2D visual tracking for sports UAV cinematography applications

Nikos Nikolaidis; Ioannis Mademlis; Orestis Zachariadis; Vasileios Mygdalis; Ioannis Pitas

doi:10.1109/GlobalSIP.2017.8308599

2D visual tracking for sports UAV cinematography applications

Nikos Nikolaidis, Ioannis Mademlis, Orestis Zachariadis, Vasileios Mygdalis, Ioannis Pitas

Department of Electrical & Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

6 Citations (Scopus)

394 Downloads (Pure)

Abstract

In this paper, we provide a preliminary study of basic requirements for autonomous UAV cinematography via 2D target tracking. Our contribution is two-fold. First, we develop a mathematical framework so as to determine hardware camera requirements (specifically, focal length), on a representative case study, i.e., orbiting a still or moving target. Second, we examine the on-board software requirements in order to successfully achieve autonomous target following. To this end, we evaluate the performance of state-of-the-art realtime 2D visual trackers in videos captured by commercial drones. Overall, it was found that state-of-the-art 2D visual trackers are dependable and fast enough to be used in drone cinematography, particularly when combined with periodic target re-detection. A proposed variant of the Staple tracker achieved the best balance between real-time performance and tracking accuracy, on a dataset composed of 31 sports videos recorded by commercial drones.

Original language	English
Title of host publication	2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP2017)
Subtitle of host publication	Proceedings of a meeting held 14-16 November 2017, Montreal, Quebec, Canada
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	36-40
Number of pages	5
ISBN (Electronic)	9781509059904
ISBN (Print)	9781509059911
DOIs	https://doi.org/10.1109/GlobalSIP.2017.8308599
Publication status	Published - Apr 2018
Event	5th IEEE Global Conference on Signal and Information Processing, GlobalSIP 2017 - Montreal, Canada Duration: 14 Nov 2017 → 16 Nov 2017

Conference

Conference	5th IEEE Global Conference on Signal and Information Processing, GlobalSIP 2017
Country/Territory	Canada
City	Montreal
Period	14/11/17 → 16/11/17

Keywords

target following
target tracking
UAV cinematography
UAV shot types
video tracking

Access to Document

10.1109/GlobalSIP.2017.8308599

Full-text PDF (accepted author manuscript)
This is the accepted author manuscript (AAM). The final published version (version of record) is available online via IEEE at https://doi.org/10.1109/GlobalSIP.2017.8308599 . Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 354 KBLicence: Other

Handle.net

Persistent link

Cite this

Nikolaidis, N., Mademlis, I., Zachariadis, O., Mygdalis, V., & Pitas, I. (2018). 2D visual tracking for sports UAV cinematography applications. In 2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP2017): Proceedings of a meeting held 14-16 November 2017, Montreal, Quebec, Canada (pp. 36-40). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/GlobalSIP.2017.8308599

@inproceedings{c8ea02f12f7a436aba3c67b84e8d23aa,

title = "2D visual tracking for sports UAV cinematography applications",

abstract = "In this paper, we provide a preliminary study of basic requirements for autonomous UAV cinematography via 2D target tracking. Our contribution is two-fold. First, we develop a mathematical framework so as to determine hardware camera requirements (specifically, focal length), on a representative case study, i.e., orbiting a still or moving target. Second, we examine the on-board software requirements in order to successfully achieve autonomous target following. To this end, we evaluate the performance of state-of-the-art realtime 2D visual trackers in videos captured by commercial drones. Overall, it was found that state-of-the-art 2D visual trackers are dependable and fast enough to be used in drone cinematography, particularly when combined with periodic target re-detection. A proposed variant of the Staple tracker achieved the best balance between real-time performance and tracking accuracy, on a dataset composed of 31 sports videos recorded by commercial drones.",

keywords = "target following, target tracking, UAV cinematography, UAV shot types, video tracking",

author = "Nikos Nikolaidis and Ioannis Mademlis and Orestis Zachariadis and Vasileios Mygdalis and Ioannis Pitas",

year = "2018",

month = apr,

doi = "10.1109/GlobalSIP.2017.8308599",

language = "English",

isbn = "9781509059911",

pages = "36--40",

booktitle = "2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP2017)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "5th IEEE Global Conference on Signal and Information Processing, GlobalSIP 2017 ; Conference date: 14-11-2017 Through 16-11-2017",

