主体结构保持的增强型图像风格传递方法

doi:10.16451/j.cnki.issn1003-6059.201811004

Abstract
Figure/Table
References
Related Citation (3)

Download: PDF (7106 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract In the existing image style transferring methods, the primary structure of the target image is often deformed while a distinctive style is transferred. Therefore, a loss function for style transferring based on DCNN is designed. In addition to the original items]two more regularization terms are introduced in the function. To maintain the primary structure of the target image, the edge information extracted by the LoG operator is used as the feature for the primary structure. The first regularization term is constituted by feature difference between the resultant image and the target image. The second regularization term is composed of features obtained by Gabor filter to enhance the description of directional style features since artistic style is closely related to directional characteristics such as strokes, texture orientation and color flow while depth features are more focused on depicting global information. This item avoids the weakening effect of transferred style due to the maintenance of the primary structure. The experimental results show that the proposed method maintains better primary structure of the target image while successfully transferring a distinctive style.

Key words： Primary Structure Maintenance Image Style Transfer Deep Convolutional Neural Network(DCNN) Laplacian of Gaussian(LoG) Operator Gabor Filter

Received: 18 May 2018

ZTFLH:

TP 391.41

Fund:Supported by National Natural Science Foundation of China(No.61672158), Natural Science Foundation of Fujian Province(No.2018J1798,2016J05155)

Corresponding Authors: CHEN Zhaojiong, master, professor. Her research interests include intelligent image processing and computational intelligence.

About author:: LIN Xing, master student. His research interests include image processing;YE Dongyi, Ph.D., professor. His research interests include computational intelligence and data mining.

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	LIN Xing
	CHEN Zhaojiong
	YE Dongyi

Cite this article:

LIN Xing,CHEN Zhaojiong,YE Dongyi. Enhanced Image Style Transferring Method with Primary Structure Maintained[J]. , 2018, 31(11): 997-1007.

URL:

http://manu46.magtech.com.cn/Jweb_prai/EN/10.16451/j.cnki.issn1003-6059.201811004 OR http://manu46.magtech.com.cn/Jweb_prai/EN/Y2018/V31/I11/997

[1]REINHARD E, ADHIKHMIN M, GOOCH B, et al. Color Transfer between Images. IEEE Computer Graphics and Applications, 2001, 21(5): 34-41.
[2]HERTZMANN A, JACOBS C E, OLIVER N, et al. Image Analogies[C/OL]. [2018-04-25]. https://mrl.nyu.edu/publications/image-analogies.
[3]ASHIKHMIN N. Fast Texture Transfer. IEEE Computer Graphics and Applications, 2003, 23(4): 38-43.
[4]王洋,刘宏志,吴中海.基于非线性滤波和纹理传输的图像风格化.计算机辅助设计与图形学学报, 2015, 27(7): 1255-1262.
(WANG Y, LIU H Z, WU Z H. Image Stylization by Non-linear Filtering and Texture Transfer. Journal of Computer-Aided Design and Computer Graphics,2015, 27(7): 1255-1262.)
[5]GATYS L A, ECKER A S, BETHGE M. Image Style Transfer Using]Convolutional Neural Networks // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 2414-2423.
[6]ULYANOV D, LEBEDEV V, VEDALDI A, et al.Texture Networks: Feed-Forward Synthesis of Textures and Stylized Images[C/OL]. [2018-04-25]. https://arxiv.org/pdf/1603.03417.pdf.
[7]JOHNSON J, ALAHI A, LI F F. Perceptual Losses for Real-Time Style Transfer and Super-Resolution // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2016: 694-711.
[8]LI Y J, FANG C, YANG J M, et al.]Diversified Texture Synthesis with Feed-Forward Networks // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 266-274.
[9]CHEN T Q, SCHMIDT M. Fast Patch-Based Style Transfer of Arbitrary Style[C/OL]. [2018-04-25]. https://arxiv.org/pdf/1612.04337v1.pdf.
[10]RISSER E, WILMOT P, BARNES C. Stable and Controllable Neu-ral Texture Synthesis and Style Transfer Using Histogram Losses[C/OL]. [2018-04-25]. https://arxiv.org/pdf/1701.08893.pdf.
[11]SIMONYAN K, ZISSERMAN A. Very Deep Convolutional Networks for Large-Scale Image Recognition[J/OL]. [2018-04-25]. https://arxiv.org/pdf/1409.1556.pdf.
[12]ZHU C Y, BYRD R H, LU P H]et al.Algorithm 778: L-BFGS-B-Fortran Subroutines for Large-Scale Bound Constrained Optimization. ACM Transactions on Mathematical Software, 1997, 23(4): 550-560.
[13]RAMAKRISHNAN A G, RAJA S K, RAM H V R. Neural Network-Based Segmentation of Textures Using Gabor Features // Proc of the 12th IEEE Workshop on Neural Networks for Signal Processing. Washington, USA: IEEE, 2002: 365-374.
[14]MANJUNATH B S, MA W Y. Texture Features for Browsing and Retrieval of Image Data. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 18(8): 837-842.
[15]HAGHIGHAT M, ZONOUZ S, ABDEL-MOTTALEB M. Identification Using Encrypted Biometrics // Proc of the International Conference on Computer Analysis of Images and Patterns. Berlin, Germany: Springer, 2013: 440-448.