Image Inpainting by End-to-End Cascaded Refinement with Mask Awareness

TL;DR

This paper introduces a mask-aware inpainting framework that dynamically adapts to missing regions using specialized filtering and normalization, significantly improving the quality of inpainted images.

Contribution

The novel mask-aware inpainting model employs dynamic filtering and point-wise normalization to better handle arbitrary missing regions in an end-to-end cascaded refinement architecture.

Findings

Outperforms existing methods on Places2, CelebA, and Paris StreetView datasets.

Effectively captures multi-scale features for missing regions.

Produces visually coherent inpainted images with higher quantitative scores.

Abstract

Inpainting arbitrary missing regions is challenging because learning valid features for various masked regions is nontrivial. Though U-shaped encoder-decoder frameworks have been witnessed to be successful, most of them share a common drawback of mask unawareness in feature extraction because all convolution windows (or regions), including those with various shapes of missing pixels, are treated equally and filtered with fixed learned kernels. To this end, we propose our novel mask-aware inpainting solution. Firstly, a Mask-Aware Dynamic Filtering (MADF) module is designed to effectively learn multi-scale features for missing regions in the encoding phase. Specifically, filters for each convolution window are generated from features of the corresponding region of the mask. The second fold of mask awareness is achieved by adopting Point-wise Normalization (PN) in our decoding phase, considering that statistical natures of features at masked points differentiate from those of unmasked points. The proposed PN can tackle this issue by dynamically assigning point-wise scaling factor and bias. Lastly, our model is designed to be an end-to-end cascaded refinement one. Supervision information such as reconstruction loss, perceptual loss and total variation loss is incrementally leveraged to boost the inpainting results from coarse to fine. Effectiveness of the proposed framework is validated both quantitatively and qualitatively via extensive experiments on three public datasets including Places2, CelebA and Paris StreetView.