mlsanityData to 7 Oct 2026

Dynamical Hyperspectral Unmixing with Variational Recurrent Neural Networks

arXiv 2303.10566 ↗First posted 19 Mar 2023eess.IV cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Multitemporal hyperspectral unmixing (MTHU) is a fundamental tool in the analysis of hyperspectral image sequences. It reveals the dynamical evolution of the materials (endmembers) and of their proportions (abundances) in a given scene. However, adequately accounting for the spatial and temporal variability of the endmembers in MTHU is challenging, and has not been fully addressed so far in unsupervised frameworks. In this work, we propose an unsupervised MTHU algorithm based on variational recurrent neural networks. First, a stochastic model is proposed to represent both the dynamical evolution of the endmembers and their abundances, as well as the mixing process. Moreover, a new model based on a low-dimensional parametrization is used to represent spatial and temporal endmember variability, significantly reducing the amount of variables to be estimated. We propose to formulate MTHU as a Bayesian inference problem. However, the solution to this problem does not have an analytical solution due to the nonlinearity and non-Gaussianity of the model. Thus, we propose a solution based on deep variational inference, in which the posterior distribution of the estimated abundances and endmembers is represented by using a combination of recurrent neural networks and a physically motivated model. The parameters of the model are learned using stochastic backpropagation. Experimental results show that the proposed method outperforms state of the art MTHU algorithms.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 31 of its 92 references that are in the map. First posted on arXiv 19 Mar 2023.

Builds on

31 of 92 references.

References · 31

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2013Auto-Encoding Variational Bayescited by 3,980
  3. 2016Tutorial on Variational Autoencoderscited by 404
  4. 2015A Critical Review of Recurrent Neural Networks for Sequence Learningcited by 223
  5. 2016Structured Inference Networks for Nonlinear State Space Modelscited by 182
  6. 2017Autoencoding Variational Inference For Topic Modelscited by 168
  7. 2018An Augmented Linear Mixing Model to Address Spectral Variability for Hyperspectral Unmixingcited by 65
  8. 2020Spectral Variability in Hyperspectral Data Unmixing: A Comprehensive Reviewcited by 45
  9. 2021Endmember-Guided Unmixing Network (EGU-Net): A General Deep Learning Framework for Self-Supervised Hyperspectral Unmixingcited by 29
  10. 2015Hyperspectral unmixing with spectral variability using a perturbed linear mixing modelcited by 17
  11. 2019Deep Generative Endmember Modeling: An Application to Unsupervised Spectral Unmixingcited by 16
  12. 2019Hyperspectral Unmixing via Deep Autoencoder Networks for a Generalized Linear-Mixture/Nonlinear-Fluctuation Modelcited by 16
19 more
  1. 2018Hyperspectral Image Unmixing with Endmember Bundles and Group Sparsity Inducing Mixed Normscited by 14
  2. 2017Tech Report: A Fast Multiscale Spatial Regularization for Sparse Hyperspectral Unmixingcited by 11
  3. 2015Dynamical spectral unmixing of multitemporal hyperspectral imagescited by 9
  4. 2018Super-Resolution for Hyperspectral and Multispectral Image Fusion Accounting for Seasonal Spectral Variabilitycited by 9
  5. 2020Kalman Filtering and Expectation Maximization for Multitemporal Spectral Unmixingcited by 8
  6. 2014Unsupervised Unmixing of Hyperspectral Images Accounting for Endmember Variabilitycited by 7
  7. 2017Generalized linear mixing model accounting for endmember variabilitycited by 7
  8. 2018A Data Dependent Multiscale Model for Hyperspectral Unmixing With Spectral Variabilitycited by 7
  9. 2015Online Unmixing of Multitemporal Hyperspectral Images accounting for Spectral Variabilitycited by 6
  10. 2017A Gaussian mixture model representation of endmember variability in hyperspectral unmixingcited by 6
  11. 2019Deep Generative Models for Library Augmentation in Multiple Endmember Spectral Mixture Analysiscited by 6
  12. 2021Model-Based Deep Autoencoder Networks for Nonlinear Hyperspectral Unmixingcited by 6
  13. 2018Hyperspectral unmixing with spectral variability using adaptive bundles and double sparsitycited by 5
  14. 2018Low-Rank Tensor Modeling for Hyperspectral Unmixing Accounting for Spectral Variabilitycited by 5
  15. 2022Hybrid Neural Network Augmented Physics-based Models for Nonlinear Filteringcited by 5
  16. 2019Learning Endmember Dynamics in Multitemporal Hyperspectral Data Using a State-Space Model Formulationcited by 4
  17. 2020Illumination invariant hyperspectral image unmixing based on a digital surface modelcited by 4
  18. 2021Fast Unmixing and Change Detection in Multitemporal Hyperspectral Datacited by 4
  19. 2016A Hierarchical Bayesian Model Accounting for Endmember Variability and Abrupt Spectral Changes to Unmix Multitemporal Hyperspectral Imagescited by 3

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 1; 2025: 5.

8 papers.

  1. 2023Neural Koopman prior for data assimilationcited by 9
  2. 2024Transformer for Multitemporal Hyperspectral Image Unmixingcited by 4
  3. 2023Learning Interpretable Deep Disentangled Neural Networks for Hyperspectral Unmixingcited by 2
  4. 2025Multitemporal Latent Dynamical Framework for Hyperspectral Images Unmixingcited by 1
  5. 2025A Tractable Two-Step Linear Mixing Model Solved with Second-Order Optimization for Spectral Unmixing under Variabilitycited by 0
  6. 2025dCMF: Learning interpretable evolving patterns from temporal multiway datacited by 0
  7. 2025SMILE: A Super-resolution Guided Multi-task Learning Method for Hyperspectral Unmixingcited by 0
  8. 2025A Copula-Guided Temporal Dependency Method for Multitemporal Hyperspectral Images Unmixingcited by 0