Diverse Image Harmonization
This work addresses the need for diverse image harmonization in computer vision, offering a novel approach to handle multiple plausible reflectances, though it is incremental in building upon existing harmonization methods.
The paper tackles the problem of generating multiple plausible harmonization results for composite images, which existing methods cannot do, by proposing a reflectance-guided harmonization network and a diverse reflectance generation network, achieving better performance as demonstrated in experiments on benchmark datasets.
Image harmonization aims to adjust the foreground illumination in a composite image to make it harmonious. The existing harmonization methods can only produce one deterministic result for a composite image, ignoring that a composite image could have multiple plausible harmonization results due to multiple plausible reflectances. In this work, we first propose a reflectance-guided harmonization network, which can achieve better performance with the guidance of ground-truth foreground reflectance. Then, we also design a diverse reflectance generation network to predict multiple plausible foreground reflectances, leading to multiple plausible harmonization results. The extensive experiments on the benchmark datasets demonstrate the effectiveness of our method.