Training Deep Neural Networks for Detecting Drinking Glasses Using Synthetic Images

  • Abdul Jabbar
  • Luke Farrawell
  • Jake Fountain
  • Stephan K. Chalup
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10635)

Abstract

This study presents an approach of using synthetically rendered images for training deep neural networks on object detection. A new plug-in for the computer graphics modelling software Blender was developed that can generate large numbers of photo-realistic ray-traced images and include meta information as training labels. The performance of the deep neural network DetectNet is evaluated using training data comprising synthetically rendered images and digital photos of drinking glasses. The detection accuracy is determined by comparing bounding boxes using intersection over union technique. The detection experiments using real-world and synthetic image data resulted in comparable results and the performance increased when using a pre-trained GoogLeNet model. The experiments demonstrated that training deep neural networks for object detection on synthetic data is effective and the proposed approach can be useful for generating large labelled image data sets to enhance the performance of deep neural networks on specific object detection tasks.

Keywords

Deep learning Data augmentation Big data Image processing Ray tracing Object detection Synthetic data generation 

Notes

Acknowledgements

AJ was supported by a UNRSC50:50 scholarship. JF was supported by an Australian Government Research Training Program scholarship. LF was supported by a summer scholarship and sponsorship through 4Tel Pty. In this paper AJ focused on data generation and deep learning, JF and LF focused on development of the Blender plugin for the generation of synthetic data and SKC supervised the project.

References

  1. 1.
    Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press (2016)Google Scholar
  2. 2.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Pereira, F., Burges, C.J.C., Bottou, L., Weinberger, K.Q. (eds.) Advances in Neural Information Processing Systems (NIPS 2012), vol. 25, pp. 1097–1105. Curran Associates, Inc. (2012)Google Scholar
  3. 3.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. CoRR abs/1409.1556 (2014), http://arxiv.org/abs/1409.1556
  4. 4.
    Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A.: Going deeper with convolutions. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1–9 (2015)Google Scholar
  5. 5.
    Imagenet, http://www.image-net.org/. Accessed 03 June 2017
  6. 6.
    Shirley, P., Morley, R.K.: Realistic Ray Tracing, 2nd edn. A. K. Peters Ltd., Natick (2003)Google Scholar
  7. 7.
    Blender Foundation: Blender, https://www.blender.org/. Accessed 27 May 2017
  8. 8.
    Jaderberg, M., Simonyan, K., Vedaldi, A., Zisserman, A.: Synthetic data and artificial neural networks for natural scene text recognition. CoRR abs/1406.2227 (2014), http://arxiv.org/abs/1406.2227
  9. 9.
    Peng, X., Sun, B., Ali, K., Saenko, K.: Learning deep object detectors from 3D models. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1278–1286 (2015)Google Scholar
  10. 10.
    Rajpura, P.S., Hegde, R.S., Bojinov, H.: Object detection using deep CNNs trained on synthetic images arXiv:1706.06782 [cs] (2017)
  11. 11.
    Xu, Y., Nagahara, H., Shimada, A., Taniguchi, R.: Transcut: Transparent object segmentation from a light-field image. CoRR abs/1511.06853 (2015), http://arxiv.org/abs/1511.06853
  12. 12.
    Klank, U., Carton, D., Beetz, M.: Transparent object detection and reconstruction on a mobile platform. In: IEEE International Conference on Robotics and Automation (ICRA), Shanghai, China, 9–13 May (2011)Google Scholar
  13. 13.
    Ihrke, I., Kutulakos, K., Lensch, H., Magnor, M., Heidrich, W.: State of the art in transparent and specular object reconstruction. In: EUROGRAPHICS Star Proceedings, pp. 87–108, EG, Crete, Greece (2008)Google Scholar
  14. 14.
    Tao, A., Barker, J., Sarathy, S.: Detectnet: Deep neural network for object detection in digits (2016), https://devblogs.nvidia.com/parallelforall/detectnet-deep-neural-network-object-detection-digits/. Accessed 20 June 2017
  15. 15.
    Barker, J., Prasanna, S.: Deep learning for object detection with digits (2016), https://devblogs.nvidia.com/parallelforall/deep-learning-object-detection-digits/. Accessed 12 June 2017
  16. 16.
    Michael, L., David, W.: Distance between sets. Nature 234, 34–35 (1971)CrossRefGoogle Scholar
  17. 17.
    Everingham, M., Van Gool, L., Williams, C.K.I., Winn, J., Zisserman, A.: The pascal visual object classes (voc) challenge. Int. J. Comput. Vision 88(2), 303–338 (2010)CrossRefGoogle Scholar
  18. 18.
  19. 19.
    Chaurasiya, R.K., Ramakrishnan, K.R.: High dynamic range imaging. In: 2013 International Conference on Communication Systems and Network Technologies. pp. 83–89, April 2013Google Scholar
  20. 20.
    Debevec, P.: A tutorial on image-based lighting. IEEE Comput. Graph. Appl. (2002), http://ict.usc.edu/pubs/Image-Based%20Lighting.pdf

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Abdul Jabbar
    • 1
  • Luke Farrawell
    • 1
  • Jake Fountain
    • 1
  • Stephan K. Chalup
    • 1
  1. 1.School of Electrical Engineering and ComputingThe University of NewcastleCallaghanAustralia

Personalised recommendations