Flow-based Gaussian Splatting for Continuous-Scale Remote Sensing Image Super-Resolution

TL;DR

FlowGS is a novel framework for continuous-scale remote sensing image super-resolution that improves inference efficiency and flexibility using flow matching and Gaussian splatting techniques.

Contribution

The paper introduces FlowGS, a new generative reconstruction method that enables arbitrary-scale super-resolution of remote sensing images with enhanced efficiency and flexibility.

Findings

FlowGS achieves competitive perceptual quality in SR tasks.

FlowGS significantly improves inference efficiency over existing diffusion-based methods.

FlowGS enables flexible reconstruction at arbitrary query locations.

Abstract

High-resolution remote sensing images (RSIs) are crucial for Earth observation applications, yet acquiring them is often limited by sensor constraints and costs. In recent years, generative super-resolution (SR) methods, particularly diffusion models, have made significant progress. However, they typically require slow iterative inference with 40--1000 steps and exhibit limited flexibility in continuous-scale SR settings. To address these issues, we propose FlowGS, a generative reconstruction framework for arbitrary-scale SR of RSIs. FlowGS models the high-frequency detail representations between high- and low-resolution images and learns a continuous probability flow from noise to detail priors via flow matching (FM) constrained by shortcut consistency, thereby reducing generative complexity and improving inference efficiency. Additionally, we employ 2D Gaussian splatting to construct a continuous feature field, thereby enabling flexible reconstruction at arbitrary query locations. Experimental results show that FlowGS delivers competitive perceptual quality compared with existing methods in both continuous-scale and fixed-scale SR settings, with substantially improved inference efficiency.