Advertisement

Classification of Icon Type and Cooldown State in Video Game Replays

  • Jeremias Eichelbaum
  • Ronny Hänsch
  • Olaf Hellwich
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10882)

Abstract

The potential to positively influence research developments in seemingly unrelated areas leads to an increasing interest in the analysis of video games. As game publishers rarely provide an open interface to gain access to in-game information, the proposed system relies on the availability of video game recordings and broadcasts and operates completely in the visual domain. The classification of video game icons and associated metadata serves as an example task to assess the potential of several image recognition methods, including Random Forests (RFs), Support Vector Machines (SVMs), and Convolutional Networks (ConvNets). The experiments show that all machine learning approaches are able to successfully classify game icons in their original state, but performance is significantly decreased for icons in a cooldown state. SVMs fail to estimate the correct cooldown state, while RFs are outperformed by ConvNets.

Keywords

eSports Image classification Convolutional networks 

References

  1. 1.
    Wagner, M.G.: On the scientific relevance of eSports. In: International Conference on Internet Computing, pp. 437–442 (2006)Google Scholar
  2. 2.
    Esports Market Report: Courtside-Playmakers of 2017. https://www.superdataresearch.com/market-data/esports-market-report/
  3. 3.
    Blattner, M., Sumikawa, D., Greenberg, R.: Earcons and icons: their structure and common design principles. Hum. Comput. Interact. 4, 11–44 (1989)CrossRefGoogle Scholar
  4. 4.
    Krizhevsky, A.: Learning multiple layers of features from tiny images. Technical Report, Department of Computer Science, University of Toronto, (2009)Google Scholar
  5. 5.
    Song, Y.: Real-Time Video Highlights for Yahoo Esports (2016). arXiv preprint: arXiv:1611.08780
  6. 6.
    LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86, 2278–2324 (1998)CrossRefGoogle Scholar
  7. 7.
    Hughes, M., Bartlett, R.: The use of performance indicators in performance analysis. J. Sports Sci. 20, 739–754 (2002)CrossRefGoogle Scholar
  8. 8.
    Spann, M., Skiera, B.: Sports forecasting: a comparison of the forecast accuracy of prediction markets, betting odds and tipsters. J. Forecast. 28, 55–72 (2009)MathSciNetCrossRefGoogle Scholar
  9. 9.
    Yang, Y., Qin, T., Lei, Y.H.: Real-time eSports Match Result Prediction (2016). arXiv preprint: arXiv:1701.03162
  10. 10.
    Cireşan, D., Meier, U., Masci, J., Schmidhuber, J.: Multi-column deep neural network for traffic sign classification. Neural Netw. 32, 333–338 (2012)CrossRefGoogle Scholar
  11. 11.
    Semenov, A., Romov, P., Korolev, S., Yashkov, D., Neklyudov, K.: Performance of machine learning algorithms in predicting game outcome from drafts in Dota 2. In: International Conference on Analysis of Images, Social Networks and Texts, pp. 26–37 (2016)Google Scholar
  12. 12.
    Srivastava, N., Hinton, G.E., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014)MathSciNetzbMATHGoogle Scholar
  13. 13.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. 60(2), 91–110 (2004)CrossRefGoogle Scholar
  14. 14.
    Bosch, A., Zisserman, A., Munoz, X.: Image classification using random forests and ferns. In: IEEE 11th International Conference on Computer Vision, pp. 1–8 (2007)Google Scholar
  15. 15.
    Ebrahimzadeh, R., Jampour, M.: Efficient handwritten digit recognition based on histogram of oriented gradients and SVM. Int. J. Comput. Appl. 104, 10–13 (2014)Google Scholar
  16. 16.
    Kleinberg, E.M.: An overtraining-resistant stochastic modeling method for pattern recognition. Ann. Stat. 24(6), 2319–2349 (1996)MathSciNetCrossRefGoogle Scholar
  17. 17.
    Zhang, B., Srihari, S.: Fast k-nearest neighbor classification using cluster-based trees. IEEE Trans. Pattern Anal. Mach. Intell. 26, 525–528 (2004)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Jeremias Eichelbaum
    • 1
  • Ronny Hänsch
    • 1
  • Olaf Hellwich
    • 1
  1. 1.Computer Vision and Remote SensingTechnische Universität BerlinBerlinGermany

Personalised recommendations