3281548.3281559

Multi-label building functions classification from ground pictures using convolutional neural networks. / Srivastava, S.; Vargas Muñoz, John E.; Swinkels, David et al.
Proceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery. New York: Association for Computing Machinery (ACM), 2018. p. 43-46.

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

TY - GEN

T1 - Multi-label building functions classification from ground pictures using convolutional neural networks

AU - Srivastava, S.

AU - Vargas Muñoz, John E.

AU - Swinkels, David

AU - Tuia, D.

PY - 2018/11/6

Y1 - 2018/11/6

N2 - We approach the problem of multi building function classification for buildings from the city of Amsterdam using a collection of Google Street View (GSV) pictures acquired at multiple zoom levels (field of views, FoV) and the corresponding governmental census data per building. Since buildings can have multiple usages, we cast the problem as multilabel classification task. To do so, we trained a CNN model end-to-end with the task of predicting multiple co-occurring building function classes per building. We fuse the individual features of three FoVs by using volumetric stacking. Our proposed model outperforms baseline CNN models that use either single or multiple FoVs.

AB - We approach the problem of multi building function classification for buildings from the city of Amsterdam using a collection of Google Street View (GSV) pictures acquired at multiple zoom levels (field of views, FoV) and the corresponding governmental census data per building. Since buildings can have multiple usages, we cast the problem as multilabel classification task. To do so, we trained a CNN model end-to-end with the task of predicting multiple co-occurring building function classes per building. We fuse the individual features of three FoVs by using volumetric stacking. Our proposed model outperforms baseline CNN models that use either single or multiple FoVs.

KW - Building Function

KW - Convolutional Neural Network

KW - Google Street View

KW - Multi-Label Classification

KW - Urban space

U2 - 10.1145/3281548.3281559

DO - 10.1145/3281548.3281559

M3 - Conference paper

SN - 9781450360364

SP - 43

EP - 46

BT - Proceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery

PB - Association for Computing Machinery (ACM)

CY - New York

T2 - 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery

Y2 - 6 November 2018 through 6 November 2018

ER -

Multi-label building functions classification from ground pictures using convolutional neural networks

Abstract

Conference/symposium

Keywords

Access to Document

Fingerprint

Cite this