An Unmixing-derived Spectral and Structural Prompting Mamba framework is proposed, termed USP-Mamba, which adapts Mamba state evolution through composition-aware spectral priors and image-dependent structural prompts to provide image-dependent local guidance.
Abstract
Hyperspectral image super-resolution aims to reconstruct high-resolution imagery while preserving dense spectral information. Recently, Mamba-based models have shown promising potential for this task by capturing long-range dependencies with linear computational complexity. Nevertheless, their causal sequence modeling requires two-dimensional hyperspectral features to be unfolded along predefined scanning orders, which disrupts spatial adjacency and restricts the effective propagation of contextual information. Moreover, state-space parameterization of existing models is predominantly derived from generic learned representations, without explicit alignment with the intrinsic characteristics of the hyperspectral image. To address this issue, we propose an Unmixing-derived Spectral and Structural Prompting Mamba framework, termed USP-Mamba, which adapts Mamba state evolution through composition-aware spectral priors and image-dependent structural prompts. Specifically, an unmixing-informed spectral prompt captures the global material composition of the input image and provides persistent conditioning throughout reconstruction. Injected into the Mamba sequence and progressively adapted across layers, it steers state evolution toward composition-consistent reconstruction. We introduce feature-level structural prompts comprising spatial and frequency components to provide image-dependent local guidance. The spatial prompt promotes structure-sensitive state encoding for local detail preservation, while the frequency prompt enables region-adaptive transitions between homogeneous regions and high-frequency details. Finally, complementary Hilbert and Semantic-Guided Neighboring scans preserve spatial continuity and strengthen non-local semantic dependency modeling. Extensive experiments on different datasets demonstrate that the proposed method consistently outperforms representative approaches.
Recently, state space models (SSMs), particularly the Mamba architecture, have emerged as a powerful paradigm for hyperspectral image (HSI) classification due to their linear complexity and ability to capture long-range dependencies. However, existing Mamba-based approaches mainly rely on predefined geometric scanning...
Jinliang An, Long-Long Dai, Deng-Ru Han et al.· IEEE Transactions on Geoscie...· 0 citations
This work proposes Two-Stage Reconstruction with Implicit Tensor Neural Representation (TSR-ITNR), a unified self-supervised framework integrating representation refinement and observation-guided calibration that demonstrates strong quantitative, visual, and spectral reconstruction performance without ground-truth HR-H...
Li-Qian Yang, Xing-Chi Chen, Xin-Feng Gui et al.· 0 citations
An innovative dual-stream spatial-frequency state-space model (SSM) is presented, termed WF-Mamba, for enhanced HU, and an adaptive synergistic fusion (ASF) mechanism is proposed for WF-Mamba to break the bottleneck of rigid feature concatenation.
Lian-Hui Liang, Jin-Rong Chen, Shao-Quan Zhang et al.· IEEE Transactions on Geoscie...· 0 citations
Hyperspectral and multispectral image fusion (HMIF) aims to generate hyperspectral images (HSIs) with high-spatial resolution and favorable spectral fidelity. Benefiting from efficient and powerful global modeling capacity, Mamba has become a prevailing solution for the HMIF task in recent years. Nevertheless, most exi...
Huanhuan Lv, Jun-Jie Feng, Hui Zhang et al.· IEEE Transactions on Geoscie...· 0 citations
Single hyperspectral image super-resolution (HSI-SR) aims to enhance the spatial resolution of hyperspectral images without modifying imaging hardware. However, existing methods still face challenges in efficiently utilizing high-dimensional spectral information, exploiting intrinsic spectral mixing characteristics, an...