Deep Image Harmonization with Globally Guided Feature Transformation and Relation Distillation

TL;DR

This paper introduces a novel deep image harmonization method that leverages global guidance and relation distillation to improve foreground-background consistency, supported by a new dataset and extensive experiments.

Contribution

It proposes a globally guided feature transformation and relation distillation approach, along with a new ccHarmony dataset for better training and evaluation.

Findings

Significant performance improvement over previous methods

Effective transfer of foreground-background relations from real to composite images

The ccHarmony dataset enhances training and evaluation

Abstract

Given a composite image, image harmonization aims to adjust the foreground illumination to be consistent with background. Previous methods have explored transforming foreground features to achieve competitive performance. In this work, we show that using global information to guide foreground feature transformation could achieve significant improvement. Besides, we propose to transfer the foreground-background relation from real images to composite images, which can provide intermediate supervision for the transformed encoder features. Additionally, considering the drawbacks of existing harmonization datasets, we also contribute a ccHarmony dataset which simulates the natural illumination variation. Extensive experiments on iHarmony4 and our contributed dataset demonstrate the superiority of our method. Our ccHarmony dataset is released at https://github.com/bcmi/Image-Harmonization-Dataset-ccHarmony.