| 基于生成对抗网络的图像动漫化 |
| |
| 引用本文： | 翟慧聪,张明,邓星,王利群. 基于生成对抗网络的图像动漫化[J]. 计算机与现代化, 2022, 0(7): 21-26 |
| |
| 作者姓名： | 翟慧聪 张明 邓星 王利群 |
| |
| 基金项目： | 国家自然科学基金青年科学基金资助项目（61902158） |
| |
| 摘 要： | 动漫风格的图像具有高度的简化和抽象等特征，为了解决将现实世界图像转化成动漫风格图像这一问题，提出一种基于生成对抗网络的图像动漫化方法。本文的生成网络是类U-Net的全卷积结构，对输入图像先下采样，并加上浅层的特征用双线性插值的方法进行上采样，判别网络则采用Patch GAN加谱归一化的结构，分别计算语义内容损失和风格损失以提高网络的稳定性。本文采用surface表征损失、structure表征损失和texture表征损失代替风格损失，使得生成动漫图像的效果更可控。写实图像选用train2014，人脸图像采用CelebA-HQ数据集。使用本文模型在这些数据集上进行实验，实验结果表明，本文模型能够有效地完成图像动漫化的过程，并生成较高质量的动漫化图像。
|
| 关 键 词： | 深度学习 生成对抗网络 图像动漫化 |
| 收稿时间： | 2022-07-25 |
Image Animation Based on Generative Adversarial Networks |
| |
| Abstract: | Anime-style images are highly simplified and abstract. In order to solve the problem of transforming real-world images into anime-style images, this paper proposes an image animation method based on generative adversarial networks. The generation network in this paper is like a U-Net fully convolutional structure. The input image is down-sampled first, and the shallow features are up-sampled by bilinear interpolation. The discriminant network uses Patch GAN and spectrum normalization. Semantic content loss and style loss are calculated separately to improve the stability of the network. Surface representation loss, structure representation loss, and texture representation loss are used to replace style loss to make the effect of generating animation pictures more controllable. We use train2014 for realistic images, and use the CelebA-HQ data set for face images. Experiments are performed on these data sets using this model. The experimental results show that the model in this paper can effectively complete the process of image animation and generate high-quality animation images. |
| |
| Keywords: | deep learning generative adversarial networks image animation |
|
| 点击此处可从《计算机与现代化》浏览原始摘要信息 |
|
点击此处可从《计算机与现代化》下载全文 |
