Understanding Patch-Based Learning of Video Data by Explaining Predictions

  • Christopher J. Anders
  • Grégoire MontavonEmail author
  • Wojciech Samek
  • Klaus-Robert Müller
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11700)


Deep neural networks have shown to learn highly predictive models of video data. Due to the large number of images in individual videos, a common strategy for training is to repeatedly extract short clips with random offsets from the video. We apply the deep Taylor/Layer-wise Relevance Propagation (LRP) technique to understand classification decisions of a deep network trained with this strategy, and identify a tendency of the classifier to look mainly at the frames close to the temporal boundaries of its input clip. This “border effect” reveals the model’s relation to the step size used to extract consecutive video frames for its input, which we can then tune in order to improve the classifier’s accuracy without retraining the model. To our knowledge, this is the first work to apply the deep Taylor/LRP technique on any neural network operating on video data.


Deep neural networks Video classification Human action recognition Explaining predictions 



This work was supported by the German Ministry for Education and Research as Berlin Big Data Centre (01IS14013A), Berlin Center for Machine Learning (01IS18037I) and TraMeExCo (01IS18056A). Partial funding by DFG is acknowledged (EXC 2046/1, project-ID: 390685689). This work was also supported by the Institute for Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (No. 2017-0-00451, No. 2017-0-01779).


