Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Hyperspectral Recovery from RGB Images using Gaussian Processes
Naveed Akhtar, Ajmal Saeed Mian
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Gaussian Processes instead of plain dictionary learning bases (ablation) | Proposed (DL variant) · CAVE Feathers | RMSE: 8.11 -> 7.89 | Improved |
| use Gaussian Processes instead of plain dictionary learning bases (ablation) | Proposed (DL variant) · CAVE Peppers | RMSE: 7.18 -> 5.74 | Improved |
| use Gaussian Processes instead of plain dictionary learning bases (ablation) | Proposed (DL variant) · iCVL domain-specific Beach | RMSE: 4.89 -> 4.33 | Improved |
| use Gaussian Processes instead of plain dictionary learning bases (ablation) | Proposed (DL variant) · iCVL domain-specific Objects | SAM: 1.10 -> 0.82 | Improved |
| replace Arad's sparse coding with the proposed GP method | Arad · CAVE | average RMSE: 16.35% reduction in RMSE | Improved |
| replace Arad's method with the proposed GP method | Arad · CAVE | average SAM: 9.39% reduction | Improved |
| replace A+ with the proposed GP method | A+ · CAVE | average RMSE: 11.98% reduction | Improved |
| replace Arad's method with the proposed GP method | Arad · iCVL domain-specific | average relative RMSE: 18.62% reduction | Improved |
| replace Arad's method with the proposed GP method | Arad · iCVL domain-specific | average SAM: 23.35% reduction | Improved |
| replace Arad's method with the proposed GP method | Arad · Harvard | average RMSE: 30.75% improvement | Improved |
| train on a different scene domain than the test domain (ablation) | Proposed · iCVL Floral | relative RMSE: 7.38 -> 9.44 | Hurt |
| train on a different scene domain than the test domain (ablation) | Proposed · iCVL Floral | SAM: 4.49 -> 5.28 | Hurt |
Closest papers by meaning
- Semantic-Embedded Unsupervised Spectral Reconstruction From Single RGB Images in the Wild (ICCV 2021 · similarity 0.87)
- Sparse Recovery of Hyperspectral Signal from Natural RGB Images (ECCV 2016 · similarity 0.87)
- PhySpec: Physically Consistent Spectral Reconstruction via Orthogonal Subspace Decomposition and Self-Supervised Meta-Auxiliary Learning (ICML 2025 · similarity 0.86)
- EigenSR: Eigenimage-Bridged Pre-Trained RGB Learners for Single Hyperspectral Image Super-Resolution (AAAI 2025 · similarity 0.85)
- Illumination Spectrum Estimation for Multispectral Images via Surface Reflectance Modeling and Spatial-Spectral Feature Generation (CVPR 2025 · similarity 0.84)
- FRN: Fractal-Based Recursive Spectral Reconstruction Network (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
