论文标题
基于模型的遮挡散布图像到图像翻译
Model-based occlusion disentanglement for image-to-image translation
论文作者
论文摘要
图像到图像的翻译受纠缠现象的影响，纠缠现象可能发生在目标数据中，包括雨滴，污垢等。我们的无监督的基于模型的学习解散场景和闭塞，同时受益于对抗性的移液，从而使对遮挡模型的物理物理参数回归。该实验证明我们的方法能够处理各种类型的闭塞并生成高度逼真的翻译，从定性和定量上胜过多个数据集上的最先进。
Image-to-image translation is affected by entanglement phenomena, which may occur in case of target data encompassing occlusions such as raindrops, dirt, etc. Our unsupervised model-based learning disentangles scene and occlusions, while benefiting from an adversarial pipeline to regress physical parameters of the occlusion model. The experiments demonstrate our method is able to handle varying types of occlusions and generate highly realistic translations, qualitatively and quantitatively outperforming the state-of-the-art on multiple datasets.
