
SatDiffMoE is a novel diffusion-based algorithm that leverages the temporal information in a sequence of low-resolution satellite images to reconstruct a high-resolution image, outperforming existing methods in perceptual quality and computational efficiency.


coremsg

satdiffmoe-fusing-time-series-low-resolution-satellite-images-for-super-resolution-using-latent-diffusion-models


SatDiffMoE: Fusing Time-Series Low-Resolution Satellite Images for Super-Resolution Using Latent Diffusion Models


title_rewrite


This paper proposes using computationally efficient Consistency Models (CMs) for super-resolution of low-quality satellite images, achieving significant speed improvements and enhanced image quality compared to traditional Denoising Diffusion Probabilistic Models (DDPMs).


improving-satellite-image-resolution-using-computationally-efficient-consistency-models


Improving Satellite Image Resolution Using Computationally Efficient Consistency Models



불규칙적으로 샘플링된 Sentinel-2 시계열 데이터를 활용하여 고해상도 영상을 복원하는 방법을 제안한다.


센서-간-불규칙적으로-샘플링된-sentinel-2-시계열의-초해상도-복원


센서 간 불규칙적으로 샘플링된 Sentinel-2 시계열의 초해상도 복원
