English

Physics-Guided Regime Unmixing

Computer Vision and Pattern Recognition 2026-05-07 v1

Abstract

The Linear Mixing Model (LMM) dominates spectral unmixing for its simplicity, but fails under multiple scattering; existing nonlinear models compensate by applying a fixed regime uniformly across entire scenes. We propose Physics-Guided Regime Unmixing (PGRU), which estimates a pixel-wise scalar ξi[0,1]\xi_i \in [0,1] from observable physical features to activate nonlinear mixing only where justified. Residuals from the Generalized Bilinear Model (GBM), the Post-Nonlinear Mixing Model (PPNM), and Hapke are combined via learned attention, yielding interpretable regime maps. Experiments on Samson, Jasper Ridge, and Urban show consistent improvements over baselines, with physical coherence ρ>0.90\rho > 0.90.

Keywords

Cite

@article{arxiv.2605.04247,
  title  = {Physics-Guided Regime Unmixing},
  author = {Paula Pacheco and Pablo Granitto and Juan B. Cabral},
  journal= {arXiv preprint arXiv:2605.04247},
  year   = {2026}
}
R2 v1 2026-07-01T12:51:46.542Z