Pan-Sharpening with Color-Aware Perceptual Loss and Guided Re-Colorization

TL;DR

This paper introduces a novel color-aware perceptual loss and guided re-colorization technique for pan-sharpening, improving image quality by focusing on structural details and realistic colors, outperforming existing methods.

Contribution

The paper proposes a new CAP loss that emphasizes spatial details over color and a guided re-colorization method for better color fidelity in pan-sharpened images.

Findings

Outperforms state-of-the-art on WorldView3 dataset

Produces images with fewer artifacts and more natural appearance

Enhances colors during training and testing using self-supervision

Abstract

We present a novel color-aware perceptual (CAP) loss for learning the task of pan-sharpening. Our CAP loss is designed to focus on the deep features of a pre-trained VGG network that are more sensitive to spatial details and ignore color information to allow the network to extract the structural information from the PAN image while keeping the color from the lower resolution MS image. Additionally, we propose "guided re-colorization", which generates a pan-sharpened image with real colors from the MS input by "picking" the closest MS pixel color for each pan-sharpened pixel, as a human operator would do in manual colorization. Such a re-colorized (RC) image is completely aligned with the pan-sharpened (PS) network output and can be used as a self-supervision signal during training, or to enhance the colors in the PS image during test. We present several experiments where our network trained with our CAP loss generates naturally looking pan-sharpened images with fewer artifacts and outperforms the state-of-the-arts on the WorldView3 dataset in terms of ERGAS, SCC, and QNR metrics.