Advertisement

Deep Learning vs. Traditional Computer Vision

  • Niall O’MahonyEmail author
  • Sean Campbell
  • Anderson Carvalho
  • Suman Harapanahalli
  • Gustavo Velasco Hernandez
  • Lenka Krpalkova
  • Daniel Riordan
  • Joseph Walsh
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 943)

Abstract

Deep Learning has pushed the limits of what was possible in the domain of Digital Image Processing. However, that is not to say that the traditional computer vision techniques which had been undergoing progressive development in years prior to the rise of DL have become obsolete. This paper will analyse the benefits and drawbacks of each approach. The aim of this paper is to promote a discussion on whether knowledge of classical computer vision techniques should be maintained. The paper will also explore how the two sides of computer vision can be combined. Several recent hybrid methodologies are reviewed which have demonstrated the ability to improve computer vision performance and to tackle problems not suited to Deep Learning. For example, combining traditional computer vision techniques with Deep Learning has been popular in emerging domains such as Panoramic Vision and 3D vision for which Deep Learning models have not yet been fully optimised.

Keywords

Computer vision Deep learning Hybrid techniques 

References

  1. 1.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Proceedings of 25th International Conference on Neural Information Processing System, NIPS 2012, vol. 1, pp. 1097–1105 (2012)Google Scholar
  2. 2.
    Nash, W., Drummond, T., Birbilis, N.: A review of deep learning in the study of materials degradation. npj Mater. Degrad. 2 (2018). Article number: 37.  https://doi.org/10.1038/s41529-018-0058-x
  3. 3.
    Bonaccorso, G.: Machine Learning Algorithms: Popular Algorithms for Data Science and Machine Learning, 2nd edn. Packt Publishing Ltd., Birmingham (2018)Google Scholar
  4. 4.
    O’Mahony, N., Murphy, T., Panduru, K., et al.: Improving controller performance in a powder blending process using predictive control. In: 2017 28th Irish Signals and Systems Conference (ISSC), pp. 1–6. IEEE (2017)Google Scholar
  5. 5.
    O’Mahony, N., Murphy, T., Panduru, K., et al.: Real-time monitoring of powder blend composition using near infrared spectroscopy. In: 2017 Eleventh International Conference on Sensing Technology (ICST), pp. 1–6. IEEE (2017)Google Scholar
  6. 6.
    O’Mahony, N., Murphy, T., Panduru, K., et al.: Adaptive process control and sensor fusion for process analytical technology. In: 2016 27th Irish Signals and Systems Conference (ISSC), pp. 1–6. IEEE (2016)Google Scholar
  7. 7.
    Koehn, P.: Combining genetic algorithms and neural networks: the encoding problem (1994)Google Scholar
  8. 8.
    Wang, J., Ma, Y., Zhang, L., Gao, R.X.: Deep learning for smart manufacturing: methods and applications. J. Manufact. Syst. 48, 144–156 (2018).  https://doi.org/10.1016/J.JMSY.2018.01.003CrossRefGoogle Scholar
  9. 9.
    Voulodimos, A., Doulamis, N., Doulamis, A., Protopapadakis, E.: Deep learning for computer vision: a brief review. Comput. Intell. Neurosci. 2018, 1–13 (2018).  https://doi.org/10.1155/2018/7068349CrossRefGoogle Scholar
  10. 10.
    Dumoulin, V., Visin, F., Box, G.E.P.: A guide to convolution arithmetic for deep learning. arXiv Prepr arXiv:1603.07285v2 (2018)
  11. 11.
    Hayou, S., Doucet, A., Rousseau, J.: On the selection of initialization and activation function for deep neural networks. arXiv Prepr arXiv:1805.08266v2 (2018)
