摘要 : Semantic image synthesis methods learn to generate new images conditioned on predefined semantic label maps. Existing methods require access to large-volume samples labeled with semantic maps, which limits their applications. We p... 展开 Semantic image synthesis methods learn to generate new images conditioned on predefined semantic label maps. Existing methods require access to large-volume samples labeled with semantic maps, which limits their applications. We propose USIS, a Unified Semantic Image Synthesis model which can be trained on only a single or multiple pairs of images and semantic maps. Once trained, a USIS model can generate new images according to unseen semantic maps, as existing semantic image synthesis methods do. Specifically, we design a hierarchical architecture to reconstruct training samples and grad-ually learn the distributions of multi-scale patches in samples from coarse to fine. To avoid the error accu-mulation across scales, we propose a mixed training strategy to stabilize the training process. Extensive experiments on one-or multiple-sample datasets show our proposed model achieves state-of-the-art performance in terms of visual fidelity.(c) 2022 Elsevier B.V. All rights reserved. 收起