Advertisement

Pixel2Field: Single Image Transformation to Physical Field of Sports Videos

  • Liang PengEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11844)

Abstract

Locating players on a 2D field plane for sports match is the first step towards developing many types of sports analytics applications. Most existing mechanisms of locating players require them to wear sensors during sports play. Sports games can be easily recorded by cameras with low cost. Current human detection and tracking techniques can be used to locate players in the video, which is typically distorted for panorama view. We propose an end-to-end system named Pixel2Field, which can transform every pixel location into their scaled 2d field image. This is done by first undistorting the image by estimating the distortion coefficients, followed by a homography recovery. Experiments using detected soccer players from a distorted video show the proposed transformation method works well. To the best of knowledge, this is the first end-to-end system that can transform frame pixel location to field location without any human intervention. This unlock a lot of opportunities for developing sports analytics applications.

Keywords

Image distortion Homograph transformation Camera calibration Sports analytic 

References

  1. 1.
    Alonso, R.A.: Sports telemetry system for collecting performance metrics and data, US Patent 8,289,185, 16 October 2012Google Scholar
  2. 2.
    Baysal, S., Duygulu, P.: Sentioscope: a soccer player tracking system using model field particles. IEEE Trans. Circuits Syst. for Video Technol. 26(7), 1350–1362 (2015)CrossRefGoogle Scholar
  3. 3.
    Carlock, C.R.: Apparatus for locating and positioning a football on a football playing field, US Patent 3,985,356, 12 October 1976Google Scholar
  4. 4.
    Fryer, J.G., Brown, D.C.: Lens distortion for close-range photogrammetry. Photogram. Eng. Remote Sens. 52(1), 51–58 (1986)Google Scholar
  5. 5.
    Lopez, M., Mari, R., Gargallo, P., Kuang, Y., Gonzalez-Jimenez, J., Haro, G.: Deep single image camera calibration with radial distortion. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 11817–11825 (2019)Google Scholar
  6. 6.
    Pavić, D., Schönefeld, V., Kobbelt, L.: Interactive image completion with perspective correction. Vis. Comput. 22(9–11), 671–681 (2006)CrossRefGoogle Scholar
  7. 7.
    Peng, L.: Enhanced camera capturing using object-detection-based autofocus on smartphones. In: 2016 4th Intl Conf on Applied Computing and Information Technology/3rd Intl Conf on Computational Science/Intelligence and Applied Informatics/1st Intl Conf on Big Data, Cloud Computing, Data Science & Engineering (ACIT-CSII-BCD), pp. 208–212. IEEE (2016)Google Scholar
  8. 8.
    Peng, L.: Gscheduler: Reducing mobile device energy consumption. In: 2016 4th Intl Conf on Applied Computing and Information Technology/3rd Intl Conf on Computational Science/Intelligence and Applied Informatics/1st Intl Conf on Big Data, Cloud Computing, Data Science & Engineering (ACIT-CSII-BCD), pp. 1–6. IEEE (2016)Google Scholar
  9. 9.
    Peng, L.: Object Recognition in Videos Utilizing Hierarchical and Temporal Objectness with Deep Neural Networks. Ph.D. dissertation, Utah State University (2017)Google Scholar
  10. 10.
    Peng, L., Qi, X.: A hierarchical model to learn object proposals and its applications. J. Intell. Fuzzy Syst. 31(5), 2543–2551 (2016)CrossRefGoogle Scholar
  11. 11.
    Peng, L., Qi, X.: Temporal objectness: model-free learning of object proposals in video. In: 2016 IEEE International Conference on Image Processing (ICIP), pp. 3663–3667. IEEE (2016)Google Scholar
  12. 12.
    Peng, L., Qian, H.: Automatic red-eye detection in consumer photographs. In: International Conference on Computer Applications in Industry and Engineering (2018)Google Scholar
  13. 13.
    Peng, L., Qian, H.: A three-stage object detection framework using compact features and deep CNN. In: International Conference on Computer Applications in Industry and Engineering (2018)Google Scholar
  14. 14.
    Peng, L., Wang, H.: Object identification system and method, US Patent 9,122,931, 1 September 2015Google Scholar
  15. 15.
    Peng, L., Yang, Y., Qi, X., Wang, H.: Highly accurate video object identification utilizing hint information. In: 2014 International Conference on Computing, Networking and Communications (ICNC), pp. 317–321. IEEE (2014)Google Scholar
  16. 16.
    Peng, L., Yang, Y., Wang, H.: Automatic face annotation method and system, US Patent 9,176,987, 3 November 2015Google Scholar
  17. 17.
    Pettersen, S.A., et al.: Soccer video and player position dataset. In: Proceedings of the 5th ACM Multimedia Systems Conference, pp. 18–23. ACM (2014)Google Scholar
  18. 18.
    Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: Unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 779–788 (2016)Google Scholar
  19. 19.
    Wang, Q., Zhang, Y.: High speed stereoscopic shadowgraph imaging and its digital 3D reconstruction. Meas. Sci. Technol. 22(6), 065302 (2011)CrossRefGoogle Scholar
  20. 20.
    Yu, J., Skaff, S., Peng, L., Imai, F.: Leveraging knowledge-based inference for material classification. In: Proceedings of the 23rd ACM International Conference on Multimedia, pp. 1243–1246. ACM (2015)Google Scholar
  21. 21.
    Open image in new window , cN Patent CN105100894A, 25 November 2015Google Scholar
  22. 22.
    Open image in new window , cN Patent CN104573706A, 29 April 2015Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Verizon MediaSunnyvaleUSA

Personalised recommendations