Resolution-robust Large Mask Inpainting with Fourier Convolutions

🤖 Plain-English Summary

Modern image inpainting systems, despite the significant progress, often struggle with large missing areas, complex geometric structures, and high-resolution images. Our model generalizes surprisingly well to resolutions that are higher than those seen at train time, and achieves this at lower parameter&time costs than the competitive baselines.

🔑 Key Findings

• We find that one of the main reasons for that is the lack of an effective receptive field in both the inpainting network and the loss function.
• To alleviate this issue, we propose a new method called large mask inpainting (LaMa).
• LaMa is based on i) a new inpainting network architecture that uses fast Fourier convolutions (FFCs), which have the image-wide receptive field; ii) a high receptive field perceptual loss; iii) large training masks, which unlocks the potential of the first two components.

🔗 Related Resources