SpectralMamba: Efficient Mamba for Hyperspectral Image Classification

TL;DR

SpectralMamba introduces a novel state space model for hyperspectral image classification that significantly improves efficiency and performance by modeling data dynamics at multiple levels and employing a piece-wise scanning mechanism.

Contribution

It proposes SpectralMamba, a new deep learning framework that models hyperspectral data efficiently without reliance on attention or recurrence, enhancing scalability and accuracy.

Findings

Achieves competitive classification accuracy on four benchmark datasets.

Reduces computational complexity compared to traditional RNNs and Transformers.

Demonstrates superior efficiency and performance balance in hyperspectral imaging tasks.

Abstract

Recurrent neural networks and Transformers have recently dominated most applications in hyperspectral (HS) imaging, owing to their capability to capture long-range dependencies from spectrum sequences. However, despite the success of these sequential architectures, the non-ignorable inefficiency caused by either difficulty in parallelization or computationally prohibitive attention still hinders their practicality, especially for large-scale observation in remote sensing scenarios. To address this issue, we herein propose SpectralMamba -- a novel state space model incorporated efficient deep learning framework for HS image classification. SpectralMamba features the simplified but adequate modeling of HS data dynamics at two levels. First, in spatial-spectral space, a dynamical mask is learned by efficient convolutions to simultaneously encode spatial regularity and spectral peculiarity, thus attenuating the spectral variability and confusion in discriminative representation learning. Second, the merged spectrum can then be efficiently operated in the hidden state space with all parameters learned input-dependent, yielding selectively focused responses without reliance on redundant attention or imparallelizable recurrence. To explore the room for further computational downsizing, a piece-wise scanning mechanism is employed in-between, transferring approximately continuous spectrum into sequences with squeezed length while maintaining short- and long-term contextual profiles among hundreds of bands. Through extensive experiments on four benchmark HS datasets acquired by satellite-, aircraft-, and UAV-borne imagers, SpectralMamba surprisingly creates promising win-wins from both performance and efficiency perspectives.