mlsanityData to 14 Oct 2026

Flow-Matching Guided Deep Unfolding for Hyperspectral Image Reconstruction

arXiv 2510.01912 ↗First posted 2 Oct 2025cs.CVcited by 0

On the map: Diffusion models → Inverse problems and image restoration

Abstract (arXiv)

Reconstructing a three-dimensional hyperspectral cube from a two-dimensional compressed measurement is a severely ill-posed inverse problem. Existing deep unfolding networks (DUNs) retain fidelity to the imaging model, but regression-trained denoisers can suppress spatial detail and smooth spectral structure under strong modulation. This paper proposes FMU, a deep unfolding framework that couples a measurement-conditioned flow-matching prior with a sensing-model-guided measurement update. A two-phase scheme first learns a compact clean-HSI latent target and then trains a conditional velocity field to generate this target from Gaussian noise. A mean-velocity regularizer additionally penalizes the first-moment error of the learned field. On the KAIST 10-scene benchmark under the optical-filter setting, FMU obtains 42.13,dB PSNR and 0.9900 SSIM, outperforming LADE-DUN by 1.16,dB in PSNR under the same training data, sensing mask, and evaluation protocol. Under the same optical-filter operator, FMU is further evaluated on held-out KAIST and ICVL scenes without fine-tuning. We also report quantitative simulated-CASSI results and qualitative reconstructions of real CASSI measurements.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models. Builds on 28 of its 76 references that are in the map. First posted on arXiv 2 Oct 2025.

Builds on

28 of 76 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Latent Diffusion Prior Enhanced Deep Unfolding for Snapshot Spectral Compressive Imaging24 Nov

Other references · 27

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  5. 2022Flow Matching for Generative Modelingcited by 5,702
  6. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,700
  7. 2019Searching for MobileNetV3cited by 1,994
  8. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,446
  9. 2021Pay Attention to MLPscited by 368
  10. 2016Plug-and-Play ADMM for Image Restoration: Fixed Point Convergence and Applicationscited by 246
  11. 2023Flow Matching in Latent Spacecited by 160
  12. 2019Deep Learning for Hyperspectral Image Classification: An Overviewcited by 123
15 more
  1. 2021Mask-guided Spectral-wise Transformer for Efficient Hyperspectral Image Reconstructioncited by 110
  2. 2018Rank Minimization for Snapshot Compressive Imagingcited by 99
  3. 2024Optimal Flow Matching: Learning Straight Trajectories in Just One Stepcited by 78
  4. 2022MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstructioncited by 72
  5. 2022Coarse-to-Fine Sparse Transformer for Hyperspectral Image Reconstructioncited by 67
  6. 2022HDNet: High-resolution Dual-domain Learning for Spectral Compressive Imagingcited by 66
  7. 2022Degradation-Aware Unfolding Half-Shuffle Transformer for Spectral Compressive Imagingcited by 66
  8. 2021Deep Gaussian Scale Mixture Prior for Spectral Compressive Imagingcited by 58
  9. 2020GAP-net for Snapshot Compressive Imagingcited by 54
  10. 2020Spectral DiffuserCam: lensless snapshot hyperspectral imaging with a spectral filter arraycited by 24
  11. 2023Binarized Spectral Compressive Imagingcited by 23
  12. 2022Residual Degradation Learning Unfolding Framework with Mixing Priors across Spectral and Spatial for Compressive Spectral Imagingcited by 22
  13. 2023Pixel Adaptive Deep Unfolding Transformer for Hyperspectral Image Reconstructioncited by 20
  14. 2017Aerial Vehicle Tracking by Adaptive Fusion of Hyperspectral Likelihood Mapscited by 13
  15. 2020Dynamic brain spectrum acquired by a real-time ultra-spectral imaging chip with reconfigurable metasurfacescited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Progressive Flow-inspired Unfolding for Spectral Compressive Imaging
  2. 2025Detail Matters: Mamba-Inspired Joint Unfolding Network for Snapshot Spectral Compressive Imaging
  3. 2024Dual Hyperspectral Mamba for Efficient Spectral Compressive Imaging
  4. 2024HIR-Diff: Unsupervised Hyperspectral Image Restoration Via Improved Diffusion Models
  5. 2025Diffusion model for gradient preconditioning in hyperspectral imaging inverse problems
  6. 2025Learning Spectral Diffusion Prior for Hyperspectral Image Reconstruction
  7. 2026Beyond Reconstruction: Reconstruction-to-Vector Diffusion for Hyperspectral Anomaly Detection
  8. 2025Using Powerful Prior Knowledge of Diffusion Model in Deep Unfolding Networks for Image Compressive Sensing