Enpowering Your Pansharpening Models with Generalizability: Unified Distribution is All You Need

TL;DR

This paper introduces UniPAN, a unified distribution strategy that normalizes data from various sources to improve the generalizability of deep learning models for remote sensing pansharpening, enabling consistent performance across different satellite sensors.

Contribution

The paper proposes a novel distribution transformation approach that enhances the generalizability of pansharpening models by normalizing data to a unified distribution during training and testing.

Findings

Significant performance improvements across diverse datasets

Enhanced model robustness to sensor-specific variations

Validation of the unified distribution approach's effectiveness

Abstract

Existing deep learning-based models for remote sensing pansharpening exhibit exceptional performance on training datasets. However, due to sensor-specific characteristics and varying imaging conditions, these models suffer from substantial performance degradation when applied to unseen satellite data, lacking generalizability and thus limiting their applicability. We argue that the performance drops stem primarily from distributional discrepancies from different sources and the key to addressing this challenge lies in bridging the gap between training and testing distributions. To validate the idea and further achieve a "train once, deploy forever" capability, this paper introduces a novel and intuitive approach to enpower any pansharpening models with generalizability by employing a unified distribution strategy (UniPAN). Specifically, we construct a distribution transformation function that normalizes the pixels sampled from different sources to conform to an identical distribution. The deep models are trained on the transformed domain, and during testing on new datasets, the new data are also transformed to match the training distribution. UniPAN aims to train and test the model on a unified and consistent distribution, thereby enhancing its generalizability. Extensive experiments validate the efficacy of UniPAN, demonstrating its potential to significantly enhance the performance of deep pansharpening models across diverse satellite sensors. Codes: https://github.com/yc-cui/UniPAN.