CVOct 20, 2018

Image Inpainting via Generative Multi-column Convolutional Neural Networks

arXiv:1810.08771v1334 citations
Originality Incremental advance
AI Analysis

This work addresses image inpainting for computer vision applications, offering an incremental improvement by combining existing techniques in a novel network architecture.

The paper tackles image inpainting by proposing a generative multi-column convolutional neural network that synthesizes image components in parallel, using a confidence-driven reconstruction loss for global structures and an implicit diversified MRF regularization for local details. The method produces visually compelling results on street view, face, natural objects, and scenes without common post-processing, as shown in extensive experiments.

In this paper, we propose a generative multi-column network for image inpainting. This network synthesizes different image components in a parallel manner within one stage. To better characterize global structures, we design a confidence-driven reconstruction loss while an implicit diversified MRF regularization is adopted to enhance local details. The multi-column network combined with the reconstruction and MRF loss propagates local and global information derived from context to the target inpainting regions. Extensive experiments on challenging street view, face, natural objects and scenes manifest that our method produces visual compelling results even without previously common post-processing.

Code Implementations2 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes