论文标题
图像用可学习的功能插补插图
Image Inpainting with Learnable Feature Imputation
论文作者
论文摘要
定期卷积层以相同的方式应用已知和未知区域的滤波器，从而导致贴有图像中的视觉伪像。几项研究以卷积输出的特征重新归一化解决了这一问题。但是，这些模型使用大量可学习的参数进行特征重新归一化，或者假设输出确定性的二进制表示。我们将缺少输入值的（层）提出（层）特征归因于卷积。与学习的特征重新归一化相反，我们的方法是有效的，并且引入了最少的参数。此外，我们提出了对图像插入图像的修订梯度惩罚，而新颖的甘恩建筑仅接受了对抗性损失的训练。我们对FDF数据集的定量评估反映了我们修订的梯度惩罚和替代卷积可显着提高产生的图像质量。我们对Celeba-HQ和Place2进行比较与当前最新的比较，以验证我们的模型。
A regular convolution layer applying a filter in the same way over known and unknown areas causes visual artifacts in the inpainted image. Several studies address this issue with feature re-normalization on the output of the convolution. However, these models use a significant amount of learnable parameters for feature re-normalization, or assume a binary representation of the certainty of an output. We propose (layer-wise) feature imputation of the missing input values to a convolution. In contrast to learned feature re-normalization, our method is efficient and introduces a minimal number of parameters. Furthermore, we propose a revised gradient penalty for image inpainting, and a novel GAN architecture trained exclusively on adversarial loss. Our quantitative evaluation on the FDF dataset reflects that our revised gradient penalty and alternative convolution improves generated image quality significantly. We present comparisons on CelebA-HQ and Places2 to current state-of-the-art to validate our model.
