Data analytics in sports is crucial to evaluate the performance of single players and the whole team. The literature proposes a number of tools for both offence and defence scenarios. Data coming from tracking location of players, in this respect, may be used to enrich the amount of useful information. In basket- ball, however, actions are interleaved with inactive periods. This paper describes a methodological approach to automatically identify active periods during a game and to classify them as offensive or defensive. The method is based on the applica- tion of thresholds to players kinematic parameters, whose values undergo a “tuning” strategy similar to Receiver Operating Characteristic curves, using a “ground truth” extracted from the video of the games.
La “data analytics” `e cruciale per valutare le prestazioni di singoli gio- catori e dei team nello sport. La letteratura accademica ha sviluppato una serie di strumenti per le situazioni di attacco e di difesa. I dati che rilevano la posizione dei giocatori possono essere utilizzati per arricchire la quantita’ di informazioni utili. Nel basket, tuttavia, le azioni sono intervallate da periodi inattivi. In questo lavoro si propone un metodo per identificare i periodi attivi e classificarli come offensivi o difensivi. Il metodo si basa sull’applicazione di soglie sui parametri cinematici dei giocatori, i cui valori vengono sottoposti ad una strategia di “tuning” simile alle curve ROC in cui la “ground truth” viene estratta da un’analisi video.
(2019). Detecting and classifying moments in basketball matches using sensor tracked data . Retrieved from http://hdl.handle.net/10446/228025
Detecting and classifying moments in basketball matches using sensor tracked data
Metulini, Rodolfo;Zuccolotto, Paola
2019-01-01
Abstract
Data analytics in sports is crucial to evaluate the performance of single players and the whole team. The literature proposes a number of tools for both offence and defence scenarios. Data coming from tracking location of players, in this respect, may be used to enrich the amount of useful information. In basket- ball, however, actions are interleaved with inactive periods. This paper describes a methodological approach to automatically identify active periods during a game and to classify them as offensive or defensive. The method is based on the applica- tion of thresholds to players kinematic parameters, whose values undergo a “tuning” strategy similar to Receiver Operating Characteristic curves, using a “ground truth” extracted from the video of the games.File | Dimensione del file | Formato | |
---|---|---|---|
26. SIS proc 2019 basket.pdf
Solo gestori di archivio
Versione:
publisher's version - versione editoriale
Licenza:
Licenza default Aisberg
Dimensione del file
744.28 kB
Formato
Adobe PDF
|
744.28 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo