Hyperspectral Image Fusion with Spectral-Band and Fusion-Scale Agnosticism

Current deep learning models for Multispectral and Hyperspectral Image Fusion (MS/HS fusion) are typically designed for fixed spectral bands and spatial scales, which limits their transferability across diverse sensors. To address this, we propose SSA, a unified framework for MS/HS fusion with spectral-band and fusion-scale agnosticism. Specifically, we introduce Matryoshka Kernel (MK), a novel operator that enables a single model to process varying ordered spectral band counts within a supported range. Meanwhile, we build SSA upon an Implicit Neural Representation (INR) backbone that models the HS signal as a continuous function, enabling reconstruction at arbitrary spatial resolutions. Together, these two mechanisms support a single MS/HS fusion model for heterogeneous sensors and arbitrary query scales. Extensive experiments demonstrate that our single jointly trained model achieves state-of-the-art performance, generalizes to unseen spatial scales, and transfers to unseen sensors with few-step adaptation. The code can be obtained at https://github.com/vg219/SSA.

Imagine merging a blurry but color-rich photograph with a sharp, standard photograph to get the best of both worlds. Current models can do this, but they are highly rigid: they only work for specific cameras (with an exact number of color bands) and fixed zoom levels (like exactly 2x or 4x). If you switch cameras, you must build and train a completely new model. We introduce a single, "universal" model that breaks these rules. First, we designed a flexible "Russian nesting doll" structure that automatically shrinks or expands to process any camera's specific color bands. Second, we treat images as continuous mathematical surfaces rather than rigid grids of pixels, allowing our model to zoom in to any arbitrary size. This flexibility lets us combine data from many different cameras to train one powerful, unified model. Our single model not only outperforms specialized models but can also instantly adapt to brand-new cameras and zoom levels it has never seen before.