  12. 12.
    Horiguchi, S., Ikami, D., Aizawa, K.: Significance of softmax-based features in comparison to distance metric learning-based features (2017)Google Scholar
  13. 13.
    Deshpande, A.: A beginner’s guide to understanding convolutional neural networks. CS Undergrad at UCLA (2019). https://adeshpande3.github.io/A-Beginner%27s-Guide-To-Understanding-Convolutional-Neural-Networks/. Accessed 19 July 2018
  14. 14.
    Karami, E., Shehata, M., Smith, A.: Image identification using SIFT algorithm: performance analysis against different image deformations (2017)Google Scholar
  15. 15.
    Bay, H., Tuytelaars, T., Van Gool, L.: SURF: speeded up robust features, pp. 404–417. Springer, Heidelberg (2006)Google Scholar
  16. 16.
    Rosten, E., Drummond, T.: Machine learning for high-speed corner detection, pp. 430–443. Springer, Heidelberg (2006)Google Scholar
  17. 17.
    Goldenshluger, A., Zeevi, A.: The hough transform estimator 32 (2004).  https://doi.org/10.1214/009053604000000760
  18. 18.
    Tsai, F.C.D.: Geometric hashing with line features. Pattern Recogn. 27, 377–389 (1994).  https://doi.org/10.1016/0031-3203(94)90115-5CrossRefGoogle Scholar
  19. 19.
    Wang, J., Perez, L.: The effectiveness of data augmentation in image classification using deep learningGoogle Scholar
  20. 20.
    Schöning, J., Faion, P., Heidemann, G.: Pixel-wise ground truth annotation in videos - an semi-automatic approach for pixel-wise and semantic object annotation. In: Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods, pp. 690–697. SCITEPRESS - Science and Technology Publications (2016)Google Scholar
  21. 21.
    Zhang, X., Lee, J.-Y., Sunkavalli, K., Wang, Z.: Photometric stabilization for fast-forward videos (2017)Google Scholar
  22. 22.
    Alhaija, H.A., Mustikovela, S.K., Mescheder, L., et al.: Augmented reality meets computer vision : efficient data generation for urban driving scenes (2017)Google Scholar
  23. 23.
    Meneghetti, G., Danelljan, M., Felsberg, M., Nordberg, K.: Image alignment for panorama stitching in sparsely structured environments, pp. 428–439. Springer, Cham (2015)Google Scholar
  24. 24.
    Alldieck, T., Kassubeck, M., Magnor, M.: Optical flow-based 3D human motion estimation from monocular video (2017)Google Scholar
  25. 25.
    Zheng, B., Zhao, Y., Yu, J., et al.: Scene understanding by reasoning stability and safety. Int. J. Comput. Vis. 112, 221–238 (2015).  https://doi.org/10.1007/s11263-014-0795-4MathSciNetCrossRefGoogle Scholar
  26. 26.
    Zheng, L., Yang, Y., Tian, Q.: SIFT meets CNN: a decade survey of instance retrievalGoogle Scholar
  27. 27.
    AlDahoul, N., Md Sabri, A.Q., Mansoor, A.M.: Real-time human detection for aerial captured video sequences via deep models. Comput. Intell. Neurosci. 2018, 1–14 (2018).  https://doi.org/10.1155/2018/1639561CrossRefGoogle Scholar
  28. 28.
    Conventional computer vision coupled with deep learning makes AI better. Network World. https://www.networkworld.com/article/3239146/internet-of-things/conventional-computer-vision-coupled-with-deep-learning-makes-ai-better.html. Accessed 12 Sept 2018
  29. 29.
    Bahrampour, S., Ramakrishnan, N., Schott, L., Shah, M.: Comparative study of deep learning software frameworks (2015)Google Scholar
  30. 30.
    An in-depth look at Google’s first tensor processing unit (TPU). Google cloud big data and machine learning blog. Google cloud platform (2017). https://cloud.google.com/blog/big-data/2017/05/an-in-depth-look-at-googles-first-tensor-processing-unit-tpu. Accessed 11 Jan 2018
