论文标题
图像插入图像的经常性特征推理
Recurrent Feature Reasoning for Image Inpainting
论文作者
论文摘要
现有的介入方法已实现有希望的性能，以恢复常规图像缺陷。但是，由于缺乏对孔中心的限制，填充大的连续孔仍然很困难。在本文中，我们设计了一个经常性特征推理（RFR）网络，该网络主要由插件的复发功能推理模块和知识一致的注意（KCA）模块构建。类似于人类如何解决难题（即，首先求解易于零件，然后将结果用作其他信息来解决困难零件），RFR模块将卷积图映射的孔边界复发，然后用它们作为线索以进一步推断。该模块逐渐增强了孔中心的约束，结果变得明确。为了捕获RFR功能地图中遥远位置的信息，我们进一步开发了KCA并将其纳入RFR。从经验上讲，我们首先将所提出的RFR-NET与现有骨架进行比较，表明RFR-NET更有效（例如，相同模型大小的4 \％SSIM改进）。然后，我们将网络放置在当前最新的上下文中，该网络表现出改善的性能。相应的源代码可在以下网址获得：https：//github.com/jingyuanli001/rfr-inpainting
Existing inpainting methods have achieved promising performance for recovering regular or small image defects. However, filling in large continuous holes remains difficult due to the lack of constraints for the hole center. In this paper, we devise a Recurrent Feature Reasoning (RFR) network which is mainly constructed by a plug-and-play Recurrent Feature Reasoning module and a Knowledge Consistent Attention (KCA) module. Analogous to how humans solve puzzles (i.e., first solve the easier parts and then use the results as additional information to solve difficult parts), the RFR module recurrently infers the hole boundaries of the convolutional feature maps and then uses them as clues for further inference. The module progressively strengthens the constraints for the hole center and the results become explicit. To capture information from distant places in the feature map for RFR, we further develop KCA and incorporate it in RFR. Empirically, we first compare the proposed RFR-Net with existing backbones, demonstrating that RFR-Net is more efficient (e.g., a 4\% SSIM improvement for the same model size). We then place the network in the context of the current state-of-the-art, where it exhibits improved performance. The corresponding source code is available at: https://github.com/jingyuanli001/RFR-Inpainting
