Advertisement

Generative Adversarial Networks with Enhanced Symmetric Residual Units for Single Image Super-Resolution

  • Xianyu Wu
  • Xiaojie Li
  • Jia He
  • Xi Wu
  • Imran Mumtaz
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11295)

Abstract

In this paper, we propose a new generative adversarial network (GAN) with enhanced symmetric residual units for single image super-resolution (ERGAN). ERGAN consists of a generator network and a discriminator network. The former can maximally reconstruct a super-resolution image similar to the original image. This lead to the discriminator network cannot distinguish the image from the training data or the generated sample. Combining residual units used in the generator network, ERGAN can retain the high-frequency features and alleviate the difficulty training in deep networks. Moreover, we constructed the symmetric skip-connections in residual units. This reused features generated from the low-level, and learned more high-frequency content. Moreover, ERGAN reconstructed the super-resolution image by four times the length and width of the original image and exhibited better visual characteristics. Experimental results on extensive benchmark evaluation showed that ERGAN significantly outperformed state-of-the-art approaches in terms of accuracy and vision.

Keywords

Super-resolution GAN Residual units Symmetric skip-connection 

Notes

Acknowledgment

This work was supported by the National Natural Science Foundation of China (Grant Nos. 61602066) and by the Project Supported by the Scientific Research Foundation of the Education Department of Sichuan Province(17ZA0063 and 2017JQ0030) and the Scientific Research Foundation (KYTZ201608) of CUIT, and partially supported by Sichuan International Science and Technology Cooperation and Exchange Research Program (2016HH0018), and Sichuan Science and Technology Program (2018GZ0184).

References

  1. 1.
    Bruna, J., Sprechmann, P., Lecun, Y.: Super-resolution with deep convolutional sufficient statistics. Comput. Sci. (2015)Google Scholar
  2. 2.
    Dong, C., Chen, C.L., He, K., Tang, X.: Image super-resolution using deep convolutional networks. IEEE Trans. Pattern Anal. Mach. Intell. 38(2), 295–307 (2016)CrossRefGoogle Scholar
  3. 3.
    Denton, E., Chintala, S., Szlam, A., Fergus, R.: Deep generative image modelsusing a Laplacian pyramid of adversarial networks. In: International Conference on Neural Information Processing Systems, pp. 1486–1494 (2015)Google Scholar
  4. 4.
    Goodfellow, I.J., et al.: Generative adversarial nets. In: International Conference on Neural Information Processing Systems, pp. 2672–2680 (2014)Google Scholar
  5. 5.
    Huang, J.B., Singh, A., Ahuja, N.: Single image super-resolution from transformed self-exemplars. In: Computer Vision and Pattern Recognition, pp. 5197–5206 (2015)Google Scholar
  6. 6.
    Lai, W.S., Huang, J.B., Ahuja, N., Yang, M.H.: Deep Laplacian pyramid networks for fast and accurate super-resolution. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 5835–5843 (2017)Google Scholar
  7. 7.
    Ledig, C., et al.: Photo-realistic single image super-resolution using a generative adversarial network. In: Computer Vision and Pattern Recognition, pp. 105–114 (2017)Google Scholar
  8. 8.
    Li, J., Liang, X., Wei, Y., Xu, T., Feng, J., Yan, S.: Perceptual generative adversarial networks for small object detection. In: Computer Vision and Pattern Recognition, pp. 1951–1959 (2017)Google Scholar
  9. 9.
    Lim, B., Son, S., Kim, H., Nah, S., Lee, K.M.: Enhanced deep residual networks for single image super-resolution. In: Computer Vision and Pattern Recognition Workshops, pp. 1132–1140 (2017)Google Scholar
  10. 10.
    Lin, Z., Shum, H.Y.: On the fundamental limits of reconstruction-based super-resolution algorithms. In: Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2001, vol. 1, pp. I-1171-I-1176 (2001)Google Scholar
  11. 11.
    Nuno-Maganda, M.A., Arias-Estrada, M.O.: Real-time FPGA-based architecture for bicubic interpolation: an application for digital image scaling. In: International Conference on Reconfigurable Computing and FPGAs, p. 1 (2005)Google Scholar
  12. 12.
    Radford, A., Metz, L., Chintala, S.: Unsupervised representation learning with deep convolutional generative adversarial networks. Comput. Sci. (2015)Google Scholar
  13. 13.
    Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015).  https://doi.org/10.1007/978-3-319-24574-4_28CrossRefGoogle Scholar
  14. 14.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. Comput. Sci. (2014)Google Scholar
  15. 15.
    Song, H., Huang, B., Liu, Q., Zhang, K.: Improving the spatial resolution of landsat TM/ETM+ through fusion with SPOT5 images via learning-based super-resolution. IEEE Trans. Geosci. Remote Sens. 53(3), 1195–1204 (2014)CrossRefGoogle Scholar
  16. 16.
    Tai, Y., Yang, J., Liu, X.: Image super-resolution via deep recursive residual network. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 2790–2798 (2017)Google Scholar
  17. 17.
    Timofte, R., De Smet, V., Van Gool, L.: A+: adjusted anchored neighborhood regression for fast super-resolution. In: Cremers, D., Reid, I., Saito, H., Yang, M.-H. (eds.) ACCV 2014. LNCS, vol. 9006, pp. 111–126. Springer, Cham (2015).  https://doi.org/10.1007/978-3-319-16817-3_8CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Xianyu Wu
    • 1
  • Xiaojie Li
    • 1
  • Jia He
    • 1
  • Xi Wu
    • 1
  • Imran Mumtaz
    • 2
  1. 1.Chengdu University of Information TechnologyChengduChina
  2. 2.University of Agriculture FaisalabadFaisalabadPakistan

Personalised recommendations