论文标题
深层肖像令人愉悦
Deep Portrait Delighting
论文作者
论文摘要
我们提出了一个深层的神经网络，用于从不受约束的肖像图像中删除不良阴影特征，从而恢复基础纹理。我们的培训计划纳入了三种正则化策略：蒙面损失，以强调高频阴影特征；软阴影损失，可以提高对照明微妙变化的敏感性；和阴影偏移估计，以监督阴影和纹理的分离。与最先进的方法相比，我们的方法表明了质量和概括的提高。我们进一步展示了我们的喜悦方法如何增强光敏的计算机视觉任务任务的性能，例如面部重新启动和语义解析，从而使它们能够处理极端的照明条件。
We present a deep neural network for removing undesirable shading features from an unconstrained portrait image, recovering the underlying texture. Our training scheme incorporates three regularization strategies: masked loss, to emphasize high-frequency shading features; soft-shadow loss, which improves sensitivity to subtle changes in lighting; and shading-offset estimation, to supervise separation of shading and texture. Our method demonstrates improved delighting quality and generalization when compared with the state-of-the-art. We further demonstrate how our delighting method can enhance the performance of light-sensitive computer vision tasks such as face relighting and semantic parsing, allowing them to handle extreme lighting conditions.