  1. 1.
    Bach, S., Binder, A., Montavon, G., Klauschen, F., Müller, K.R., Samek, W.: On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLoS ONE 10(7), e0130140 (2015)CrossRefGoogle Scholar
  2. 2.
    Baldi, P., Sadowski, P., Whiteson, D.: Searching for exotic particles in high-energy physics with deep learning. Nat. Commun. 5, 4308 (2014)CrossRefGoogle Scholar
  3. 3.
    Bosse, S., Maniry, D., Müller, K.R., Wiegand, T., Samek, W.: Deep neural networks for no-reference and full-reference image quality assessment. IEEE Trans. Image Process. 27(1), 206–219 (2018)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Donahue, J., et al.: Long-term recurrent convolutional networks for visual recognition and description. In: IEEE CVPR, pp. 2625–2634 (2015)Google Scholar
  5. 5.
    Gevrey, M., Dimopoulos, I., Lek, S.: Review and comparison of methods to study the contribution of variables in artificial neural network models. Ecol. Model. 160(3), 249–264 (2003)CrossRefGoogle Scholar
  6. 6.
    Graves, A., Mohamed, A., Hinton, G.: Speech recognition with deep recurrent neural networks. In: IEEE ICASSP, pp. 6645–6649 (2013)Google Scholar
  7. 7.
    Hou, L., Samaras, D., Kurc, T.M., Gao, Y., Davis, J.E., Saltz, J.H.: Patch-based convolutional neural network for whole slide tissue image classification. In: IEEE CVPR, pp. 2424–2433 (2016)Google Scholar
  8. 8.
    Hu, K.T., Leou, J.J., Hsiao, H.H.: Spatiotemporal saliency detection and salient region determination for H.264 videos. JVCIR 24(7), 760–772 (2013)Google Scholar
  9. 9.
    Ji, S., Xu, W., Yang, M., Yu, K.: 3D convolutional neural networks for human action recognition. IEEE TPAMI 35(1), 221–231 (2013)CrossRefGoogle Scholar
  10. 10.
    Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., Fei-Fei, L.: Large-scale video classification with convolutional neural networks. In: IEEE CVPR, pp. 1725–1732 (2014)Google Scholar
  11. 11.
    Kauffmann, J., Esders, M., Montavon, G., Samek, W., Müller, K.R.,: From Clustering to Cluster Explanations via Neural Networks. arXiv preprint arXiv:1906.07633 (2019)
  12. 12.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Advances in NIPS, pp. 1097–1105 (2012)Google Scholar
  13. 13.
    Landecker, W., Thomure, M.D., Bettencourt, L.M., Mitchell, M., Kenyon, G.T., Brumby, S.P.: Interpreting individual classifications of hierarchical networks. In: IEEE Symposium CIDM, pp. 32–38 (2013)Google Scholar
  14. 14.
    Lapuschkin, S., Binder, A., Montavon, G., Müller, K.R., Samek, W.: The LRP toolbox for artificial neural networks. J. Mach. Learn. Res. 17(114), 1–5 (2016)MathSciNetzbMATHGoogle Scholar
  15. 15.
    Lapuschkin, S., Wäldchen, S., Binder, A., Montavon, G., Samek, W., Müller, K.R.: Unmasking Clever Hans predictors and assessing what machines really learn. Nat. Commun. 10, 1096 (2019)CrossRefGoogle Scholar
  16. 16.
    Li, J., Liu, Z., Zhang, X., Le Meur, O., Shen, L.: Spatiotemporal saliency detection based on superpixel-level trajectory. Sig. Process. Image Commun. 38, 100–114 (2015)CrossRefGoogle Scholar
  17. 17.
    Montavon, G., Lapuschkin, S., Binder, A., Samek, W., Müller, K.R.: Explaining nonlinear classification decisions with deep Taylor decomposition. Pattern Recogn. 65, 211–222 (2017)CrossRefGoogle Scholar
  18. 18.
    Montavon, G., Samek, W., Müller, K.R.: Methods for interpreting and understanding deep neural networks. Digit. Signal Proc. 73, 1–15 (2018)MathSciNetCrossRefGoogle Scholar
  19. 19.
    van den Oord, A., et al.: WaveNet: a generative model for raw audio. In: The 9th ISCA Speech Synthesis Workshop, Sunnyvale, CA, USA, 13–15 September 2016, p. 125 (2016)Google Scholar
  20. 20.
    Pohlen, T., Hermans, A., Mathias, M., Leibe, B.: Full-resolution residual networks for semantic segmentation in street scenes. In: IEEE CVPR, pp. 4151–4160 (2017)Google Scholar
  21. 21.
    Poulin, B., et al.: Visual explanation of evidence with additive classifiers. In: Proceedings, The Twenty-First National Conference on Artificial Intelligence and the Eighteenth Innovative Applications of Artificial Intelligence Conference, 16–20 July 2006, Boston, Massachusetts, USA, pp. 1822–1829 (2006)Google Scholar
  22. 22.
    Reed, S.E., Lee, H., Anguelov, D., Szegedy, C., Erhan, D., Rabinovich, A.: Training deep neural networks on noisy labels with bootstrapping. In: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, 7–9 May 2015, Workshop Track Proceedings (2015)Google Scholar
  23. 23.
    Ribeiro, M.T., Singh, S., Guestrin, C.: “Why should I trust you?”: explaining the predictions of any classifier. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016, pp. 1135–1144 (2016)Google Scholar
  24. 24.
    Samek, W., Binder, A., Montavon, G., Lapuschkin, S., Müller, K.: Evaluating the visualization of what a deep neural network has learned. IEEE Trans. Neural Netw. Learn. Syst. 28(11), 2660–2673 (2017)MathSciNetCrossRefGoogle Scholar
  25. 25.
    Schütt, K., Kindermans, P.J., Felix, H.E.S., Chmiela, S., Tkatchenko, A., Müller, K.R.: SchNet: a continuous-filter convolutional neural network for modeling quantum interactions. In: Advances in NIPS, pp. 992–1002 (2017)Google Scholar
  26. 26.
    Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., Batra, D.: Grad-CAM: visual explanations from deep networks via gradient-based localization. In: IEEE CVPR, pp. 618–626 (2017)Google Scholar
  27. 27.
    Sharma, S., Kiros, R., Salakhutdinov, R.: Action recognition using visual attention. CoRR abs/1511.04119 (2015)Google Scholar
  28. 28.
    Simonyan, K., Vedaldi, A., Zisserman, A.: Deep inside convolutional networks: visualising image classification models and saliency maps. In: 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, 14–16 April 2014, Workshop Track Proceedings (2014)Google Scholar
  29. 29.
    Springenberg, J.T., Dosovitskiy, A., Brox, T., Riedmiller, M.A.: Striving for simplicity: The all convolutional net. In: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, 7–9 May 2015, Workshop Track Proceedings (2015)Google Scholar
  30. 30.
    Srinivasan, V., Lapuschkin, S., Hellge, C., Müller, K.R., Samek, W.: Interpretable human action recognition in compressed domain. In: IEEE ICASSP, pp. 1692–1696 (2017)Google Scholar
  31. 31.
    Sturm, I., Lapuschkin, S., Samek, W., Müller, K.R.: Interpretable deep neural networks for single-trial EEG classification. J. Neurosci. Methods 274, 141–145 (2016)CrossRefGoogle Scholar
  32. 32.
    Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., et al.: Going deeper with convolutions. In: IEEE CVPR, pp. 1–9 (2015)Google Scholar
  33. 33.
    Tran, D., Bourdev, L., Fergus, R., Torresani, L., Paluri, M.: Learning spatiotemporal features with 3D convolutional networks. In: IEEE ICCV, pp. 4489–4497 (2015)Google Scholar
  34. 34.
    Yang, C., Rangarajan, A., Ranka, S.: Visual explanations from deep 3D convolutional neural networks for Alzheimer’s disease classification. CoRR abs/1803.02544 (2018)Google Scholar
  35. 35.
    Yosinski, J., Clune, J., Nguyen, A.M., Fuchs, T.J., Lipson, H.: Understanding neural networks through deep visualization. CoRR abs/1506.06579 (2015)Google Scholar
  36. 36.
    Yue-Hei Ng, J., Hausknecht, M., Vijayanarasimhan, S., Vinyals, O., Monga, R., Toderici, G.: Beyond short snippets: deep networks for video classification. In: IEEE CVPR, pp. 4694–4702 (2015)Google Scholar
  37. 37.
    Zeiler, M.D., Fergus, R.: Visualizing and understanding convolutional networks. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8689, pp. 818–833. Springer, Cham (2014). Scholar
  38. 38.
    Zhang, C., Tian, Y.: Automatic video description generation via LSTM with joint two-stream encoding. In: ICPR, pp. 2924–2929 (2016)Google Scholar
  39. 39.
    Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., Torralba, A.: Learning deep features for discriminative localization. In: IEEE CVPR, pp. 2921–2929 (2016)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Christopher J. Anders
    • 1
  • Grégoire Montavon
    • 1
    Email author
  • Wojciech Samek
    • 2
  • Klaus-Robert Müller
    • 1
    • 3
    • 4
  1. 1.Technische Universität BerlinBerlinGermany
  2. 2.Fraunhofer Heinrich Hertz InstituteBerlinGermany
  3. 3.Korea UniversitySeongbuk-gu, SeoulKorea
  4. 4.Max Planck Institute for InformaticsSaarbrückenGermany

Personalised recommendations