Image harmonization has been significantly advanced with large-scale harmonization dataset. However, the current way to build dataset is still labor-intensive, which adversely affects the extendability of dataset. To address this problem, we propose to construct rendered harmonization dataset with fewer human efforts to augment the existing real-world dataset. To leverage both real-world images and rendered images, we propose a cross-domain harmonization network to bridge the domain gap between two domains. Moreover, we also employ well-designed style classifiers and losses to facilitate cross-domain knowledge transfer. Extensive experiments demonstrate the potential of using rendered images for image harmonization and the effectiveness of our proposed network.
@article{arxiv.2103.17104,
title = {Deep Image Harmonization by Bridging the Reality Gap},
author = {Junyan Cao and Wenyan Cong and Li Niu and Jianfu Zhang and Liqing Zhang},
journal= {arXiv preprint arXiv:2103.17104},
year = {2022}
}