Label-Efficient Hyperspectral Image Classification via Spectral FiLM Modulation of Low-Level Pretrained Diffusion Features

TL;DR

This paper introduces a label-efficient hyperspectral image classification method that uses pretrained diffusion model features and a spectral FiLM fusion module to improve performance with sparse labels.

Contribution

It proposes a novel framework combining frozen diffusion features with spectral FiLM modulation for hyperspectral classification, enhancing label efficiency and transferability.

Findings

Outperforms state-of-the-art methods on hyperspectral datasets

Diffusion-derived features improve classification accuracy

Spectral-aware fusion enhances multimodal learning under sparse supervision

Abstract

Hyperspectral imaging (HSI) enables detailed land cover classification, yet low spatial resolution and sparse annotations pose significant challenges. We present a label-efficient framework that leverages spatial features from a frozen diffusion model pretrained on natural images. Our approach extracts low-level representations from high-resolution decoder layers at early denoising timesteps, which transfer effectively to the low-texture structure of HSI. To integrate spectral and spatial information, we introduce a lightweight FiLM-based fusion module that adaptively modulates frozen spatial features using spectral cues, enabling robust multimodal learning under sparse supervision. Experiments on two recent hyperspectral datasets demonstrate that our method outperforms state-of-the-art approaches using only the provided sparse training labels. Ablation studies further highlight the benefits of diffusion-derived features and spectral-aware fusion. Overall, our results indicate that pretrained diffusion models can support domain-agnostic, label-efficient representation learning for remote sensing and broader scientific imaging tasks.