Skip to main content
SpringerLink
Log in

Iterative Feature Transformation for Fast and Versatile Universal Style Transfer

  • Conference paper
  • First Online:
Computer Vision – ECCV 2020 (ECCV 2020)

Part of the book series: Lecture Notes in Computer Science ((LNIP,volume 12364))

Included in the following conference series:

Abstract

The general framework for fast universal style transfer consists of an autoencoder and a feature transformation at the bottleneck. We propose a new transformation that iteratively stylizes features with analytical gradient descent (Implementation is open-sourced at https://github.com/chiutaiyin/Iterative-feature-transformation-for-style-transfer). Experiments show this transformation is advantageous in part because it is fast. With control knobs to balance content preservation and style effect transferal, we also show this method can switch between artistic and photo-realistic style transfers and reduce distortion and artifacts. Finally, we show it can be used for applications requiring spatial control and multiple-style transfer.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 84.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 109.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    Previous analysis [3] shows that the feature \(\mathbf {F}_{wct}\) derived by applying WCT to \(\mathbf {F}_{N,c}\) and \(\mathbf {F}_{N,s}\) makes the value of the style loss in Eq. 4 go to zero, and hence could serve as an approximate solution. However, WCT does not consider the balance between soft proximity loss and style loss.

  2. 2.

    We found \(n_{iter}=3\) is sufficient for convergence with little difference from \(n_{iter}=2\).

  3. 3.

    Our reported times include the computation time for style image encoding and its relevant terms, as done for some papers [2, 8, 13, 15] but not others [21].

  4. 4.

    While a fancy autoencoder such as WCT2 [25] using wavelet pooling can further prevent distortion, that is beyond the scope of this paper.

  5. 5.

    Our aim was to capture the distortion level caused by different transformations. While a user study can reflect aesthetics, it is hard for users to notice all distortions in an image and the score scale can be only coarse-grained (eg, 0 = no distortion, 5 = worst distortion). That motivated our choice to use quantitative metrics.

  6. 6.

    We use the MATLAB implementation, setting the luminance, contrast, and structural exponents to 1 and regularization constants to \(0.01^2\), \(0.03^2\), and \(0.03^2/2\).

  7. 7.

    We adopt the official implementation with default hyper-parameters, which computes phase congruency with Kovesi’s method and log-Gabor filters and the gradient magnitude based on the Scharr operator.

  8. 8.

    Due to limited space and a similar trend of linear transition from one style to the other, we show results from AdaIN and Avatar-net in the Supplementary Materials.

References

  1. Champandard, A.J.: Semantic style transfer and turning two-bit doodles into fine artworks. arXiv preprint arXiv:1603.01768 (2016)

  2. Chen, T.Q., Schmidt, M.: Fast patch-based style transfer of arbitrary style. arXiv preprint arXiv:1612.04337 (2016)

  3. Chiu, T.Y.: Understanding generalized whitening and coloring transform for universal style transfer. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 4452–4460 (2019)

    Google Scholar 

  4. Dumoulin, V., Shlens, J., Kudlur, M.: A learned representation for artistic style. In: 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, 24–26 April 2017, Conference Track Proceedings. OpenReview.net (2017). https://openreview.net/forum?id=BJO-BuT1g

  5. Gatys, L.A., Ecker, A.S., Bethge, M.: Image style transfer using convolutional neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2414–2423 (2016)

    Google Scholar 

  6. Gatys, L.A., Ecker, A.S., Bethge, M., Hertzmann, A., Shechtman, E.: Controlling perceptual factors in neural style transfer. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3985–3993 (2017)

    Google Scholar 

  7. Ghiasi, G., Lee, H., Kudlur, M., Dumoulin, V., Shlens, J.: Exploring the structure of a real-time, arbitrary neural artistic stylization network. In: Kim, T.K., Zafeiriou, S., Brostow, G., Mikolajczyk, K. (eds.) Proceedings of the British Machine Vision Conference (BMVC), pp. 114.1–114.12. BMVA Press, September 2017. http://doi.org/10.5244/C.31.114

  8. Huang, X., Belongie, S.: Arbitrary style transfer in real-time with adaptive instance normalization. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1501–1510 (2017)

    Google Scholar 

  9. Johnson, J., Alahi, A., Fei-Fei, L.: Perceptual losses for real-time style transfer and super-resolution. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9906, pp. 694–711. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46475-6_43

    Chapter  Google Scholar 

  10. Li, C., Wand, M.: Combining Markov random fields and convolutional neural networks for image synthesis. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2479–2486 (2016)

