Abstract
In sports applications, vision-based player detection is essential. For real-time activities like broadcasts and player identification, accuracy, efficiency, and minimal memory use are needed. The major difficulties in deploying object detection networks to embedded devices are the high computation and memory requirements. This paper proposes a mechanism of deep learning lightweight player detection pre-trained network (MobileNet) for Single-Shot Multibox Detector (SSD), which reduces the architecture weight file by reducing the number of convolutional layers and improves the computation speed. Therefore, MobileNetv1 is concatenated with the SSD framework of an efficient and accurate deep learning model for player detection in sports. The fundamental objective of this paper is to examine the accuracy and computation speed of the player detection approach (SSD), as well as the significance of a pre-trained deep learning model (MobileNet). The proposed model achieves 92.1% of precision, 81.3% of f1-score, and 12.4 MB of network weight file with an average frame rate of 57.2 frames per second (FPS) on the basketball dataset. The experimental results show that the MobileNetv1 + SSD designed for the purpose is more suitable for deployment in embedded devices for real-time player detection in sports.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hartley R, Zisserman A (2003) Multiple view geometry in computer vision. Cambridge University Press, Cambridge
Thomas G, Gade R, Moeslund TB, Carr P, Hilton A (2017) Computer vision for sports: current applications and research topics. Comput Vis Image Underst 159:3–18
Chen J, Le HM, Carr P, Yue Y, Little JJ (2016) Learning online smooth predictors for realtime camera planning using recurrent decision trees. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4688–4696
Chen J, Little JJ (2017) Where should cameras look at soccer games: improving smoothness using the overlapped hidden Markov model. Comput Vis Image Underst 159:59–73
Ibrahim MS, Muralidharan S, Deng Z, Vahdat A, Mori G (2016) A hierarchical deep temporal model for group activity recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1971–1980
Lu W-L, Ting J-A, Little JJ, Murphy KP (2013) Learning to track and identify players from broadcast sports videos. IEEE Trans Pattern Anal Mach Intell 35(7):1704–1716
Lucey P, Oliver D, Carr P, Roth J, Matthews I (2013) Assessing team strategy using spatiotemporal data. In: Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining, pp 1366–1374
Carr P, Sheikh Y, Matthews I (2012) Monocular object detection using 3d geometric primitives. In: European conference on computer vision. Springer, Berlin, Heidelberg, pp 864–878
Liu J, Tong X, Li W, Wang T, Zhang Y, Wang H (2009) Automatic player detection, labeling and tracking in broadcast soccer video. Pattern Recogn Lett 30(2):103–113
Parisot P, De Vleeschouwer C (2017) Scene-specific classifier for effective and efficient team sport players detection from a single calibrated camera. Comput Vis Image Understand 159:74–88
Ren S, He K, Girshick R, Sun J (2016) Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 39(6):1137–1149
Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 779–788
Howard AG, Zhu M, Chen B, Kalenichenko D, Wang W, Weyand T, Andreetto M, Adam H (2017) Mobilenets: efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861
Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C-Y, Berg AC (2016) Ssd: Single shot multibox detector. In: European conference on computer vision. Springer, Cham, pp 21–37
Zeiler MD, Fergus R (2014) Visualizing and understanding convolutional networks. In: European conference on computer vision. Springer, Cham, pp 818–833
Citraro L, Márquez-Neila P, Savare S, Jayaram V, Dubout C, Renaut F, Hasfura A, Shitrit HB, Fua P (2020) Real-time camera pose estimation for sports fields. Mach Vis Appl 31(3):1–13
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Naik, B.T., Hashmi, M.F. (2023). MobileNet + SSD: Lightweight Network for Real-Time Detection of Basketball Player. In: Yadav, R.P., Nanda, S.J., Rana, P.S., Lim, MH. (eds) Proceedings of the International Conference on Paradigms of Computing, Communication and Data Sciences. Algorithms for Intelligent Systems. Springer, Singapore. https://doi.org/10.1007/978-981-19-8742-7_2
Download citation
DOI: https://doi.org/10.1007/978-981-19-8742-7_2
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-8741-0
Online ISBN: 978-981-19-8742-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)