基于SASGAN的戏剧脸谱多样化生成
Diversified generation of theatrical masks based on SASGAN
古天骏 1熊苏雅 2林晓3
作者信息
- 1. 上海师范大学信息与机电工程学院,上海 200234
- 2. 上海交通大学海洋装备研究院,上海 200240
- 3. 上海师范大学信息与机电工程学院,上海 200234;上海师范大学上海智能教育大数据工程技术研究中心,上海 200234;上海市中小学在线教育研究基地,上海 200234
- 折叠
摘要
为解决现有自动生成的戏剧脸谱在分辨率和真实性上效果不佳的问题,提出了基于自注意力机制的风格化生成对抗网络(SASGAN).首先在 StyleGAN 的基础上引入了自注意力机制以及矢量量化方法,增强了对脸谱图案几何结构特征的提取,接着通过多样化差异性增强(DDG)扩充数据,采用脸谱色调辅助算法对 DDG方法进行补充,建立了包含12 599张图像的戏剧脸谱数据集,最后在此数据集上进行训练,生成了兼顾多样性和真实性的脸谱图像.实验结果表明,对于戏剧脸谱图像,DDG 方法较传统方法在数据增广方面有着较大提升,而SASGAN则提升了戏剧脸谱图像的分辨率和真实性,在主观视觉上得到了理想的效果.
Abstract
To address the problem of low resolution and lack of realism in existing automatically generated theatrical masks,a stylized generative adversarial network(SASGAN)based on a self-attentive mechanism was proposed.Firstly,SASGAN introduced the self-attentive mechanism and vector quantization method based on StyleGAN,thereby enhancing the extraction of geometric structure features of mask patterns.Subsequently,the diversified differentiation generation(DDG)method was supplemented with a mask hue-assisted algorithm by expanding the data with DDG to build a theatrical mask dataset containing 12,599 images.The final training was performed on this dataset to generate mask images with both diversity and realism.The experimental results demonstrated significant improvement in data augmentation for theatrical masks using the DDG method compared to the traditional methods,while SASGAN enhanced the resolution and realism of theatrical masks,achieving the desired effect in subjective visualization.
关键词
戏剧脸谱/生成对抗网络/图像生成/注意力机制/矢量量化Key words
theatrical masks/generation adversarial network/image generation/attention mechanism/vector quantization引用本文复制引用
基金项目
国家自然科学基金项目(61502220)
国家自然科学基金项目(61572326)
国家自然科学基金项目(61775139)
国家自然科学基金项目(61872242)
出版年
2024
NETL
NSTL
万方数据