Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics

Adrià Arbués Sangüesa, Thomas B. Moeslund, Chris Holmberg Bahnsen, Raul Benítez Iglesias

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

2 Citations (Scopus)
277 Downloads (Pure)

Abstract

Advanced statistics have proved to be a crucial tool for basketball coaches in order to improve training skills. Indeed, the performance of the team can be further optimized by studying the behaviour of players under certain conditions. In the United States of America, companies such as STATS or Second Spectrum use a complex multi-camera setup to deliver advanced statistics to all NBA teams, but the price of this service is far beyond the budget of the vast majority of European teams. For this reason, a first prototype based on positioning sensors is presented. An experimental dataset has been created and meaningful basketball features have been extracted. 97.9% accuracy is obtained using Support Vector Machines when identifying 5 different classic plays: floppy offense, pick and roll, press break, post-up situation and fast breaks. After recognizing these plays in video sequences, advanced statistics could be extracted with ease.
Original languageEnglish
Title of host publication2017 IEEE International Conference on Data Mining Workshops (ICDMW)
PublisherIEEE
Publication date17 Nov 2017
ISBN (Print)978-1-5386-3801-9
DOIs
Publication statusPublished - 17 Nov 2017
EventIEEE International Conference on Data Mining : Workshop on Data mining for the Analysis of Performance and Success - New Orleans, United States
Duration: 18 Nov 201721 Nov 2017

Workshop

WorkshopIEEE International Conference on Data Mining
CountryUnited States
CityNew Orleans
Period18/11/201721/11/2017
SeriesIEEE International Conference on Data Mining Workshops (ICDMW)
ISSN2375-9259

Fingerprint

Statistics
Sensors
Costs
Support vector machines
Cameras
Industry

Keywords

  • Advanced Statistics
  • Play Classification
  • Player Tracking
  • Basketball
  • Acceleration Wearable Sensora

Cite this

Sangüesa, A. A., Moeslund, T. B., Bahnsen, C. H., & Iglesias, R. B. (2017). Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics. In 2017 IEEE International Conference on Data Mining Workshops (ICDMW) IEEE. IEEE International Conference on Data Mining Workshops (ICDMW) https://doi.org/10.1109/ICDMW.2017.123
Sangüesa, Adrià Arbués ; Moeslund, Thomas B. ; Bahnsen, Chris Holmberg ; Iglesias, Raul Benítez. / Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics. 2017 IEEE International Conference on Data Mining Workshops (ICDMW). IEEE, 2017. (IEEE International Conference on Data Mining Workshops (ICDMW)).
@inproceedings{2bec33b572094364824d18beefe4d931,
title = "Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics",
abstract = "Advanced statistics have proved to be a crucial tool for basketball coaches in order to improve training skills. Indeed, the performance of the team can be further optimized by studying the behaviour of players under certain conditions. In the United States of America, companies such as STATS or Second Spectrum use a complex multi-camera setup to deliver advanced statistics to all NBA teams, but the price of this service is far beyond the budget of the vast majority of European teams. For this reason, a first prototype based on positioning sensors is presented. An experimental dataset has been created and meaningful basketball features have been extracted. 97.9{\%} accuracy is obtained using Support Vector Machines when identifying 5 different classic plays: floppy offense, pick and roll, press break, post-up situation and fast breaks. After recognizing these plays in video sequences, advanced statistics could be extracted with ease.",
keywords = "Advanced Statistics, Play Classification, Player Tracking, Basketball, Acceleration Wearable Sensora",
author = "Sang{\"u}esa, {Adri{\`a} Arbu{\'e}s} and Moeslund, {Thomas B.} and Bahnsen, {Chris Holmberg} and Iglesias, {Raul Ben{\'i}tez}",
year = "2017",
month = "11",
day = "17",
doi = "10.1109/ICDMW.2017.123",
language = "English",
isbn = "978-1-5386-3801-9",
series = "IEEE International Conference on Data Mining Workshops (ICDMW)",
publisher = "IEEE",
booktitle = "2017 IEEE International Conference on Data Mining Workshops (ICDMW)",
address = "United States",

}

Sangüesa, AA, Moeslund, TB, Bahnsen, CH & Iglesias, RB 2017, Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics. in 2017 IEEE International Conference on Data Mining Workshops (ICDMW). IEEE, IEEE International Conference on Data Mining Workshops (ICDMW), New Orleans, United States, 18/11/2017. https://doi.org/10.1109/ICDMW.2017.123

Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics. / Sangüesa, Adrià Arbués; Moeslund, Thomas B.; Bahnsen, Chris Holmberg; Iglesias, Raul Benítez.

2017 IEEE International Conference on Data Mining Workshops (ICDMW). IEEE, 2017. (IEEE International Conference on Data Mining Workshops (ICDMW)).

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics

AU - Sangüesa, Adrià Arbués

AU - Moeslund, Thomas B.

AU - Bahnsen, Chris Holmberg

AU - Iglesias, Raul Benítez

PY - 2017/11/17

Y1 - 2017/11/17

N2 - Advanced statistics have proved to be a crucial tool for basketball coaches in order to improve training skills. Indeed, the performance of the team can be further optimized by studying the behaviour of players under certain conditions. In the United States of America, companies such as STATS or Second Spectrum use a complex multi-camera setup to deliver advanced statistics to all NBA teams, but the price of this service is far beyond the budget of the vast majority of European teams. For this reason, a first prototype based on positioning sensors is presented. An experimental dataset has been created and meaningful basketball features have been extracted. 97.9% accuracy is obtained using Support Vector Machines when identifying 5 different classic plays: floppy offense, pick and roll, press break, post-up situation and fast breaks. After recognizing these plays in video sequences, advanced statistics could be extracted with ease.

AB - Advanced statistics have proved to be a crucial tool for basketball coaches in order to improve training skills. Indeed, the performance of the team can be further optimized by studying the behaviour of players under certain conditions. In the United States of America, companies such as STATS or Second Spectrum use a complex multi-camera setup to deliver advanced statistics to all NBA teams, but the price of this service is far beyond the budget of the vast majority of European teams. For this reason, a first prototype based on positioning sensors is presented. An experimental dataset has been created and meaningful basketball features have been extracted. 97.9% accuracy is obtained using Support Vector Machines when identifying 5 different classic plays: floppy offense, pick and roll, press break, post-up situation and fast breaks. After recognizing these plays in video sequences, advanced statistics could be extracted with ease.

KW - Advanced Statistics

KW - Play Classification

KW - Player Tracking

KW - Basketball

KW - Acceleration Wearable Sensora

U2 - 10.1109/ICDMW.2017.123

DO - 10.1109/ICDMW.2017.123

M3 - Article in proceeding

SN - 978-1-5386-3801-9

T3 - IEEE International Conference on Data Mining Workshops (ICDMW)

BT - 2017 IEEE International Conference on Data Mining Workshops (ICDMW)

PB - IEEE

ER -

Sangüesa AA, Moeslund TB, Bahnsen CH, Iglesias RB. Identifying Basketball Plays from Sensor Data; towards a Low-Cost Automatic Extraction of Advanced Statistics. In 2017 IEEE International Conference on Data Mining Workshops (ICDMW). IEEE. 2017. (IEEE International Conference on Data Mining Workshops (ICDMW)). https://doi.org/10.1109/ICDMW.2017.123