Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark

Bohao Huang; Kangkang Lu; Nicolas Audebert; Andrew Khalel; Yuliya Tarabalka; Jordan Malof; Alexandre Boulch; Bertrand Le Saux; Leslie Collins; Kyle Bradbury; Sébastien Lefèvre; Motaz El-Saban

doi:10.1109/IGARSS.2018.8518525

Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark

Bohao Huang, Kangkang Lu, Nicolas Audebert, Andrew Khalel, Yuliya Tarabalka, Jordan Malof, Alexandre Boulch, Bertrand Le Saux, Leslie Collins, Kyle Bradbury, Sébastien Lefèvre, Motaz El-Saban

Computer Science

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

91 Scopus citations

Abstract

Over the recent years, there has been an increasing interest in large-scale classification of remote sensing images. In this context, the Inria Aerial Image Labeling Benchmark has been released online in December 2016. In this paper, we discuss the outcomes of the first year of the benchmark contest, which consisted in dense labeling of aerial images into building/not building classes, covering areas of five cities not present in the training set. We present four methods with the highest numerical accuracies, all four being convolutional neural network approaches. It is remarkable that three of these methods use the U-net architecture, which has thus proven to become a new standard in image dense labeling.

Original language	English
Title of host publication	2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	6947-6950
Number of pages	4
ISBN (Electronic)	9781538671504
DOIs	https://doi.org/10.1109/IGARSS.2018.8518525
State	Published - Oct 31 2018
Event	38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Valencia, Spain Duration: Jul 22 2018 → Jul 27 2018

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)
Volume	2018-July

Conference

Conference	38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018
Country/Territory	Spain
City	Valencia
Period	07/22/18 → 07/27/18

Keywords

Aerial images
Classification benchmark
Convolutional neural networks
Deep learning
U-net

Access to Document

10.1109/IGARSS.2018.8518525

Cite this

Huang, B., Lu, K., Audebert, N., Khalel, A., Tarabalka, Y., Malof, J., Boulch, A., Saux, B. L., Collins, L., Bradbury, K., Lefèvre, S., & El-Saban, M. (2018). Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark. In 2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings (pp. 6947-6950). Article 8518525 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2018-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IGARSS.2018.8518525

Huang, Bohao ; Lu, Kangkang ; Audebert, Nicolas et al. / Large-scale semantic classification : Outcome of the first year of inria aerial image labeling benchmark. 2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 6947-6950 (International Geoscience and Remote Sensing Symposium (IGARSS)).

@inproceedings{753ebb626f054e81b046f241c1825cc4,

title = "Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark",

abstract = "Over the recent years, there has been an increasing interest in large-scale classification of remote sensing images. In this context, the Inria Aerial Image Labeling Benchmark has been released online in December 2016. In this paper, we discuss the outcomes of the first year of the benchmark contest, which consisted in dense labeling of aerial images into building/not building classes, covering areas of five cities not present in the training set. We present four methods with the highest numerical accuracies, all four being convolutional neural network approaches. It is remarkable that three of these methods use the U-net architecture, which has thus proven to become a new standard in image dense labeling.",

keywords = "Aerial images, Classification benchmark, Convolutional neural networks, Deep learning, U-net",

author = "Bohao Huang and Kangkang Lu and Nicolas Audebert and Andrew Khalel and Yuliya Tarabalka and Jordan Malof and Alexandre Boulch and Saux, {Bertrand Le} and Leslie Collins and Kyle Bradbury and S{\'e}bastien Lef{\`e}vre and Motaz El-Saban",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 ; Conference date: 22-07-2018 Through 27-07-2018",

year = "2018",

month = oct,

day = "31",

doi = "10.1109/IGARSS.2018.8518525",

language = "English",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "6947--6950",

booktitle = "2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings",

}

Huang, B, Lu, K, Audebert, N, Khalel, A, Tarabalka, Y, Malof, J, Boulch, A, Saux, BL, Collins, L, Bradbury, K, Lefèvre, S & El-Saban, M 2018, Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark. in 2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings., 8518525, International Geoscience and Remote Sensing Symposium (IGARSS), vol. 2018-July, Institute of Electrical and Electronics Engineers Inc., pp. 6947-6950, 38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018, Valencia, Spain, 07/22/18. https://doi.org/10.1109/IGARSS.2018.8518525

Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark. / Huang, Bohao; Lu, Kangkang; Audebert, Nicolas et al.
2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 6947-6950 8518525 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2018-July).

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

TY - GEN

T1 - Large-scale semantic classification

T2 - 38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018

AU - Huang, Bohao

AU - Lu, Kangkang

AU - Audebert, Nicolas

AU - Khalel, Andrew

AU - Tarabalka, Yuliya

AU - Malof, Jordan

AU - Boulch, Alexandre

AU - Saux, Bertrand Le

AU - Collins, Leslie

AU - Bradbury, Kyle

AU - Lefèvre, Sébastien

AU - El-Saban, Motaz

PY - 2018/10/31

Y1 - 2018/10/31

N2 - Over the recent years, there has been an increasing interest in large-scale classification of remote sensing images. In this context, the Inria Aerial Image Labeling Benchmark has been released online in December 2016. In this paper, we discuss the outcomes of the first year of the benchmark contest, which consisted in dense labeling of aerial images into building/not building classes, covering areas of five cities not present in the training set. We present four methods with the highest numerical accuracies, all four being convolutional neural network approaches. It is remarkable that three of these methods use the U-net architecture, which has thus proven to become a new standard in image dense labeling.

AB - Over the recent years, there has been an increasing interest in large-scale classification of remote sensing images. In this context, the Inria Aerial Image Labeling Benchmark has been released online in December 2016. In this paper, we discuss the outcomes of the first year of the benchmark contest, which consisted in dense labeling of aerial images into building/not building classes, covering areas of five cities not present in the training set. We present four methods with the highest numerical accuracies, all four being convolutional neural network approaches. It is remarkable that three of these methods use the U-net architecture, which has thus proven to become a new standard in image dense labeling.

KW - Aerial images

KW - Classification benchmark

KW - Convolutional neural networks

KW - Deep learning

KW - U-net

UR - http://www.scopus.com/inward/record.url?scp=85060884859&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2018.8518525

DO - 10.1109/IGARSS.2018.8518525

M3 - Conference contribution

AN - SCOPUS:85060884859

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 6947

EP - 6950

BT - 2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 22 July 2018 through 27 July 2018

ER -

Huang B, Lu K, Audebert N, Khalel A, Tarabalka Y, Malof J et al. Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark. In 2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 6947-6950. 8518525. (International Geoscience and Remote Sensing Symposium (IGARSS)). doi: 10.1109/IGARSS.2018.8518525

Large-scale semantic classification: Outcome of the first year of inria aerial image labeling benchmark

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this