Texture Synthesis with Spatial Generative Adversarial Networks
Texture Synthesis with Spatial Generative Adversarial Networks
复制标题
DOI：
--
复制
发表时间：
2016-11
期刊：
影响因子：
--
通讯作者：
Nikolay Jetchev;Urs M. Bergmann;Roland Vollgraf
中科院分区：
文献类型：
--
作者：
Nikolay Jetchev;Urs M. Bergmann;Roland Vollgraf
生成对抗网络（GAN）是一种训练数据生成模型的最新方法，已被证明在图像数据上特别有效。在本文中，我们介绍了一种新的模型，纹理合成的基础上GAN学习。通过将输入噪声分布空间从单个向量扩展到整个空间张量，我们创建了一个具有非常适合纹理合成任务的属性的架构，我们称之为空间GAN（SGAN）。据我们所知，这是第一个成功的基于GAN的完全数据驱动的纹理合成方法。我们的方法具有以下特点，这使它成为一个国家的最先进的纹理合成算法：高图像质量的生成纹理，非常高的可扩展性w.r.t.输出纹理大小，快速实时向前生成，在复杂纹理中融合多个不同源图像的能力。为了说明这些功能，我们提出了不同类别的纹理图像和用例的多个实验。我们还讨论了我们的方法的一些限制，它可以合成的纹理图像的类型，并将其与其他神经技术的纹理生成。
Generative adversarial networks (GANs) are a recent approach to train generative models of data, which have been shown to work particularly well on image data. In the current paper we introduce a new model for texture synthesis based on GAN learning. By extending the input noise distribution space from a single vector to a whole spatial tensor, we create an architecture with properties well suited to the task of texture synthesis, which we call spatial GAN (SGAN). To our knowledge, this is the first successful completely data-driven texture synthesis method based on GANs. Our method has the following features which make it a state of the art algorithm for texture synthesis: high image quality of the generated textures, very high scalability w.r.t. the output texture size, fast real-time forward generation, the ability to fuse multiple diverse source images in complex textures. To illustrate these capabilities we present multiple experiments with different classes of texture images and use cases. We also discuss some limitations of our method with respect to the types of texture images it can synthesize, and compare it to other neural techniques for texture generation.