Abstract
This paper describes the methodology our team followed for our submission to the SoccerNet Calibration Challenge - Soccer Pitch Markings and Goal Posts Localization. The goal of the challenge is to detect the extremities of soccer pitch lines present in the image. Our method directly infers the line extremities’ localization in the image, using a Deep Learning Convolutional Neural Network based on U-Net with a hierarchical output and unbalanced loss weights. The hierarchical output contains three outputs composed of different segmentation masks. Our team’s name is 2Ai-IPCA and our method achieved third place in the Pitch Localization task of the SoccerNet Calibration Challenge, with 71.01%, 76.18%, and 77.60% accuracies for the 5 pixel, 10 pixel, and 20 pixel thresholds respectively, and a global average accuracy score of 73.81% on the challenge set.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Soccernet camera calibration challenge development kit. https://github.com/SoccerNet/sn-calibration. Accessed 30 Jan 2023
Soccernet field localization challenge. https://www.soccer-net.org/tasks/field-localization. Accessed 30 Jan 2023
Cioppa, A., et al.: Camera calibration and player localization in SoccerNet-v2 and investigation of their representations for action spotting. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pp. 4537–4546 (2021)
Ekin, A., Tekalp, A., Mehrotra, R.: Automatic soccer video analysis and summarization. IEEE Trans. Image Process. 12(7), 796–807 (2003)
Fu, X., Zhang, K., Wang, C., Fan, C.: Multiple player tracking in basketball court videos. J. Real-Time Image Proc. 17(6), 1811–1828 (2020). https://doi.org/10.1007/s11554-020-00968-x
Gao, X., Niu, Z., Tao, D., Li, X.: Non-goal scene analysis for soccer video. Neurocomputing 74(4), 540–548 (2011)
Khatoonabadi, S.H., Rahmati, M.: Automatic soccer players tracking in goal scenes by camera motion elimination. Image Vis. Comput. 27(4), 469–479 (2009)
Liu, Y., Liang, D., Huang, Q., Gao, W.: Extracting 3D information from broadcast soccer video. Image Vis. Comput. 24(10), 1146–1162 (2006)
Niu, Z., Gao, X., Tian, Q.: Tactic analysis based on real-world ball trajectory in soccer video. Pattern Recogn. 45(5), 1937–1947 (2012)
Ren, J., Orwell, J., Jones, G.A., Xu, M.: Tracking the soccer ball using multiple fixed cameras. Comput. Vis. Image Underst. 113(5), 633–642 (2009)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Yu, X., Xu, C., Leong, H.W., Tian, Q., Tang, Q., Wan, K.W.: Trajectory-based ball detection and tracking with applications to semantic analysis of broadcast soccer video. In: Proceedings of the Eleventh ACM International Conference on Multimedia, pp. 11–20. Association for Computing Machinery (2003)
zhixuhao: Unet. https://github.com/zhixuhao/unet (2017). Accessed 30 Jan 2023
Åkeborg, E.: Automating feature-extraction for camera calibration through machine learning and computer vision (2021). Master thesis. https://lup.lub.lu.se/student-papers/search/publication/9054961. Accessed 31 Jan 2023
Acknowledgments
This work was partially funded by the project “POCI-01-0247-FEDER-046964”, supported by Operational Program for Competitiveness and Internationalization (COMPETE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF 1). This work was also partially funded by national funds (PIDDAC), through the FCT - Fundação para a Ciência e Tecnologia and FCT/MCTES under the scope of the projects UIDB/05549/2020 and UIDP/05549/2020. This paper was also partially funded by national funds, through the FCT - Fundação para a Ciência e a Tecnologia and FCT/MCTES under the scope of the project LASI-LA/P/0104/2020.
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 Switzerland AG
About this paper
Cite this paper
Santos Marques, M., Gomes Faria, R., Brito, J.H. (2023). Hierarchical Line Extremity Segmentation U-Net for the SoccerNet 2022 Calibration Challenge - Pitch Localization. In: Pertusa, A., Gallego, A.J., Sánchez, J.A., Domingues, I. (eds) Pattern Recognition and Image Analysis. IbPRIA 2023. Lecture Notes in Computer Science, vol 14062. Springer, Cham. https://doi.org/10.1007/978-3-031-36616-1_35
Download citation
DOI: https://doi.org/10.1007/978-3-031-36616-1_35
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-36615-4
Online ISBN: 978-3-031-36616-1
eBook Packages: Computer ScienceComputer Science (R0)
