Existing state-of-the-art techniques in exemplar-based image-to-image
translation hold several critical concerns. Existing methods related to
exemplar-based image-to-image translation are impossible to translate on an
image tuple input (source, target) that is not aligned. Additionally, we can
confirm that the existing method exhibits limited generalization ability to
unseen images. In order to overcome this limitation, we propose Multiple GAN
Inversion for Exemplar-based Image-to-Image Translation. Our novel Multiple GAN
Inversion avoids human intervention by using a self-deciding algorithm to
choose the number of layers using Fr\'echet Inception Distance(FID), which
selects more plausible image reconstruction results among multiple hypotheses
without any training or supervision. Experimental results have in fact, shown
the advantage of the proposed method compared to existing state-of-the-art
exemplar-based image-to-image translation methods.

本研究提出了一种名为多重 GAN 反演的方法，以解决基于样本的图像到图像翻译中存在的对未对齐图像元组输入和泛化能力不足的问题，并通过实验验证了其相比现有技术的优势。

基于样例的图像到图像翻译的多 GAN 反演

Multiple GAN Inversion for Exemplar-based Image-to-Image Translation

Recently, unsupervised exemplar-based image-to-image translation, conditioned
on a given exemplar without the paired data, has accomplished substantial
advancements. In order to transfer the information from an exemplar to an input
image, existing methods often use a normalization technique, e.g., adaptive
instance normalization, that controls the channel-wise statistics of an input
activation map at a particular layer, such as the mean and the variance.
Meanwhile, style transfer approaches similar task to image translation by
nature, demonstrated superior performance by using the higher-order statistics
such as covariance among channels in representing a style. In detail, it works
via whitening (given a zero-mean input feature, transforming its covariance
matrix into the identity). followed by coloring (changing the covariance matrix
of the whitened feature to those of the style feature). However, applying this
approach in image translation is computationally intensive and error-prone due
to the expensive time complexity and its non-trivial backpropagation. In
response, this paper proposes an end-to-end approach tailored for image
translation that efficiently approximates this transformation with our novel
regularization methods. We further extend our approach to a group-wise form for
memory and time efficiency as well as image quality. Extensive qualitative and
quantitative experiments demonstrate that our proposed method is fast, both in
training and inference, and highly effective in reflecting the style of an
exemplar. Finally, our code is available at
this https URL

该论文提出了一种针对图像翻译的端对端方法，通过创新的正则化方法有效地逼近样本的风格转换，缩短了运算时间复杂度和误差率，进一步扩展了方法的适用范围，实验证明该方法在训练和推断时速度快，且高效地反映了样本的风格。