    Google Scholar 

  11. Li, C., Wand, M.: Precomputed real-time texture synthesis with Markovian generative adversarial networks. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9907, pp. 702–716. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46487-9_43

    Chapter  Google Scholar 

  12. Li, P., Zhao, L., Xu, D., Lu, D.: Optimal transport of deep feature for image style transfer. In: Proceedings of the 2019 4th International Conference on Multimedia Systems and Signal Processing, pp. 167–171 (2019)

    Google Scholar 

  13. Li, X., Liu, S., Kautz, J., Yang, M.H.: Learning linear transformations for fast image and video style transfer. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3809–3817 (2019)

    Google Scholar 

  14. Li, Y., Fang, C., Yang, J., Wang, Z., Lu, X., Yang, M.H.: Diversified texture synthesis with feed-forward networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3920–3928 (2017)

    Google Scholar 

  15. Li, Y., Fang, C., Yang, J., Wang, Z., Lu, X., Yang, M.H.: Universal style transfer via feature transforms. In: Guyon, I., et al. (eds.) Advances in Neural Information Processing Systems, pp. 386–396 (2017)

    Google Scholar 

  16. Li, Y., Liu, M.-Y., Li, X., Yang, M.-H., Kautz, J.: A closed-form solution to photorealistic image stylization. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11207, pp. 468–483. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01219-9_28

    Chapter  Google Scholar 

  17. Lu, M., Zhao, H., Yao, A., Chen, Y., Xu, F., Zhang, L.: A closed-form solution to universal style transfer. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 5952–5961 (2019)

    Google Scholar 

  18. Luan, F., Paris, S., Shechtman, E., Bala, K.: Deep photo style transfer. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4990–4998 (2017)

    Google Scholar 

  19. Mroueh, Y.: Wasserstein style transfer. arXiv preprint arXiv:1905.12828 (2019)

  20. Risser, E., Wilmot, P., Barnes, C.: Stable and controllable neural texture synthesis and style transfer using histogram losses. arXiv preprint arXiv:1701.08893 (2017)

  21. Sheng, L., Lin, Z., Shao, J., Wang, X.: Avatar-Net: multi-scale zero-shot style transfer by feature decoration. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 8242–8250 (2018)

    Google Scholar 

  22. Ulyanov, D., Lebedev, V., Vedaldi, A., Lempitsky, V.S.: Texture networks: feed-forward synthesis of textures and stylized images. In: Balcan, M.F., Weinberger, K.Q. (eds.) ICML, vol. 1, p. 4 (2016)

    Google Scholar 

  23. Ulyanov, D., Vedaldi, A., Lempitsky, V.: Improved texture networks: maximizing quality and diversity in feed-forward stylization and texture synthesis. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 6924–6932 (2017)

    Google Scholar 

  24. Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)

    Article  Google Scholar 

  25. Yoo, J., Uh, Y., Chun, S., Kang, B., Ha, J.W.: Photorealistic style transfer via wavelet transforms. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 9036–9045 (2019)

    Google Scholar 

  26. Zhang, H., Dana, K.: Multi-style generative network for real-time transfer. In: Leal-Taixé, L., Roth, S. (eds.) ECCV 2018. LNCS, vol. 11132, pp. 349–365. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-11018-5_32

    Chapter  Google Scholar 

  27. Zhang, L., Zhang, L., Mou, X., Zhang, D.: FSIM: a feature similarity index for image quality assessment. IEEE Trans. Image Process. 20(8), 2378–2386 (2011)

    Article  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. University of Texas at Austin, Austin, USA

    Tai-Yin Chiu & Danna Gurari

Authors
  1. Tai-Yin Chiu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Danna Gurari
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Tai-Yin Chiu .

Editor information

Editors and Affiliations

  1. University of Oxford, Oxford, UK

    Andrea Vedaldi

  2. Graz University of Technology, Graz, Austria

    Horst Bischof

  3. University of Freiburg, Freiburg im Breisgau, Germany

    Thomas Brox

  4. University of North Carolina at Chapel Hill, Chapel Hill, NC, USA

    Jan-Michael Frahm

1 Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 24492 KB)

Rights and permissions

Reprints and permissions

Copyright information

© 2020 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Chiu, TY., Gurari, D. (2020). Iterative Feature Transformation for Fast and Versatile Universal Style Transfer. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, JM. (eds) Computer Vision – ECCV 2020. ECCV 2020. Lecture Notes in Computer Science(), vol 12364. Springer, Cham. https://doi.org/10.1007/978-3-030-58529-7_11

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-58529-7_11

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-58528-0

  • Online ISBN: 978-3-030-58529-7

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation