SPG-Net: Segmentation Prediction and Guidance Network for Image Inpainting

TL;DR

SPG-Net introduces a novel approach for image inpainting by incorporating semantic segmentation to improve boundary clarity and texture consistency, outperforming existing methods and enabling multi-modal predictions.

Contribution

The paper proposes a two-step segmentation-guided inpainting framework that leverages semantic segmentation to enhance boundary sharpness and texture quality in image inpainting.

Findings

Outperforms existing inpainting methods in quality metrics.

Provides clearer boundaries between different semantic regions.

Enables multi-modal inpainting predictions through interactive segmentation guidance.

Abstract

In this paper, we focus on image inpainting task, aiming at recovering the missing area of an incomplete image given the context information. Recent development in deep generative models enables an efficient end-to-end framework for image synthesis and inpainting tasks, but existing methods based on generative models don't exploit the segmentation information to constrain the object shapes, which usually lead to blurry results on the boundary. To tackle this problem, we propose to introduce the semantic segmentation information, which disentangles the inter-class difference and intra-class variation for image inpainting. This leads to much clearer recovered boundary between semantically different regions and better texture within semantically consistent segments. Our model factorizes the image inpainting process into segmentation prediction (SP-Net) and segmentation guidance (SG-Net) as two steps, which predict the segmentation labels in the missing area first, and then generate segmentation guided inpainting results. Experiments on multiple public datasets show that our approach outperforms existing methods in optimizing the image inpainting quality, and the interactive segmentation guidance provides possibilities for multi-modal predictions of image inpainting.