A Deep Equilibrium Network for Hyperspectral Unmixing

arXiv:2604.11279v1 Announce Type: new Abstract: Hyperspectral unmixing (HU) is crucial for analyzing hyperspectral imagery, yet achieving accurate unmixing remains challenging. While traditional methods struggle to effectively model complex spectral-spatial features, deep learning approaches often lack physical interpretability. Unrolling-based methods, despite offering network interpretability, inadequately exploit spectral-spatial information and incur high memory costs and numerical precision issues during backpropagation. To address these limitations, we propose DEQ-Unmix, which reformulates abundance estimation as a deep equilibrium model, enabling efficient constant-memory training via implicit differentiation. It replaces the gradient operator of the data reconstruction term with a trainable convolutional network to capture spectral-spatial information. By leveraging implicit differentiation, DEQ-Unmix enables efficient and constant-memory backpropagation. Experiments on synthetic and two real-world datasets demonstrate that DEQ-Unmix achieves superior unmixing performance while maintaining constant memory cost.

Related

• RS-OVC: Open-Vocabulary Counting for Remote-Sensing Data
• Automatic Uncertainty-Aware Synthetic Data Bootstrapping for Historical Map Segmentation
• Efficient Transceiver Design for Aerial Image Transmission and Large-scale Scene Reconstruction

Source: arXiv cs.CV | 2026-04-14