Combining inertial and visual sensing for human action recognition in tennis

Ciarán Ó Conaire; Damien Connaghan; Philip Kelly; Noel E. O'Connor; Mark Gaffney; John Buckley

doi:10.1145/1877868.1877882

Combining inertial and visual sensing for human action recognition in tennis

Ciarán Ó Conaire
, Damien Connaghan
, Philip Kelly
, Noel E. O'Connor
, Mark Gaffney
, John Buckley

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

Abstract

In this paper, we present a framework for both the automatic extraction of the temporal location of tennis strokes within a match and the subsequent classification of these as being either a serve, forehand or backhand. We employ the use of low-cost visual sensing and low-cost inertial sensing to achieve these aims, whereby a single modality can be used or a fusion of both classification strategies can be adopted if both modalities are available within a given capture scenario. This flexibility allows the framework to be applicable to a variety of user scenarios and hardware infrastructures. Our proposed approach is quantitatively evaluated using data captured from elite tennis players. Results point to the extremely accurate performance of the proposed approach irrespective of input modality configuration.

Original language	English
Title of host publication	ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010
Pages	51-56
Number of pages	6
DOIs	https://doi.org/10.1145/1877868.1877882
Publication status	Published - 2010
Event	1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, ARTEMIS'10, Co-located with ACM Multimedia 2010 - Firenze, Italy Duration: 29 Oct 2010 → 29 Oct 2010

Publication series

Name	ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010

Conference

Conference	1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, ARTEMIS'10, Co-located with ACM Multimedia 2010
Country/Territory	Italy
City	Firenze
Period	29/10/10 → 29/10/10

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Accelerometers
Activity classification
Data fusion
Image processing
Inertial measurement units

Access to Document

10.1145/1877868.1877882

Cite this

Conaire, C. Ó., Connaghan, D., Kelly, P., O'Connor, N. E., Gaffney, M., & Buckley, J. (2010). Combining inertial and visual sensing for human action recognition in tennis. In ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010 (pp. 51-56). (ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010). https://doi.org/10.1145/1877868.1877882

Conaire, Ciarán Ó ; Connaghan, Damien ; Kelly, Philip et al. / Combining inertial and visual sensing for human action recognition in tennis. ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010. 2010. pp. 51-56 (ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010).

@inproceedings{1b2d55b1885f4b0ca9875e873dbb38b1,

title = "Combining inertial and visual sensing for human action recognition in tennis",

abstract = "In this paper, we present a framework for both the automatic extraction of the temporal location of tennis strokes within a match and the subsequent classification of these as being either a serve, forehand or backhand. We employ the use of low-cost visual sensing and low-cost inertial sensing to achieve these aims, whereby a single modality can be used or a fusion of both classification strategies can be adopted if both modalities are available within a given capture scenario. This flexibility allows the framework to be applicable to a variety of user scenarios and hardware infrastructures. Our proposed approach is quantitatively evaluated using data captured from elite tennis players. Results point to the extremely accurate performance of the proposed approach irrespective of input modality configuration.",

keywords = "Accelerometers, Activity classification, Data fusion, Image processing, Inertial measurement units",

author = "Conaire, \{Ciar{\'a}n {\'O}\} and Damien Connaghan and Philip Kelly and O'Connor, \{Noel E.\} and Mark Gaffney and John Buckley",

year = "2010",

doi = "10.1145/1877868.1877882",

language = "English",

isbn = "9781450301633",

series = "ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010",

pages = "51--56",

booktitle = "ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010",

note = "1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, ARTEMIS'10, Co-located with ACM Multimedia 2010 ; Conference date: 29-10-2010 Through 29-10-2010",

}

Conaire, CÓ, Connaghan, D, Kelly, P, O'Connor, NE, Gaffney, M & Buckley, J 2010, Combining inertial and visual sensing for human action recognition in tennis. in ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010. ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010, pp. 51-56, 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, ARTEMIS'10, Co-located with ACM Multimedia 2010, Firenze, Italy, 29/10/10. https://doi.org/10.1145/1877868.1877882

Combining inertial and visual sensing for human action recognition in tennis. / Conaire, Ciarán Ó; Connaghan, Damien; Kelly, Philip et al.
ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010. 2010. p. 51-56 (ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010).

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

TY - GEN

T1 - Combining inertial and visual sensing for human action recognition in tennis

AU - Conaire, Ciarán Ó

AU - Connaghan, Damien

AU - Kelly, Philip

AU - O'Connor, Noel E.

AU - Gaffney, Mark

AU - Buckley, John

PY - 2010

Y1 - 2010

N2 - In this paper, we present a framework for both the automatic extraction of the temporal location of tennis strokes within a match and the subsequent classification of these as being either a serve, forehand or backhand. We employ the use of low-cost visual sensing and low-cost inertial sensing to achieve these aims, whereby a single modality can be used or a fusion of both classification strategies can be adopted if both modalities are available within a given capture scenario. This flexibility allows the framework to be applicable to a variety of user scenarios and hardware infrastructures. Our proposed approach is quantitatively evaluated using data captured from elite tennis players. Results point to the extremely accurate performance of the proposed approach irrespective of input modality configuration.

AB - In this paper, we present a framework for both the automatic extraction of the temporal location of tennis strokes within a match and the subsequent classification of these as being either a serve, forehand or backhand. We employ the use of low-cost visual sensing and low-cost inertial sensing to achieve these aims, whereby a single modality can be used or a fusion of both classification strategies can be adopted if both modalities are available within a given capture scenario. This flexibility allows the framework to be applicable to a variety of user scenarios and hardware infrastructures. Our proposed approach is quantitatively evaluated using data captured from elite tennis players. Results point to the extremely accurate performance of the proposed approach irrespective of input modality configuration.

KW - Accelerometers

KW - Activity classification

KW - Data fusion

KW - Image processing

KW - Inertial measurement units

UR - https://www.scopus.com/pages/publications/78650440841

U2 - 10.1145/1877868.1877882

DO - 10.1145/1877868.1877882

M3 - Conference proceeding

AN - SCOPUS:78650440841

SN - 9781450301633

T3 - ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010

SP - 51

EP - 56

BT - ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010

T2 - 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, ARTEMIS'10, Co-located with ACM Multimedia 2010

Y2 - 29 October 2010 through 29 October 2010

ER -

Conaire CÓ, Connaghan D, Kelly P, O'Connor NE, Gaffney M, Buckley J. Combining inertial and visual sensing for human action recognition in tennis. In ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010. 2010. p. 51-56. (ARTEMIS'10 - Proceedings of the 1st ACM Workshop on Analysis and Retrieval of Tracked Events and Motion in Imagery Streams, Co-located with ACM Multimedia 2010). doi: 10.1145/1877868.1877882

Combining inertial and visual sensing for human action recognition in tennis

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this