StructureFlow: Image Inpainting via Structure-aware Appearance Flow
This addresses the challenge of restoring both reasonable structures and fine-grained textures in images, which is important for applications in computer vision and graphics, though it is incremental as it builds on existing deep learning approaches.
The paper tackles the problem of image inpainting by proposing a two-stage model that separates structure reconstruction and texture generation, resulting in superior performance on multiple datasets.
Image inpainting techniques have shown significant improvements by using deep neural networks recently. However, most of them may either fail to reconstruct reasonable structures or restore fine-grained textures. In order to solve this problem, in this paper, we propose a two-stage model which splits the inpainting task into two parts: structure reconstruction and texture generation. In the first stage, edge-preserved smooth images are employed to train a structure reconstructor which completes the missing structures of the inputs. In the second stage, based on the reconstructed structures, a texture generator using appearance flow is designed to yield image details. Experiments on multiple publicly available datasets show the superior performance of the proposed network.