An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric

Mark David Jenkins; Peter Barrie; Thomas Buggy; Gordon Morison

doi:10.1109/EDERC.2014.6924377

An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric

Mark David Jenkins
, Peter Barrie
, Thomas Buggy
, Gordon Morison

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

3 Citations (Scopus)

Abstract

This paper presents an extended algorithm for Real-time Compressive Tracking using Cosine Similarity Metric for object tracking. The method utilises a weighted multi-frame cosine similarity metric with the ground truth bounding box and a recently computed target bounding box. In comparison to the original algorithm it is capable of handling fast motion with a greater degree of accuracy. The proposed algorithm has been benchmarked on a desktop computer and subsequently implemented on a Texas Instruments ARM based DM3730 Beagleboard-xM. The proposed algorithm demonstrates a significant performance increase in fast motion video sequences. In addition, the low computational complexity of the algorithm makes it well suited for embedded applications.

Original language	English
Title of host publication	2014 6th European Embedded Design in Education and Research Conference (EDERC)
Editors	John J. Soraghan, Gaetano Di Caterina, Djordje Marinkovic, Nuria Llin, Daniel Wicks
Publisher	IEEE
Pages	147-151
Number of pages	5
ISBN (Electronic)	9781479968435
ISBN (Print)	9781479968411
DOIs	https://doi.org/10.1109/EDERC.2014.6924377
Publication status	Published - 16 Oct 2014
Event	6th European Embedded Design in Education and Research Conference (EDERC 2014) - Milan, Italy Duration: 11 Sept 2014 → 12 Sept 2014

Conference

Conference	6th European Embedded Design in Education and Research Conference (EDERC 2014)
Country/Territory	Italy
City	Milan
Period	11/09/14 → 12/09/14

Keywords

extended algorithm
real time compressive tracking
cosine similarity metric
fast motion video sequence

Documents and Links

10.1109/EDERC.2014.6924377

Cite this

Jenkins, M. D., Barrie, P., Buggy, T., & Morison, G. (2014). An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric. In J. J. Soraghan, G. Di Caterina, D. Marinkovic, N. Llin, & D. Wicks (Eds.), 2014 6th European Embedded Design in Education and Research Conference (EDERC) (pp. 147-151). IEEE. https://doi.org/10.1109/EDERC.2014.6924377

@inbook{9993cee4e1994364a8200e6294f4adb2,

title = "An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric",

abstract = "This paper presents an extended algorithm for Real-time Compressive Tracking using Cosine Similarity Metric for object tracking. The method utilises a weighted multi-frame cosine similarity metric with the ground truth bounding box and a recently computed target bounding box. In comparison to the original algorithm it is capable of handling fast motion with a greater degree of accuracy. The proposed algorithm has been benchmarked on a desktop computer and subsequently implemented on a Texas Instruments ARM based DM3730 Beagleboard-xM. The proposed algorithm demonstrates a significant performance increase in fast motion video sequences. In addition, the low computational complexity of the algorithm makes it well suited for embedded applications.",

keywords = "extended algorithm, real time compressive tracking, cosine similarity metric, fast motion video sequence",

author = "Jenkins, \{Mark David\} and Peter Barrie and Thomas Buggy and Gordon Morison",

year = "2014",

month = oct,

day = "16",

doi = "10.1109/EDERC.2014.6924377",

language = "English",

isbn = "9781479968411",

pages = "147--151",

editor = "Soraghan, \{John J.\} and \{Di Caterina\}, Gaetano and Djordje Marinkovic and Nuria Llin and Daniel Wicks",

booktitle = "2014 6th European Embedded Design in Education and Research Conference (EDERC)",

publisher = "IEEE",

address = "United States",

note = "6th European Embedded Design in Education and Research Conference (EDERC 2014) ; Conference date: 11-09-2014 Through 12-09-2014",

}

Jenkins, MD, Barrie, P, Buggy, T & Morison, G 2014, An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric. in JJ Soraghan, G Di Caterina, D Marinkovic, N Llin & D Wicks (eds), 2014 6th European Embedded Design in Education and Research Conference (EDERC). IEEE, pp. 147-151, 6th European Embedded Design in Education and Research Conference (EDERC 2014), Milan, Italy, 11/09/14. https://doi.org/10.1109/EDERC.2014.6924377

An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric. / Jenkins, Mark David; Barrie, Peter; Buggy, Thomas et al.
2014 6th European Embedded Design in Education and Research Conference (EDERC). ed. / John J. Soraghan; Gaetano Di Caterina; Djordje Marinkovic; Nuria Llin; Daniel Wicks. IEEE, 2014. p. 147-151.

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

TY - CHAP

T1 - An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric

AU - Jenkins, Mark David

AU - Barrie, Peter

AU - Buggy, Thomas

AU - Morison, Gordon

PY - 2014/10/16

Y1 - 2014/10/16

N2 - This paper presents an extended algorithm for Real-time Compressive Tracking using Cosine Similarity Metric for object tracking. The method utilises a weighted multi-frame cosine similarity metric with the ground truth bounding box and a recently computed target bounding box. In comparison to the original algorithm it is capable of handling fast motion with a greater degree of accuracy. The proposed algorithm has been benchmarked on a desktop computer and subsequently implemented on a Texas Instruments ARM based DM3730 Beagleboard-xM. The proposed algorithm demonstrates a significant performance increase in fast motion video sequences. In addition, the low computational complexity of the algorithm makes it well suited for embedded applications.

AB - This paper presents an extended algorithm for Real-time Compressive Tracking using Cosine Similarity Metric for object tracking. The method utilises a weighted multi-frame cosine similarity metric with the ground truth bounding box and a recently computed target bounding box. In comparison to the original algorithm it is capable of handling fast motion with a greater degree of accuracy. The proposed algorithm has been benchmarked on a desktop computer and subsequently implemented on a Texas Instruments ARM based DM3730 Beagleboard-xM. The proposed algorithm demonstrates a significant performance increase in fast motion video sequences. In addition, the low computational complexity of the algorithm makes it well suited for embedded applications.

KW - extended algorithm

KW - real time compressive tracking

KW - cosine similarity metric

KW - fast motion video sequence

U2 - 10.1109/EDERC.2014.6924377

DO - 10.1109/EDERC.2014.6924377

M3 - Chapter (peer-reviewed)

SN - 9781479968411

SP - 147

EP - 151

BT - 2014 6th European Embedded Design in Education and Research Conference (EDERC)

A2 - Soraghan, John J.

A2 - Di Caterina, Gaetano

A2 - Marinkovic, Djordje

A2 - Llin, Nuria

A2 - Wicks, Daniel

PB - IEEE

T2 - 6th European Embedded Design in Education and Research Conference (EDERC 2014)

Y2 - 11 September 2014 through 12 September 2014

ER -

An extended real-time compressive tracking method using weighted multi-frame cosine similarity metric

Abstract

Conference

Keywords

Documents and Links

Fingerprint

Cite this