|本期目录/Table of Contents|

[1]祝亮亮,郭业才.基于双重注意力网络和内容修复损失的艺术风格迁移[J].浙江理工大学学报,2026,55-56(自科一):105-113.
　ZHU Liangliang,GUO Yecai.Artistic style transfer based on dual attention network and content restoration loss[J].Journal of Zhejiang Sci-Tech University,2026,55-56(自科一):105-113.
点击复制

基于双重注意力网络和内容修复损失的艺术风格迁移()
分享到：

浙江理工大学学报[ISSN:1673-3851/CN:33-1338/TS]

卷:
55-56
期数:
2026年自科第一期
页码:
105-113
栏目:
出版日期:
2026-01-10

文章信息/Info

Title:
Artistic style transfer based on dual attention network and content restoration loss
文章编号:
1673-3851(2026) 01-0105-09
作者:
祝亮亮 ; 郭业才
1. 安徽建筑大学电子与信息工程学院 ,合肥 230601;2. 南京信息工程大学电子与信息工程学院 , 南京 210044
Author(s):
ZHU Liangliang;  GUO Yecai
1. School of Electronic and Information Engineering, Anhui Jianzhu University, Hefei 230601, China; 2. School of Electronic & Information Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, China
关键词:
深度学习; 卷积神经网络; 风格迁移; 空间注意力 ; 通道注意力
分类号:
TP394.41
文献标志码:
A
摘要:
针对现有艺术风格迁移网络在迁移过程中难以精确保持生成图像的结构细节 , 以及生成图像中来自风格图像映射痕迹明显的问题 ,提出了一种基于双重注意力网络和内容修复损失的艺术风格迁移网络 DatNet。该网络通过卷积核可动态调整的轻量化通道注意力模块 , 实现对图像特征分布的再优化;同时在空间注意模块中 ,通过学习相关矩阵的高阶特征 , 实现对风格特征的精细建模 。 另外 ,设计了一种内容修复损失函数 , 以 内容图像为双输入生成图像 , 并与原始内容图像在多层特征空间中进行差异约束 , 以增强网络对生成图像结构特征的保持能力。 DatNet与主流网络在客观指标上进行横向对比实验 ,结果表明 ,基于双重注意力网络和内容修复损失的艺术风格迁移生成的图像 ,在结构相似性(Structure similarity index measure,SSIM) 和峰值信噪比(Peak signal-to-noise ratio, PSNR)上较 MicroAST分别提升了 0.01和 0.66。该网络将通道维度特征动态优化与空间相关矩阵的高阶特征匹配相结合 ,计算以内容图像为双输入的生成图像与内容图像之间多层特征的差异 ,在显著提升生成图像内容结构清晰度的同时 ,有效降低了风格图像对生成图像的映射干扰 ,展现出较高的实用价值。

参考文献/References:

[1] 胡琦瑶, 刘乾龙, 彭先霖, 等. SN-CLPGAN:基于谱归一化的中国传统山水画风格迁移方法[J]. 西北大学学报(自然科学版), 2025, 55(1): 63-74.
[2] 姚琳涵, 张颖, 姚岚, 等. 基于多尺度纹理合成的刺绣风格迁移模型[J]. 纺织学报, 2023, 44(9): 84-90.
[3] 邱雪琳, 孙迎, 刘正, 等. 基于色彩优化的云锦局部风格迁移[J]. 丝绸, 2022, 59(12): 64-70.
[4] 胡景浩, 耿国华, 熊美均, 等. 基于细节特征提取融合的中国山水画风格迁移[J]. 工程科学与技术, 2025, 57(1): 98-106.
[5] 邓海欣, 张凤全, 王楠, 等. 基于潜在空间的动漫人脸风格迁移与编辑方法[J]. 系统仿真学报, 2024, 36(12): 2834-2849.
[6] 李云红, 朱景坤, 刘杏瑞, 等. 基于改进生成对抗网络的动漫图像风格迁移算法[J]. 北京邮电大学学报, 2024, 47(4): 117-123.
[7] Gatys L A, Ecker A S, Bethge M. Image style transfer using convolutional neural networks[C] // 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 27-30, 2016, Las Vegas, NV, USA. IEEE, 2016: 2414-2423.
[8] Johnson J, Alahi A, Fei-Fei L. Perceptual losses for real-time style transfer and super-resolution[C] // Computer Vision-ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part II 14. Cham: Springer International Publishing, 2016: 694-711.
[9] Ulyanov D, Vedaldi A, Lempitsky V. Instance normalization: The missing ingredient for fast stylization[EB/OL]. (2017-11-6)[2025-05-15]. https://arxiv.org/abs/1607.08022.
[10] Ulyanov D, Vedaldi A, Lempitsky V. Improved texture networks: Maximizing quality and diversity in feed-forward stylization and texture synthesis[C] // 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). July 21-26, 2017, Honolulu, HI, USA. IEEE, 2017: 4105-4113.
[11] Dumoulin V, Shlens J, Kudlur M. A learned representation for artistic style[EB/OL]. (2017-02-16)[2025-05-15]. https://arxiv.org/abs/1610.07629.
[12] Chen D, Yuan L, Liao J, et al. Stylebank: An explicit representation for neural image style transfer[C] // 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). July 21-26, 2017, Honolulu, HI, USA. IEEE, 2017: 2770-2779.
[13] Huang X, Belongie S. Arbitrary style transfer in real-time with adaptive instance normalization[C] // 2017 IEEE International Conference on Computer Vision (ICCV). October 22-29, 2017, Venice, Italy. IEEE, 2017: 1510-1519.
[14] Park D Y, Lee K H. Arbitrary style transfer with style-attentional networks[C] // 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 15-20, 2019, Long Beach, CA, USA. IEEE, 2019: 5873-5881.
[15] Liu S H, Lin T W, He D L, et al. AdaAttN: Revisit attention mechanism in arbitrary neural style transfer[C] // 2021 IEEE/CVF International Conference on Computer Vision (ICCV). October 10-17, 2021, Montreal, QC, Canada. IEEE, 2021: 6629-6638.
[16] Li Y, Fang C, Yang J, et al. Universal style transfer via feature transforms[C] // Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS’17). Red Hook, NY, USA: Curran Associates Inc., 2017: 385-395.
[17] Wang Z, Zhao L, Chen H, et al. Diversified arbitrary style transfer via deep feature perturbation[C] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 13-19, 2020, Seattle, WA, USA. IEEE, 2020: 7786-7795.
[18] Chen Y Q, Zheng Y T, Yan Z W. Decoupling and coupling transformation for arbitrary style transfer[C] // 2021 6th International Conference on Image, Vision and Computing (ICIVC). July 23-25, 2021, Qingdao, China. IEEE, 2021: 227-232.
[19] Wang Z, Zhao L, Zuo Z, et al. MicroAST: Towards super-fast ultra-resolution arbitrary style transfer[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(3): 2742-2750.
[20] Wang Q L, Wu B G, Zhu P F, et al. ECA-net: Efficient channel attention for deep convolutional neural networks[C] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 13-19, 2020, Seattle, WA, USA. IEEE, 2020: 11531-11539.
[21] An J, Huang S Y, Song Y B, et al. ArtFlow: Unbiased image style transfer via reversible neural flows[C] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 20-25, 2021, Nashville, TN, USA. IEEE, 2021: 862-871.
[22] Deng Y Y, Tang F, Dong W M, et al. Arbitrary style transfer via multi-adaptation network[C] // Proceedings of the 28th ACM International Conference on Multimedia. Seattle, WA, USA: ACM, 2020: 2719-2727.
[23] Deng Y Y, Tang F, Dong W M, et al. Arbitrary video style transfer via multi-channel correlation[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(2): 1210-1217.

