一种改进CycleGAN的素描头像彩色化算法  

A Colorization Algorithm for Sketch Heads with Improved CycleGAN

作　　者：廖振 林国军 黄丹 胡鑫 游松 兰江海 周旭 金若水 LIAO Zhen;LIN Guojun;Huang Dan;HU Xin;YOU Song;LAN Jianghai;ZHOU Xu;JIN Ruoshui(School of Automation and Information Engineering,Sichuan University of Science&Engineering,Yibin,Sichuan 644000,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,四川宜宾644000

出　　处：《宜宾学院学报》2024年第6期21-26,共6页Journal of Yibin University

基　　金：四川省科技厅项目“基于慢性帕金森病猕猴模型的运动皮层神经元编码规律研究”(2022YFSY0056)。

摘　　要：针对现阶段由素描头像生成的彩色头像图像清晰度低、人脸识别率不高和视觉质量不佳等问题,提出一种改进CycleGAN的素描头像彩色化算法:对U-Net自编码器的第一个特征提取模块进行优化,设计一种多尺度自注意力机制特征提取模块,从多个尺度提取输入图像以减少输入图像的细节信息丢失,将提取的特征用通道堆叠的方式进行特征融合,对融合的特征嵌入SENet自注意力机制,以引导模型对特征重点区域的关注度,最后再降低融合特征的通道维数;对生成头像与真实头像添加L1像素损失和感知损失,以进一步提升生成头像的质量.实验结果表明:较基础模型CycleGAN生成的彩色头像,在CUHK数据集FID值降低了22.23、Rank-1值提高了16%,在AR数据集FID值降低了15.34、Rank-1值提高了9.3%.Aiming at the problems of low clarity,low face recognition rate and poor visual quality of color avatar images gener⁃ated from sketch avatars at the present stage,a colorization algorithm for sketch avatars improving CycleGAN was proposed:by optimizing the first feature extraction module of the U-Net self-encoder,a multi-scale self-attention mechanism feature extrac⁃tion module was designed to extract the input image from multiple scales to reduce the loss of detail information of the input im⁃age.The extracted features were fused by means of channel stacking,and the fused features were embedded with SENet selfattention mechanism to direct the models attention to the feature focus area.Finally,the dimension of fused features was reduced.L1 pixel loss and perceptual loss were added to the generated and real avatars to further improve the quality of the generated ava⁃tars.The experimental results show that compared with the color avatar generated by the base model CycleGAN,the FID value of the CUHK dataset is reduced by 22.23 and Rank-1 value is improved by 16%,and the FID value of the AR dataset is reduced by 15.34 and Rank-1 value is improved by 9.3%.

关 键 词：CycleGAN 多尺度特征提取 SENet 监督学习 L_1像素损失 感知损失 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]