Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles

Andreas  Kamilaris; Corjan  van den Brink; Savvas Karatsiolis

doi:10.1007/978-3-030-29930-9_8

Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles

Andreas Kamilaris, Corjan van den Brink, Savvas Karatsiolis

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

7 Citations (Scopus)

88 Downloads (Pure)

Abstract

This paper describes preliminary work in the recent promising approach of generating synthetic training data for facilitating the learning procedure of deep learning (DL) models, with a focus on aerial photos produced by unmanned aerial vehicles (UAV). The general concept and methodology are described, and preliminary results are presented, based on a classification problem of fire identification in forests as well as a counting problem of estimating number of houses in urban areas. The proposed technique constitutes a new possibility for the DL community, especially related to UAV-based imagery analysis, with much potential, promising results, and unexplored ground for further research.

Original language	English
Title of host publication	Computer Analysis of Images and Patterns
Subtitle of host publication	CAIP 2019 International Workshops, ViMaBi and DL-UAV, Salerno, Italy, September 6, 2019, Proceedings
Editors	Mario Vento, Gennaro Percannella, Sara Colantonio, Daniela Giorgi, Bogdan J. Matuszewski, Hamideh Kerdegari, Manzoor Razaak
Place of Publication	Cham
Publisher	Springer
Pages	81-90
ISBN (Electronic)	978-3-030-29930-9
ISBN (Print)	978-3-030-29929-3
DOIs	https://doi.org/10.1007/978-3-030-29930-9_8
Publication status	Published - 23 Aug 2019
Event	Workshop on Deep-Learning Based Computer Vision for UAV, DL-UAV19 - Grand Hotel Salerno, Salerno, Italy Duration: 6 Sept 2019 → 6 Sept 2019

Publication series

Name	Communications in Computer and Information Science
Publisher	Springer
Volume	1089
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Workshop

Workshop	Workshop on Deep-Learning Based Computer Vision for UAV, DL-UAV19
Abbreviated title	DL-UAV19
Country/Territory	Italy
City	Salerno
Period	6/09/19 → 6/09/19

Keywords

UAV
Deep learning
Generative data
Aerial imagery

Access to Document

10.1007/978-3-030-29930-9_8

Kamilaris2019trainingFinal published version, 1.1 MBLicence: Taverne

7 Citations
1 Preprint

Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles
Kamilaris, A., van den Brink, C. & Karatsiolis, S., 18 Aug 2019, ArXiv.org.
Research output: Working paper › Preprint › Academic

Open Access
File
14 Downloads (Pure)

Cite this

Kamilaris, A., van den Brink, C., & Karatsiolis, S. (2019). Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles. In M. Vento, G. Percannella, S. Colantonio, D. Giorgi, B. J. Matuszewski, H. Kerdegari, & M. Razaak (Eds.), Computer Analysis of Images and Patterns : CAIP 2019 International Workshops, ViMaBi and DL-UAV, Salerno, Italy, September 6, 2019, Proceedings (pp. 81-90). (Communications in Computer and Information Science; Vol. 1089). Springer. https://doi.org/10.1007/978-3-030-29930-9_8

Kamilaris, Andreas ; van den Brink, Corjan ; Karatsiolis, Savvas. / Training Deep Learning Models via Synthetic Data : Application in Unmanned Aerial Vehicles. Computer Analysis of Images and Patterns : CAIP 2019 International Workshops, ViMaBi and DL-UAV, Salerno, Italy, September 6, 2019, Proceedings. editor / Mario Vento ; Gennaro Percannella ; Sara Colantonio ; Daniela Giorgi ; Bogdan J. Matuszewski ; Hamideh Kerdegari ; Manzoor Razaak. Cham : Springer, 2019. pp. 81-90 (Communications in Computer and Information Science).

@inproceedings{67b01aa0eff142e189f9aded4740ff81,

title = "Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles",

abstract = "This paper describes preliminary work in the recent promising approach of generating synthetic training data for facilitating the learning procedure of deep learning (DL) models, with a focus on aerial photos produced by unmanned aerial vehicles (UAV). The general concept and methodology are described, and preliminary results are presented, based on a classification problem of fire identification in forests as well as a counting problem of estimating number of houses in urban areas. The proposed technique constitutes a new possibility for the DL community, especially related to UAV-based imagery analysis, with much potential, promising results, and unexplored ground for further research.",

keywords = "UAV, Deep learning, Generative data, Aerial imagery",

author = "Andreas Kamilaris and {van den Brink}, Corjan and Savvas Karatsiolis",

year = "2019",

month = aug,

day = "23",

doi = "10.1007/978-3-030-29930-9_8",

language = "English",

isbn = "978-3-030-29929-3",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "81--90",

editor = "Mario Vento and Gennaro Percannella and Sara Colantonio and Daniela Giorgi and Matuszewski, {Bogdan J.} and Hamideh Kerdegari and Manzoor Razaak",

booktitle = "Computer Analysis of Images and Patterns",

address = "Germany",

note = "Workshop on Deep-Learning Based Computer Vision for UAV, DL-UAV19, DL-UAV19 ; Conference date: 06-09-2019 Through 06-09-2019",

