Neural Style Transfer for Image Stylization

Qifeng Xiang

doi:10.54097/m99phd60

Authors

Qifeng Xiang

DOI:

https://doi.org/10.54097/m99phd60

Keywords:

Neural Style Transfer, Artistic Creativity, Image Style Transfer, Deep Learning in Art.

Abstract

Image style migration, exploring the transformation of visual styles from one image to another, has become a focal point in computer vision research. The semantic and stylistic features of images are difficult to express directly through mathematical models, which greatly increases the difficulty of image stylization. Fortunately, approaches based on deep learning have shown promise in extracting deep semantic information from images, facilitating notable advancements in image style transfer. However, achieving a balance between content preservation and style transformation remains a formidable challenge. This paper introduces a neural style transfer network (NSTN) that aims to maintain image semantics while performing style transfer effectively. The NSTN framework comprises a process block, a style block, and an ascension decoder, working in concert to achieve nuanced style shifts while preserving the content integrity. Implementation results on the WikiArt and COCO datasets demonstrate the model's effectiveness in achieving a harmonious balance between content preservation and style integration.

Downloads

Download data is not yet available.

References

[1] Gatys, L.A., Bethge, M., Hertzmann, A., & Shechtman, E., Preserving Color in Neural Artistic Style Transfer, ArXiv, abs/1606.05897, (2016).

[2] Li, Y., Wang, N., Liu, J., & Hou, X., Demystifying Neural Style Transfer, International Joint Conference on Artificial Intelligence, (2017).

[3] Jing, Y., Yang, Y., Feng, Z., Ye, J., & Song, M., Neural Style Transfer: A Review, IEEE Transactions on Visualization and Computer Graphics, 26, 3365 - 3385 (2017).

[4] Wu, H., Sun, Z., & Yuan, W., Direction-aware Neural Style Transfer, Proceedings of the 26th ACM international conference on Multimedia, (2018).

[5] Liu, X., Cheng, M., Lai, Y., & Rosin, P.L., Depth-aware neural style transfer, International Symposium on Non-Photorealistic Animation and Rendering, (2017).

[6] Gatys, L.A., Ecker, A.S., Bethge, M., Hertzmann, A., & Shechtman, E., Controlling Perceptual Factors in Neural Style Transfer, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 3730 - 3738 (2016).

[7] Gong, X., Huang, H., Ma, L., Shen, F., Liu, W., & Zhang, T., Neural Stereoscopic Image Style Transfer, European Conference on Computer Vision, (2018).

[8] Cheng, M., Liu, X., Wang, J., Lu, S., Lai, Y., & Rosin, P.L., Structure-Preserving Neural Style Transfer, IEEE Transactions on Image Processing, 29, 909 - 920 (2020).

[9] Wu, H., Sun, Z., Zhang, Y., & Li, Q., Direction-aware neural style transfer with texture enhancement, Neurocomputing, 370, 39 - 55 (2019).

[10] Li, J., Wang, Q., Chen, H., An, J., & Li, S., A Review on Neural Style Transfer, Journal of Physics: Conference Series, 1651 (2020).

[11] Shibly, K.H., Rahman, S., Dey, S.K., & Shamim, S.H., Advanced Artistic Style Transfer Using Deep Neural Network, (2020).

[12] Xu, Z., Tao, D., Zhang, Y., Wu, J., & Tsoi, A.C., Architectural Style Classification Using Multinomial Latent Logistic Regression, European Conference on Computer Vision, (2014).

[13] Lin, T., Maire, M., Belongie, S.J., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C.L., Microsoft COCO: Common Objects in Context, European Conference on Computer Vision, (2014).

[14] Vaswani, A., Shazeer, N.M., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., & Polosukhin, I., Attention is All you Need, Neural Information Processing Systems, (2017).

[15] Phillips, F.Y., & Mackintosh, B., Wiki Art Gallery, Inc.: A Case for Critical Thinking, Issues in Accounting Education, 26, 593 - 608 (2011).

[16] Kingma, D.P., & Ba, J., Adam: A Method for Stochastic Optimization, CoRR, abs/1412.6980 (2014).