arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1807.04409cs.CV
收藏

Sem-GAN:语义一致的图像到图像翻译

Sem-GAN: Semantically-Consistent Image-to-Image Translation

Mitsubishi Electric Research Labs (MERL)(三菱电机研究实验室(MERL))

Anoop Cherian, Alan Sullivan

摘要:

不成对图像到图像转换是将源域中的图像映射到目标域中的图像的问题,而不需要对应的图像对。为了确保翻译的图像是真实可信的,最近的作品,如Cycle-GAN,要求这种映射是可逆的。虽然这个要求在域是单峰的情况下展示了有希望的结果,但在多模态场景(例如图像分割任务)中,其性能是不可预测的。这是因为,可逆性不一定强制语义正确性。为此,我们提出了一个语义一致的GAN框架,被称为Sem-GAN,其中的语义定义的类身份的图像段在源域中产生的语义分割算法。我们提出的框架包括翻译任务的一致性约束,连同GAN损失和周期约束,强制执行翻译时的图像将继承目标域的外观,同时(近似)保持它们的身份从源域。我们在几个图像到图像的翻译任务上进行了实验,并证明Sem-GAN显着提高了翻译图像的质量,有时在FCN分数上提高了20%以上。此外,我们表明,语义分割模型,通过Sem-GAN翻译的合成图像训练,导致比其他变体更好的分割结果。

英文摘要:

Unpaired image-to-image translation is the problem of mapping an image in the source domain to one in the target domain, without requiring corresponding image pairs. To ensure the translated images are realistically plausible, recent works, such as Cycle-GAN, demands this mapping to be invertible. While, this requirement demonstrates promising results when the domains are unimodal, its performance is unpredictable in a multi-modal scenario such as in an image segmentation task. This is because, invertibility does not necessarily enforce semantic correctness. To this end, we present a semantically-consistent GAN framework, dubbed Sem-GAN, in which the semantics are defined by the class identities of image segments in the source domain as produced by a semantic segmentation algorithm. Our proposed framework includes consistency constraints on the translation task that, together with the GAN loss and the cycle-constraints, enforces that the images when translated will inherit the appearances of the target domain, while (approximately) maintaining their identities from the source domain. We present experiments on several image-to-image translation tasks and demonstrate that Sem-GAN improves the quality of the translated images significantly, sometimes by more than 20% on the FCN score. Further, we show that semantic segmentation models, trained with synthetic images translated via Sem-GAN, leads to significantly better segmentation results than other variants.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