  31. 31.
    Vision Processing Unit: Machine vision technology. Movidius. https://www.movidius.com/solutions/vision-processing-unit. Accessed 11 Jan 2018
  32. 32.
    Ng, H.-W., Nguyen, D., Vonikakis, V., Winkler, S.: Deep learning for emotion recognition on small datasets using transfer learning.  https://doi.org/10.1145/2818346.2830593
  33. 33.
    Pepik, B., Stark, M., Gehler, P., Schiele, B.: Teaching 3D geometry to deformable part models. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2012)Google Scholar
  34. 34.
    Russakovsky, O., Deng, J., Su, H., et al.: ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. 115, 211–252 (2015).  https://doi.org/10.1007/s11263-015-0816-yMathSciNetCrossRefGoogle Scholar
  35. 35.
    Lin, T.-Y., Maire, M., Belongie, S., et al.: Microsoft COCO: common objects in context (2014)Google Scholar
  36. 36.
    CS231n convolutional neural networks for visual recognition. http://cs231n.github.io/transfer-learning/. Accessed 9 Mar 2018
  37. 37.
    Highlander, T.C.: Efficient training of small kernel convolutional neural networks using fast fourier transformGoogle Scholar
  38. 38.
    Highlander, T., Rodriguez, A.: Very efficient training of convolutional neural networks using fast fourier transform and overlap-and-add (2016)Google Scholar
  39. 39.
    Li, F., Wang, C., Liu, X., et al.: A composite model of wound segmentation based on traditional methods and deep neural networks. Comput. Intell. Neurosci. 2018, 1–12 (2018).  https://doi.org/10.1155/2018/4149103CrossRefGoogle Scholar
  40. 40.
    Nijhawan, R., Das, J., Raman, B.: A hybrid of deep learning and hand-crafted features based approach for snow cover mapping. Int. J. Remote Sens. 1–15 (2018).  https://doi.org/10.1080/01431161.2018.1519277
  41. 41.
    Zeng, G., Zhou, J., Jia, X., et al.: Hand-crafted feature guided deep learning for facial expression recognition. In: 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018), pp. 423–430. IEEE (2018)Google Scholar
  42. 42.
    Burchfiel, B., Konidaris, G.: Hybrid Bayesian eigenobjects: combining linear subspace and deep network methods for 3D robot visionGoogle Scholar
  43. 43.
    Marcus, G.: Deep learning: a critical appraisalGoogle Scholar
  44. 44.
    Nalisnick, E., Smyth, P.: Learning priors for invariance, pp. 366–375 (2018)Google Scholar
  45. 45.
    Diligenti, M., Roychowdhury, S., Gori, M.: Integrating prior knowledge into deep learning. In: 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 920–923. IEEE (2017)Google Scholar
  46. 46.
    Zhu, H., Nie, Y., Yue, T., Cao, X.: The role of prior in image based 3D modeling: a survey. Front. Comput. Sci. 11, 175–191 (2017).  https://doi.org/10.1007/s11704-016-5520-8CrossRefGoogle Scholar
  47. 47.
    Tran, D., Bourdev, L., Fergus, R., et al.: Learning spatiotemporal features with 3D convolutional networks. arXiv Prepr arXiv:1412.0767 (2015)
  48. 48.
    Pang, G., Neumann, U.: 3D point cloud object detection with multi-view convolutional neural network. In: 2016 23rd International Conference on Pattern Recognition (ICPR), pp. 585–590. IEEE (2016)Google Scholar
  49. 49.
    Lan, Q., Wang, Z., Wen, M., et al.: High performance implementation of 3D convolutional neural networks on a GPU. Comput. Intell. Neurosci. 2017, 1–8 (2017).  https://doi.org/10.1155/2017/8348671CrossRefGoogle Scholar
