G-ZAP: A Generalizable Zero-Shot Framework for Arbitrary-Scale Pansharpening

TL;DR

G-ZAP is a versatile zero-shot framework for arbitrary-scale pansharpening that generalizes well across different resolutions, scenes, and sensors, achieving state-of-the-art results without per-image retraining.

Contribution

It introduces a feature-based implicit neural representation fusion network with a multi-scale semi-supervised training scheme for improved generalization.

Findings

State-of-the-art performance on multiple datasets.

Supports weight reuse across image pairs.

Effective for arbitrary-scale pansharpening.

Abstract

Pansharpening aims to fuse a high-resolution panchromatic (PAN) image and a low-resolution multispectral (LRMS) image to produce a high-resolution multispectral (HRMS) image. Recent deep models have achieved strong performance, yet they typically rely on large-scale pretraining and often generalize poorly to unseen real-world image pairs.Prior zero-shot approaches improve real-scene generalization but require per-image optimization, hindering weight reuse, and the above methods are usually limited to a fixed scale.To address this issue, we propose G-ZAP, a generalizable zero-shot framework for arbitrary-scale pansharpening, designed to handle cross-resolution, cross-scene, and cross-sensor generalization.G-ZAP adopts a feature-based implicit neural representation (INR) fusion network as the backbone and introduces a multi-scale, semi-supervised training scheme to enable robust generalization.Extensive experiments on multiple real-world datasets show that G-ZAP achieves state-of-the-art results under PAN-scale fusion in both visual quality and quantitative metrics.Notably, G-ZAP supports weight reuse across image pairs while maintaining competitiveness with per-pair retraining, demonstrating strong potential for efficient real-world deployment.