TY - GEN
T1 - Prediction of future shot direction using pose and position of tennis player
AU - Shimizu, Tomohiro
AU - Hachiuma, Ryo
AU - Saito, Hideo
AU - Yoshikawa, Takashi
AU - Lee, Chonho
N1 - Publisher Copyright:
© 2019 Copyright held by the owner/author(s).
PY - 2019/10/15
Y1 - 2019/10/15
N2 - In this paper, we propose a method to predict the future shot direction in a tennis match using pose information and player position. As far as we know, there is no work that deals with such a predictive task, so there is no shot direction dataset as yet. Therefore, using a YouTube tennis match video, we construct an time of impact and shot direction dataset. To reduce annotation costs, we propose a method to automatically label the shot direction. Moreover, we propose a method to predict the future shot direction using the constructed dataset. The shot direction is predicted using LSTM(long short-time memory), from sequential pose information up to the time of impact and the player position. We employ OpenPose to extract the position of skeleton joints. In the experiment, we evaluate the accuracy of shot direction prediction and verify the effectiveness of the proposed method. Since there are no studies that predict future shot direction, we set four baseline methods to evaluate the effectiveness of our proposed method.
AB - In this paper, we propose a method to predict the future shot direction in a tennis match using pose information and player position. As far as we know, there is no work that deals with such a predictive task, so there is no shot direction dataset as yet. Therefore, using a YouTube tennis match video, we construct an time of impact and shot direction dataset. To reduce annotation costs, we propose a method to automatically label the shot direction. Moreover, we propose a method to predict the future shot direction using the constructed dataset. The shot direction is predicted using LSTM(long short-time memory), from sequential pose information up to the time of impact and the player position. We employ OpenPose to extract the position of skeleton joints. In the experiment, we evaluate the accuracy of shot direction prediction and verify the effectiveness of the proposed method. Since there are no studies that predict future shot direction, we set four baseline methods to evaluate the effectiveness of our proposed method.
KW - Activity recognition in tennis
KW - Long short-term memory
KW - Shot direction prediction
UR - http://www.scopus.com/inward/record.url?scp=85075726640&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075726640&partnerID=8YFLogxK
U2 - 10.1145/3347318.3355523
DO - 10.1145/3347318.3355523
M3 - Conference contribution
AN - SCOPUS:85075726640
T3 - MMSports 2019 - Proceedings of the 2nd International Workshop on Multimedia Content Analysis in Sports, co-located with MM 2019
SP - 59
EP - 66
BT - MMSports 2019 - Proceedings of the 2nd International Workshop on Multimedia Content Analysis in Sports, co-located with MM 2019
PB - Association for Computing Machinery, Inc
T2 - 2nd ACM International Workshop on Multimedia Content Analysis in Sports, MMSports 2019, co-located with ACM Multimedia 2019
Y2 - 25 October 2019
ER -