Research map / Label-efficient and robust learning / Semi-supervised learning
Unraveling Metameric Dilemma for Spectral Reconstruction: A High-Fidelity Approach via Semi-Supervised Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes Diff-Spectra, a semi-supervised spectral reconstruction method that combines supervised physics-aware spectral estimation with unsupervised high-fidelity spectral regularization to address the metameric dilemma in RGB-to-hyperspectral reconstruction.
Problem
Spectral reconstruction from RGB images suffers from metamerism: distinct spectral distributions map to nearly identical RGB values, so current models cannot distinguish them and produce unreliable reconstructions.
Approach
Integrates supervised physics-aware spectral estimation with unsupervised high-fidelity spectral regularization (diffusion-based, per the name Diff-Spectra) in a semi-supervised framework, so unlabeled RGB images can constrain the spectral solution beyond paired data.
Closest papers by meaning
- EigenSR: Eigenimage-Bridged Pre-Trained RGB Learners for Single Hyperspectral Image Super-Resolution (AAAI 2025 · similarity 0.88)
- FRN: Fractal-Based Recursive Spectral Reconstruction Network (NeurIPS 2025 · similarity 0.88)
- Unmixing Diffusion for Self-Supervised Hyperspectral Image Denoising (CVPR 2024 · similarity 0.86)
- Spectral Compressive Imaging via Chromaticity-Intensity Decomposition (NeurIPS 2025 · similarity 0.86)
- Sp3ctralMamba: Physics-Driven Joint State Space Model for Hyperspectral Image Reconstruction (AAAI 2025 · similarity 0.86)
- Hyperspectral Image Reconstruction via Combinatorial Embedding of Cross-Channel Spatio-Spectral Clues (AAAI 2024 · similarity 0.86)
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.