Unifying conditional and unconditional semantic image synthesis with OCO-GAN

arxiv(2022)

引用 0|浏览44
暂无评分
摘要
Generative image models have been extensively studied in recent years. In the unconditional setting, they model the marginal distribution from unlabelled images. To allow for more control, image synthesis can be conditioned on semantic segmentation maps that instruct the generator the position of objects in the image. While these two tasks are intimately related, they are generally studied in isolation. We propose OCO-GAN, for Optionally COnditioned GAN, which addresses both tasks in a unified manner, with a shared image synthesis network that can be conditioned either on semantic maps or directly on latents. Trained adversarially in an end-to-end approach with a shared discriminator, we are able to leverage the synergy between both tasks. We experiment with Cityscapes, COCO-Stuff, ADE20K datasets in a limited data, semi-supervised and full data regime and obtain excellent performance, improving over existing hybrid models that can generate both with and without conditioning in all settings. Moreover, our results are competitive or better than state-of-the art specialised unconditional and conditional models.
更多
查看译文
关键词
unconditional semantic image synthesis,conditional,oco-gan
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要