Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest

Kacper Skawinski; Ferran Montraveta Roca; Rainhard Dieter Findling; Stephan Sigg

doi:10.1007/978-3-030-20521-8_29

Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest

Kacper Skawinski, Ferran Montraveta Roca, Rainhard Dieter Findling^*, Stephan Sigg

^*Tämän työn vastaava kirjoittaja

Ambient Intelligence

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference contribution › Scientific › vertaisarvioitu

12 Sitaatiot (Scopus)

211 Lataukset (Pure)

Abstrakti

Sports and workout activities have become important parts of modern life. Nowadays, many people track characteristics about their sport activities with their mobile devices, which feature inertial measurement unit (IMU) sensors. In this paper we present a methodology to detect and recognize workout, as well as to count repetitions done in a recognized type of workout, from a single 3D accelerometer worn at the chest. We consider four different types of workout (pushups, situps, squats and jumping jacks). Our technical approach to workout type recognition and repetition counting is based on machine learning with a convolutional neural network. Our evaluation utilizes data of 10 subjects, which wear a Movesense sensors on their chest during their workout. We thereby find that workouts are recognized correctly on average 89.9% of the time, and the workout repetition counting yields an average detection accuracy of 97.9% over all types of workout.

Alkuperäiskieli	Englanti
Otsikko	Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings
Toimittajat	Ignacio Rojas, Gonzalo Joya, Andreu Catala
Sivut	347-359
Sivumäärä	13
DOI - pysyväislinkit	https://doi.org/10.1007/978-3-030-20521-8_29
Tila	Julkaistu - 2019
OKM-julkaisutyyppi	A4 Artikkeli konferenssijulkaisuussa
Tapahtuma	International Work Conference on Artificial Neural Networks - Gran Canaria, Espanja Kesto: 12 kesäk. 2019 → 14 kesäk. 2019 Konferenssinumero: 15

Julkaisusarja

Nimi	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Vuosikerta	11506 LNCS
ISSN (painettu)	0302-9743
ISSN (elektroninen)	1611-3349

Conference

Conference	International Work Conference on Artificial Neural Networks
Lyhennettä	IWANN
Maa/Alue	Espanja
Kaupunki	Gran Canaria
Ajanjakso	12/06/2019 → 14/06/2019

Pääsy asiakirjaan

10.1007/978-3-030-20521-8_29

ELEC_Skawinski_WorkoutTypeRecognition_LNCSAccepted author manuscript, 2,54 MB

SFX-saatavuus

Koko teksti

Muut tiedostot ja linkit

Link to publication in Scopus

Siteeraa tätä

Skawinski, K., Montraveta Roca, F., Findling, R. D., & Sigg, S. (2019). Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest. teoksessa I. Rojas, G. Joya, & A. Catala (Toimittajat), Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings (Sivut 347-359). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vuosikerta 11506 LNCS). https://doi.org/10.1007/978-3-030-20521-8_29

Skawinski, Kacper ; Montraveta Roca, Ferran ; Findling, Rainhard Dieter et al. / Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest. Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. Toimittaja / Ignacio Rojas ; Gonzalo Joya ; Andreu Catala. 2019. Sivut 347-359 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{4993d9d5a5fa412c802398d59572e966,

title = "Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest",

abstract = "Sports and workout activities have become important parts of modern life. Nowadays, many people track characteristics about their sport activities with their mobile devices, which feature inertial measurement unit (IMU) sensors. In this paper we present a methodology to detect and recognize workout, as well as to count repetitions done in a recognized type of workout, from a single 3D accelerometer worn at the chest. We consider four different types of workout (pushups, situps, squats and jumping jacks). Our technical approach to workout type recognition and repetition counting is based on machine learning with a convolutional neural network. Our evaluation utilizes data of 10 subjects, which wear a Movesense sensors on their chest during their workout. We thereby find that workouts are recognized correctly on average 89.9% of the time, and the workout repetition counting yields an average detection accuracy of 97.9% over all types of workout.",

keywords = "Acceleration, Activity recognition, CNN, Deep learning, Movesense, Neural Networks, Sensors, Workout",

author = "Kacper Skawinski and {Montraveta Roca}, Ferran and Findling, {Rainhard Dieter} and Stephan Sigg",

year = "2019",

doi = "10.1007/978-3-030-20521-8_29",

language = "English",

isbn = "978-3-030-20520-1",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "347--359",

editor = "Ignacio Rojas and Gonzalo Joya and Andreu Catala",

booktitle = "Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings",

note = "International Work Conference on Artificial Neural Networks, IWANN ; Conference date: 12-06-2019 Through 14-06-2019",