相似文献/References:

[1]李斯凡,高法钦.基于卷积神经网络的手写数字识别[J].浙江理工大学学报,2017,37-38(自科3):438.
　LI Sifan,GAO Faqin.Handwritten Numeral Recognition Based on Convolution Neural Network[J].Journal of Zhejiang Sci-Tech University,2017,37-38(自科一):438.
[2]张玮,张华熊.基于卷积神经网络的纺织面料主成分分类[J].浙江理工大学学报,2019,41-42(自科一):1.
　ZHANG Wei,ZHANG Huaxiong.Classification of main components of textile fabrics based on convolutional neural network[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):1.
[3]包晓安,徐海,张娜,等.基于深度学习的语音识别模型及其在智能家居中的应用[J].浙江理工大学学报,2019,41-42(自科二):217.
　BAO Xiaoan,XU Hai,ZHANG Na,et al.Speech recognition model based on deep learning and its application in smart home[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):217.
[4]包晓安,高春波,张娜,等.基于生成对抗网络的图像超分辨率方法[J].浙江理工大学学报,2019,41-42(自科四):499.
　BAO Xiaoan,GAO Chunbo,ZHANG Na,et al.Image superresolution method based ongenerative adversarial network[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):499.
[5]陈巧红,王磊,孙麒,等.基于混合神经网络的中文短文本分类模型[J].浙江理工大学学报,2019,41-42(自科四):509.
　CHEN Qiaohong,WANG Lei,SUN Qi,et al.Chinese short text classification model based on hybrid neural network[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):509.
[6]陈巧红,陈翊,孙麒,等.服装图像分类技术综述[J].浙江理工大学学报,2019,41-42(自科五):631.
　CHEN Qiaohong,CHEN Yi,SUN Qi,et al.Overview of clothing image classification technology[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):631.
[7]程诚,任佳.基于自适应卷积核的改进CNN数值型数据分类算法[J].浙江理工大学学报,2019,41-42(自科五):657.
　CHENG Cheng,REN Jia.Improved CNN classification algorithm based on adaptive convolution kernel for numerical data[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):657.
[8]包晓安,涂小妹,徐璐,等.基于扩展卷积神经网络与度量学习的指静脉识别[J].浙江理工大学学报,2020,43-44(自科二):232.
　BAO Xiaoan,TU Xiaomei,XU Lu,et al.Finger vein recognition based on extended convolutional neural networks and metric learning[J].Journal of Zhejiang Sci-Tech University,2020,43-44(自科一):232.
[9]潘海鹏,郝慧,苏雯.基于注意力机制与多尺度特征融合的人脸表情识别[J].浙江理工大学学报,2022,47-48(自科三):382.
　PAN Haipeng,HAO Hui,SU Wen.Facial expression recognition based on attention mechanism and multiscale feature fusion[J].Journal of Zhejiang Sci-Tech University,2022,47-48(自科一):382.
[10]郭波,吕文涛,余序宜,等.基于改进YOLOv5模型的织物疵点检测算法[J].浙江理工大学学报,2022,47-48(自科五):755.
　GUO Bo,LV Wentao,YU Xuyi,et al.Fabric defect detection algorithm based on improved YOLOv5 Model[J].Journal of Zhejiang Sci-Tech University,2022,47-48(自科一):755.
[11]邓远远,沈炜.基于注意力反馈机制的深度图像标注模型[J].浙江理工大学学报,2019,41-42(自科二):208.
　DENG Yuanyuan,SHEN Wei.Depth image caption model based on  attention feedback mechanism[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):208.
[12]邓远远,沈炜.基于注意力反馈机制的深度图像标注模型[J].浙江理工大学学报,2019,41-42(自科二):208.
　DENG Yuanyuan,SHEN Wei.Depth image caption model based on attention feedback mechanism[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):208.
[13]陈巧红,董雯,孙麒,等.基于混合神经网络的单文档自动文摘模型[J].浙江理工大学学报,2019,41-42(自科四):489.
　CHEN Qiaohong,DONG Wen,SUN Qi,et al.Single document automatic summarization model based on hybrid neural network[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科一):489.

备注/Memo

备注/Memo:
基金项目 : 国家自然科学基金项目(61673222)收稿日期 : 2025-05-15 网络出版日期 : 2025-09-08
作者简介 : 祝亮亮(1999— ) ,男 ,贵州遵义人 ,硕士研究生 ,主要从事深度学习方面的研究。通信作者 : 郭业才 ,E-mail:guo-yecai@163. com
更新日期/Last Update: 2026-01-08