Abstract
3D players tracking plays an important role in sports analysis. Tracking of players contributes to high level game analysis such as tactic analysis and commercial applications such as TV contents. Many services like sports live and broadcasting have strict limitation on processing time, thus real-time implementation for 3D players tracking is necessary. This paper proposes a particle filter based 60 fps multi-view volleyball players tracking system on GPU platform. There are three proposals: body region constraint prediction, spatial pixels selection and inter-frame combined likelihood. The body region constraint prediction uses player’s body region as limitation in prediction to increase tracking accuracy. The spatial pixels selection method selects pixels for likelihood calculating to reduce calculation amount in spatial space. The inter-frame observation method does particle filter algorithm with two frames each time to reduce calculation amount in temporal space. Our experiments are based on videos of the Final and Semi-Final Game of 2014 Japan Inter High School Games of Men’s Volleyball in Tokyo Metropolitan Gymnasium. On the GPU device GeForce GTX 1080Ti, our tracking system achieves real-time on 60 fps videos and keeps the tracking accuracy higher than 97%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Buddubariki, V., Tulluri, G.S., Mukherjee, S.: Multiple object tracking by improved KLT tracker over SURF features. In: 5th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG), pp. 1–4 (2015)
Tseng, T., Liu, A., Hsiao, P., Huang, C., Fu, L.: Real-time people detection and tracking for indoor surveillance using multiple top-view depth cameras. In: IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 4077–4082 (2014)
Huang, S., Zhuang, X., Cheng, X., Ikoma, N., Honda, M., Ikenaga, T.: Player feature based multi-likelihood and spatial relationship based multi-view elimination with least square fitting prediction for volleyball players tracking in 3D space. IIEEJ Trans. Image Electr. Vis. Comput. 4(2), 145–155 (2016)
Deng, Z., Hou, Y., Cheng, X., Ikenaga, T.: Vectorized data combination and binary search oriented reweight for CPU-GPU based real-time 3D ball tracking. In: Zeng, B., Huang, Q., El Saddik, A., Li, H., Jiang, S., Fan, X. (eds.) PCM 2017. LNCS, vol. 10736, pp. 508–516. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-77383-4_50
Nishikawa, Y., Sato, H., Ozawa, J.: Multiple sports player tracking system based on graph optimization using low-cost cameras. In: IEEE International Conference on Consumer Electronics (ICCE), pp. 1–4 (2018)
Petsas, P., Kaimakis, P.: Soccer player tracking using particle filters. In: IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), pp. 57–62 (2016)
Hou, Y., Cheng, X., Ikenaga, T.: Real-time 3D ball tracking with CPU-GPU acceleration using particle filter with multi-command queues and stepped parallelism iteration. In: International Conference on Multimedia and Image Processing (ICMIP), pp. 235–239 (2017)
Acknowledgements
This work was supported by KAKENHI (16K 13006) and Waseda University Grant for Special Research Projects (2018K-302).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhao, Y., Cheng, X., Ikenaga, T. (2018). Spatial Pixels Selection and Inter-frame Combined Likelihood Based Observation for 60 fps 3D Tracking of Twelve Volleyball Players on GPU. In: Hong, R., Cheng, WH., Yamasaki, T., Wang, M., Ngo, CW. (eds) Advances in Multimedia Information Processing – PCM 2018. PCM 2018. Lecture Notes in Computer Science(), vol 11165. Springer, Cham. https://doi.org/10.1007/978-3-030-00767-6_66
Download citation
DOI: https://doi.org/10.1007/978-3-030-00767-6_66
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-00766-9
Online ISBN: 978-3-030-00767-6
eBook Packages: Computer ScienceComputer Science (R0)