论文标题
图像到图像翻译的层的幂
Powers of layers for image-to-image translation
论文作者
论文摘要
我们提出了一个简单的体系结构，以解决未配对的图像到图像翻译任务：样式或类传输，变形，脱毛，拆卸等。我们从具有固定权重的图像自动编码器体系结构开始。对于每个任务，我们学习一个在潜在空间中运行的残留块，该块被迭代地调用，直到达到目标域。需要特定的培训时间表来减轻迭代的凸起效应。在测试时，它提供了几个优点：重量参数的数量有限，并且组成设计使一个人可以通过迭代次数调节转换的强度。例如，当不知道要抑制的噪声类型或数量时，这很有用。在实验上，我们提供了概念的证据，显示了我们方法对许多转变的兴趣。我们的模型的性能比Cyclegan具有可比性或更好的参数。
We propose a simple architecture to address unpaired image-to-image translation tasks: style or class transfer, denoising, deblurring, deblocking, etc. We start from an image autoencoder architecture with fixed weights. For each task we learn a residual block operating in the latent space, which is iteratively called until the target domain is reached. A specific training schedule is required to alleviate the exponentiation effect of the iterations. At test time, it offers several advantages: the number of weight parameters is limited and the compositional design allows one to modulate the strength of the transformation with the number of iterations. This is useful, for instance, when the type or amount of noise to suppress is not known in advance. Experimentally, we provide proofs of concepts showing the interest of our method for many transformations. The performance of our model is comparable or better than CycleGAN with significantly fewer parameters.
