Gesture-based communication is commonly used by soccer players during matches to exchange information with teammates. Among the possible forms of gesture-based interaction, hand signals are the most used. In this paper, we present a deep learning method for recognizing robot-to-robot hand signals exchanged during a soccer game. A neural network for estimating human body, face, hands, and foot position has been adapted for the application in the robot soccer scenario. Quantitative experiments carried out on NAO V6 robots demonstrate the effectiveness of the proposed approach. Source code and data used in this work are made publicly available for the community.
On Field Gesture-Based Robot-to-Robot Communication with NAO Soccer Players / Di Giambattista, V.; Fawakherji, M.; Suriani, V.; Bloisi, D. D.; Nardi, D.. - 11531:(2019), pp. 367-375. (Intervento presentato al convegno 23rd Annual RoboCup International Symposium, RoboCup 2019 tenutosi a Sydney; Australia) [10.1007/978-3-030-35699-6_28].
On Field Gesture-Based Robot-to-Robot Communication with NAO Soccer Players
Di Giambattista V.
;Fawakherji M.
;Suriani V.
;Bloisi D. D.
;Nardi D.
2019
Abstract
Gesture-based communication is commonly used by soccer players during matches to exchange information with teammates. Among the possible forms of gesture-based interaction, hand signals are the most used. In this paper, we present a deep learning method for recognizing robot-to-robot hand signals exchanged during a soccer game. A neural network for estimating human body, face, hands, and foot position has been adapted for the application in the robot soccer scenario. Quantitative experiments carried out on NAO V6 robots demonstrate the effectiveness of the proposed approach. Source code and data used in this work are made publicly available for the community.File | Dimensione | Formato | |
---|---|---|---|
DiGiambattista_On-Field-Gesture-based_2019.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
2.97 MB
Formato
Adobe PDF
|
2.97 MB | Adobe PDF | Contatta l'autore |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.