InGAN: Capturing and Retargeting the “DNA” of a Natural Image
InGAN: Capturing and Retargeting the “DNA” of a Natural Image
复制标题
DOI：
10.1109/iccv.2019.00459
复制
发表时间：
2019-10
期刊：
影响因子：
--
通讯作者：
Assaf Shocher;Shai Bagon;Phillip Isola;M. Irani
中科院分区：
文献类型：
--
作者：
Assaf Shocher;Shai Bagon;Phillip Isola;M. Irani
生成对抗网络（GAN）通常会学习大型图像数据集中的图像分布，然后能够根据该分布生成新图像。然而，每个自然图像都有自己的内部统计数据，通过其独特的补丁分布捕获。在本文中，我们提出了一种“内部 GAN”（InGAN）——一种特定于图像的 GAN——它在单个输入图像上进行训练并学习其补丁的内部分布。然后，它能够合成大量尺寸、形状和纵横比显着不同的新自然图像，所有图像都具有与输入图像相同的内部补丁分布（相同的“DNA”）。特别是，尽管图像的全局尺寸/形状发生很大变化，图像内的所有元素仍保持其局部尺寸/形状。 InGAN 是完全无监督的，除了输入图像本身之外不需要额外的数据。一旦对输入图像进行训练，它就可以在单次前馈传递中将输入重新映射到任何大小或形状，同时保留相同的内部补丁分布。 InGAN 为各种任务提供了统一的框架，弥合了纹理和自然图像之间的差距。
Generative Adversarial Networks (GANs) typically learn a distribution of images in a large image dataset, and are then able to generate new images from this distribution. However, each natural image has its own internal statistics, captured by its unique distribution of patches. In this paper we propose an ``Internal GAN'' (InGAN) -- an image-specific GAN -- which trains on a single input image and learns its internal distribution of patches. It is then able to synthesize a plethora of new natural images of significantly different sizes, shapes and aspect-ratios – all with the same internal patch-distribution (same ``DNA'') as the input image. In particular, despite large changes in global size/shape of the image, all elements inside the image maintain their local size/shape. InGAN is fully unsupervised, requiring no additional data other than the input image itself. Once trained on the input image, it can remap the input to any size or shape in a single feedforward pass, while preserving the same internal patch distribution. InGAN provides a unified framework for a variety of tasks, bridging the gap between textures and natural images.