}

Kamilaris, A, van den Brink, C & Karatsiolis, S 2019, Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles. in M Vento, G Percannella, S Colantonio, D Giorgi, BJ Matuszewski, H Kerdegari & M Razaak (eds), Computer Analysis of Images and Patterns : CAIP 2019 International Workshops, ViMaBi and DL-UAV, Salerno, Italy, September 6, 2019, Proceedings. Communications in Computer and Information Science, vol. 1089, Springer, Cham, pp. 81-90, Workshop on Deep-Learning Based Computer Vision for UAV, DL-UAV19, Salerno, Italy, 6/09/19. https://doi.org/10.1007/978-3-030-29930-9_8

Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles. / Kamilaris, Andreas ; van den Brink, Corjan ; Karatsiolis, Savvas.
Computer Analysis of Images and Patterns : CAIP 2019 International Workshops, ViMaBi and DL-UAV, Salerno, Italy, September 6, 2019, Proceedings. ed. / Mario Vento; Gennaro Percannella; Sara Colantonio; Daniela Giorgi; Bogdan J. Matuszewski; Hamideh Kerdegari; Manzoor Razaak. Cham: Springer, 2019. p. 81-90 (Communications in Computer and Information Science; Vol. 1089).

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

TY - GEN

T1 - Training Deep Learning Models via Synthetic Data

T2 - Workshop on Deep-Learning Based Computer Vision for UAV, DL-UAV19

AU - Kamilaris, Andreas

AU - van den Brink, Corjan

AU - Karatsiolis, Savvas

PY - 2019/8/23

Y1 - 2019/8/23

N2 - This paper describes preliminary work in the recent promising approach of generating synthetic training data for facilitating the learning procedure of deep learning (DL) models, with a focus on aerial photos produced by unmanned aerial vehicles (UAV). The general concept and methodology are described, and preliminary results are presented, based on a classification problem of fire identification in forests as well as a counting problem of estimating number of houses in urban areas. The proposed technique constitutes a new possibility for the DL community, especially related to UAV-based imagery analysis, with much potential, promising results, and unexplored ground for further research.

AB - This paper describes preliminary work in the recent promising approach of generating synthetic training data for facilitating the learning procedure of deep learning (DL) models, with a focus on aerial photos produced by unmanned aerial vehicles (UAV). The general concept and methodology are described, and preliminary results are presented, based on a classification problem of fire identification in forests as well as a counting problem of estimating number of houses in urban areas. The proposed technique constitutes a new possibility for the DL community, especially related to UAV-based imagery analysis, with much potential, promising results, and unexplored ground for further research.

KW - UAV

KW - Deep learning

KW - Generative data

KW - Aerial imagery

U2 - 10.1007/978-3-030-29930-9_8

DO - 10.1007/978-3-030-29930-9_8

M3 - Conference contribution

SN - 978-3-030-29929-3

T3 - Communications in Computer and Information Science

SP - 81

EP - 90

BT - Computer Analysis of Images and Patterns

A2 - Vento, Mario

A2 - Percannella, Gennaro

A2 - Colantonio, Sara

A2 - Giorgi, Daniela

A2 - Matuszewski, Bogdan J.

A2 - Kerdegari, Hamideh

A2 - Razaak, Manzoor

PB - Springer

CY - Cham

Y2 - 6 September 2019 through 6 September 2019

ER -

Kamilaris A, van den Brink C, Karatsiolis S. Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles. In Vento M, Percannella G, Colantonio S, Giorgi D, Matuszewski BJ, Kerdegari H, Razaak M, editors, Computer Analysis of Images and Patterns : CAIP 2019 International Workshops, ViMaBi and DL-UAV, Salerno, Italy, September 6, 2019, Proceedings. Cham: Springer. 2019. p. 81-90. (Communications in Computer and Information Science). doi: 10.1007/978-3-030-29930-9_8

Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles

Abstract

Publication series

Workshop

Keywords

Access to Document

Fingerprint

Research output

Training Deep Learning Models via Synthetic Data: Application in Unmanned Aerial Vehicles

Cite this