Deakin University
Browse
ma-generativeimageinpaintingwith-2019.pdf (2.35 MB)

Generative image inpainting with submanifold alignment

Download (2.35 MB)
Version 2 2024-06-06, 10:41
Version 1 2020-06-16, 15:07
conference contribution
posted on 2024-06-06, 10:41 authored by Ang Li, Jianzhong Qi, Rui Zhang, Xingjun Ma, Kotagiri Ramamohanarao
Image inpainting aims at restoring missing regions of corrupted images, which has many applications such as image restoration and object removal. However, current GAN-based generative inpainting models do not explicitly exploit the structural or textural consistency between restored contents and their surrounding contexts. To address this limitation, we propose to enforce the alignment (or closeness) between the local data submanifolds (subspaces) around restored images and those around the original (uncorrupted) images during the learning process of GAN-based inpainting models. We exploit Local Intrinsic Dimensionality (LID) to measure, in deep feature space, the alignment between data submanifolds learned by a GAN model and those of the original data, from a perspective of both images (denoted as iLID) and local patches (denoted as pLID) of images. We then apply iLID and pLID as regularizations for GAN-based inpainting models to encourage two different levels of submanifold alignments: 1) an image-level alignment to improve structural consistency, and 2) a patch-level alignment to improve textural details. Experimental results on four benchmark datasets show that our proposed model can generate more accurate results than state-of-the-art models.

History

Pagination

811-817

Location

Macao, China

Start date

2019-08-10

End date

2019-08-16

ISBN-13

9780999241141

Language

eng

Publication classification

E1.1 Full written paper - refereed

Editor/Contributor(s)

Unknown

Title of proceedings

IJCAI 2019 : Proceedings of the 28th International Joint Conference on Artificial Intelligence

Event

Artificial Intelligence. International Joint Conference (28th : 2019 : Macao, China)

Publisher

International Joint Conferences on Artificial Intelligence Organization

Place of publication

[Macao, China]

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC