CoMoGAN is a continuous GAN relying on the unsupervised reorganization of the target data on a functional manifold. To that matter, we introduce a new Functional Instance Normalization layer and residual mechanism, which together disentangle image content from position on target manifold. We rely on naive physics-inspired models to guide the training while allowing private model/translations features. CoMoGAN can be used with any GAN backbone and allows new types of image translation, such as cyclic image translation like timelapse generation, or detached linear translation. On all datasets and metrics, it outperforms the literature. Our code is available at http://github.com/cv-rits/CoMoGAN .

翻译：COMOGAN是一个连续的GAN,依靠对功能方方面面的目标数据进行不受监督的重组;为此,我们引入了新的功能性机构正常化层和留守机制,将图像内容与目标方位分离开来;我们依靠天真的物理学启发型模型来指导培训,同时允许私人模式/翻译特征;CoMOGAN可以与任何GAN骨干一起使用,允许新的图像翻译类型,例如像时间折叠生成那样的周期性图像翻译,或独立的线性翻译;关于所有数据集和指标,它优于文献。我们的代码可在http://github.com/cv-rits/CoMoGAN上查阅。