Arbitrary‑Scale Spatial–Spectral Fusion using Kernel Integral and Progressive Resampling

doi:10.1016/j.inffus.2026.104143

Arbitrary‑Scale Spatial–Spectral Fusion using Kernel Integral and Progressive Resampling

2026-01-12

0

PRE

AI

DOI：10.1016/j.inffus.2026.104143

原文链接

原文求助

摘要

En 中文

• SFNO: high-performance spatial-spectral fusion with 0.32M parameters, 61.20G FLOPs. • SFNO learns the mapping between function spaces, obviating the need to retrain for different scales. • Efficient Galerkin and resampling integration for SSF multiscale information. • ADE enhances spectral fusion and enables efficient dual-input processing for SSF. • Experiments on CAVE, Chikusei, Pavia Centre, Harvard and Real-world dataset outperform state-of-the-art models.

Arbitrary‑Scale Spatial–Spectral Fusion using Kernel Integral and Progressive Resampling

摘要

期刊

机构

引用论文