Hierarchical Detailed Intermediate Supervision for Image-to-Image Translation

The image-to-image translation aims to learn a mapping between the source and target domains. For improving visual quality, the majority of previous works adopt multi-stage techniques to refine coarse results in a progressive manner. In this work, we present a novel approach for generating plausible details by only introducing a group of intermediate supervisions without cascading multiple stages. Specifically, we propose a Laplacian Pyramid Transformation Generative Adversarial Network (Lap-TransGAN) to simultaneously transform components in different frequencies from the source domain to the target domain within only one stage. Hierarchical perceptual and gradient penalization are utilized for learning consistent semantic structures and details at each pyramid level. The pro-posed model is evaluated based on various metrics, including the similarity in feature maps, reconstruction quality, segmentation accuracy, similarity in details, and qualitative appearances. Our experiments show that LapTrans-GAN can achieve a much better quantitative performance than both the supervised pix2pix model and the unsupervised CycleGAN model. Comprehensive ablation experiments are conducted to study the contribution of each component.