}

Nikolaidis, N, Mademlis, I, Zachariadis, O, Mygdalis, V & Pitas, I 2018, 2D visual tracking for sports UAV cinematography applications. in 2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP2017): Proceedings of a meeting held 14-16 November 2017, Montreal, Quebec, Canada. Institute of Electrical and Electronics Engineers (IEEE), pp. 36-40, 5th IEEE Global Conference on Signal and Information Processing, GlobalSIP 2017, Montreal, Canada, 14/11/17. https://doi.org/10.1109/GlobalSIP.2017.8308599

2D visual tracking for sports UAV cinematography applications. / Nikolaidis, Nikos; Mademlis, Ioannis; Zachariadis, Orestis et al.
2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP2017): Proceedings of a meeting held 14-16 November 2017, Montreal, Quebec, Canada. Institute of Electrical and Electronics Engineers (IEEE), 2018. p. 36-40.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - 2D visual tracking for sports UAV cinematography applications

AU - Nikolaidis, Nikos

AU - Mademlis, Ioannis

AU - Zachariadis, Orestis

AU - Mygdalis, Vasileios

AU - Pitas, Ioannis

PY - 2018/4

Y1 - 2018/4

N2 - In this paper, we provide a preliminary study of basic requirements for autonomous UAV cinematography via 2D target tracking. Our contribution is two-fold. First, we develop a mathematical framework so as to determine hardware camera requirements (specifically, focal length), on a representative case study, i.e., orbiting a still or moving target. Second, we examine the on-board software requirements in order to successfully achieve autonomous target following. To this end, we evaluate the performance of state-of-the-art realtime 2D visual trackers in videos captured by commercial drones. Overall, it was found that state-of-the-art 2D visual trackers are dependable and fast enough to be used in drone cinematography, particularly when combined with periodic target re-detection. A proposed variant of the Staple tracker achieved the best balance between real-time performance and tracking accuracy, on a dataset composed of 31 sports videos recorded by commercial drones.

AB - In this paper, we provide a preliminary study of basic requirements for autonomous UAV cinematography via 2D target tracking. Our contribution is two-fold. First, we develop a mathematical framework so as to determine hardware camera requirements (specifically, focal length), on a representative case study, i.e., orbiting a still or moving target. Second, we examine the on-board software requirements in order to successfully achieve autonomous target following. To this end, we evaluate the performance of state-of-the-art realtime 2D visual trackers in videos captured by commercial drones. Overall, it was found that state-of-the-art 2D visual trackers are dependable and fast enough to be used in drone cinematography, particularly when combined with periodic target re-detection. A proposed variant of the Staple tracker achieved the best balance between real-time performance and tracking accuracy, on a dataset composed of 31 sports videos recorded by commercial drones.

KW - target following

KW - target tracking

KW - UAV cinematography

KW - UAV shot types

KW - video tracking

UR - http://www.scopus.com/inward/record.url?scp=85048031483&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2017.8308599

DO - 10.1109/GlobalSIP.2017.8308599

M3 - Conference Contribution (Conference Proceeding)

AN - SCOPUS:85048031483

SN - 9781509059911

SP - 36

EP - 40

BT - 2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP2017)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 5th IEEE Global Conference on Signal and Information Processing, GlobalSIP 2017

Y2 - 14 November 2017 through 16 November 2017

ER -

Nikolaidis N, Mademlis I, Zachariadis O, Mygdalis V, Pitas I. 2D visual tracking for sports UAV cinematography applications. In 2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP2017): Proceedings of a meeting held 14-16 November 2017, Montreal, Quebec, Canada. Institute of Electrical and Electronics Engineers (IEEE). 2018. p. 36-40 Epub 2018 Mar 8. doi: 10.1109/GlobalSIP.2017.8308599

2D visual tracking for sports UAV cinematography applications

Abstract

Conference

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this