}

Skawinski, K, Montraveta Roca, F, Findling, RD & Sigg, S 2019, Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest. julkaisussa I Rojas, G Joya & A Catala (toim), Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vuosikerta 11506 LNCS, Sivut 347-359, International Work Conference on Artificial Neural Networks, Gran Canaria, Espanja, 12/06/2019. https://doi.org/10.1007/978-3-030-20521-8_29

Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest. / Skawinski, Kacper; Montraveta Roca, Ferran; Findling, Rainhard Dieter et al.

Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. toim. / Ignacio Rojas; Gonzalo Joya; Andreu Catala. 2019. s. 347-359 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vuosikerta 11506 LNCS).

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference contribution › Scientific › vertaisarvioitu

TY - GEN

T1 - Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest

AU - Skawinski, Kacper

AU - Montraveta Roca, Ferran

AU - Findling, Rainhard Dieter

AU - Sigg, Stephan

N1 - Conference code: 15

PY - 2019

Y1 - 2019

N2 - Sports and workout activities have become important parts of modern life. Nowadays, many people track characteristics about their sport activities with their mobile devices, which feature inertial measurement unit (IMU) sensors. In this paper we present a methodology to detect and recognize workout, as well as to count repetitions done in a recognized type of workout, from a single 3D accelerometer worn at the chest. We consider four different types of workout (pushups, situps, squats and jumping jacks). Our technical approach to workout type recognition and repetition counting is based on machine learning with a convolutional neural network. Our evaluation utilizes data of 10 subjects, which wear a Movesense sensors on their chest during their workout. We thereby find that workouts are recognized correctly on average 89.9% of the time, and the workout repetition counting yields an average detection accuracy of 97.9% over all types of workout.

AB - Sports and workout activities have become important parts of modern life. Nowadays, many people track characteristics about their sport activities with their mobile devices, which feature inertial measurement unit (IMU) sensors. In this paper we present a methodology to detect and recognize workout, as well as to count repetitions done in a recognized type of workout, from a single 3D accelerometer worn at the chest. We consider four different types of workout (pushups, situps, squats and jumping jacks). Our technical approach to workout type recognition and repetition counting is based on machine learning with a convolutional neural network. Our evaluation utilizes data of 10 subjects, which wear a Movesense sensors on their chest during their workout. We thereby find that workouts are recognized correctly on average 89.9% of the time, and the workout repetition counting yields an average detection accuracy of 97.9% over all types of workout.

KW - Acceleration

KW - Activity recognition

KW - CNN

KW - Deep learning

KW - Movesense

KW - Neural Networks

KW - Sensors

KW - Workout

UR - http://www.scopus.com/inward/record.url?scp=85067455886&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-20521-8_29

DO - 10.1007/978-3-030-20521-8_29

M3 - Conference contribution

AN - SCOPUS:85067455886

SN - 978-3-030-20520-1

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 347

EP - 359

BT - Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings

A2 - Rojas, Ignacio

A2 - Joya, Gonzalo

A2 - Catala, Andreu

T2 - International Work Conference on Artificial Neural Networks

Y2 - 12 June 2019 through 14 June 2019

ER -

Skawinski K, Montraveta Roca F, Findling RD, Sigg S. Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest. julkaisussa Rojas I, Joya G, Catala A, toimittajat, Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. 2019. s. 347-359. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-20521-8_29

Workout Type Recognition and Repetition Counting with CNNs from 3D Acceleration Sensed on the Chest

Abstrakti

Julkaisusarja

Conference

Pääsy asiakirjaan

SFX-saatavuus

Muut tiedostot ja linkit

Sormenjälki

Siteeraa tätä