  50. 50.
    Ahmed, E., Saint, A., Shabayek, A.E.R., et al.: Deep learning advances on different 3D data representations: a survey. arXiv Prepr arXiv:1808.01462 (2018)
  51. 51.
    Zhou, Y., Tuzel, O.: VoxelNet: end-to-end learning for point cloud based 3D object detection. arXiv Prepr arXiv:1711.06396 (2017)
  52. 52.
    Qi, C.R., Yi, L., Su, H., Guibas, L.J.: PointNet ++: deep hierarchical feature learning on point sets in a metric space. arXiv Prepr arXiv:1706.02413v1 (2017)
  53. 53.
    Braeger, S., Foroosh, H.: Curvature augmented deep learning for 3D object recognition. In: 2018 25th IEEE International Conference on Image Processing (ICIP), pp. 3648–3652. IEEE (2018)Google Scholar
  54. 54.
    O’Mahony, N., Campbell, S., Krpalkova, L., et al.: Deep learning for visual navigation of unmanned ground vehicles; a review (2018)Google Scholar
  55. 55.
    Karami, E., Prasad, S., Shehata, M.: Image matching using SIFT, SURF, BRIEF and ORB: performance comparison for distorted imagesGoogle Scholar
  56. 56.
    Angelina Uy, M., Hee Lee, G.: PointNetVLAD: deep point cloud based retrieval for large-scale place recognitionGoogle Scholar
  57. 57.
    Camposeco, F., Cohen, A., Pollefeys, M., Sattler, T.: Hybrid scene compression for visual localizationGoogle Scholar
  58. 58.
    Loghmani, M.R., Planamente, M., Caputo, B., Vincze, M.: Recurrent convolutional fusion for RGB-D object recognitionGoogle Scholar
  59. 59.
    Clément, M., Kurtz, C., Wendling, L.: Learning spatial relations and shapes for structural object description and scene recognition. Pattern Recogn. 84, 197–210 (2018).  https://doi.org/10.1016/J.PATCOG.2018.06.017CrossRefGoogle Scholar
  60. 60.
    Ran, L., Zhang, Y., Zhang, Q., et al.: Convolutional neural network-based robot navigation using uncalibrated spherical images. Sensors 17, 1341 (2017).  https://doi.org/10.3390/s17061341CrossRefGoogle Scholar
  61. 61.
    Silva, R.M.A., Feijó, B., Gomes, P.B., et al.: Real time 360° video stitching and streaming. In: ACM SIGGRAPH 2016 Posters on - SIGGRAPH 2016, pp. 1–2. ACM Press, New York (2016)Google Scholar
  62. 62.
    Fernandez-Labrador, C., Perez-Yus, A., Lopez-Nicolas, G., Guerrero, J.J.: Layouts from panoramic images with geometry and deep learningGoogle Scholar
  63. 63.
    Schöning, J., Faion, P., Heidemann, G.: Pixel-wise ground truth annotation in videos - an semi-automatic approach for pixel-wise and semantic object annotation. In: Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods, pp. 690–697. SCITEPRESS - Science and and Technology Publications (2016)Google Scholar
  64. 64.
    Ioannidou, A., Chatzilari, E., Nikolopoulos, S., Kompatsiaris, I.: Deep learning advances in computer vision with 3D data. ACM Comput. Surv. 50, 1–38 (2017).  https://doi.org/10.1145/3042064CrossRefGoogle Scholar
  65. 65.
    Devries, T., Taylor, G.W.: Dataset augmentation in feature space. arXiv Prepr arXiv:1702.05538v1 (2017)
  66. 66.
    Dvornik, N., Mairal, J., Schmid, C.: Modeling visual context is key to augmenting object detection datasetsGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Niall O’Mahony
    • 1
    Email author
  • Sean Campbell
    • 1
  • Anderson Carvalho
    • 1
  • Suman Harapanahalli
    • 1
  • Gustavo Velasco Hernandez
    • 1
  • Lenka Krpalkova
    • 1
  • Daniel Riordan
    • 1
  • Joseph Walsh
    • 1
  1. 1.IMaR Technology GatewayInstitute of Technology TraleeTraleeIreland

Personalised recommendations