Get Started
Home
Topics
Search
Library
Research questionHow can semi-supervised hyperspectral classification use scarce labels while preserving boundaries and stabilizing pseudo-labels?Hyperspectral labels are expensive and often sparse. Propagating labels across spatial neighborhoods can cross class boundaries, while changing predictions on unlabeled pixels can create unstable training targets.
Computer Vision
Image & Video Processing
Machine Learning
Statistical Machine Learning
Latest papersRecent research connected to this question, newest first.Semi-Supervised Hyperspectral Image Classification with Edge-Aware Superpixel Label Propagation and Adaptive Pseudo-LabelingThe source presents an integrated framework for boundary-aware spatial propagation and more consistent pseudo-label selection and refinement. Evidence comes from experiments on benchmark hyperspectral datasets.research paper · Sep 4, 2026
Related questions
How can semi-supervised medical image segmentation prevent appearance variation from corrupting structural cues and pseudo-labels?How can we fuse low-resolution hyperspectral and high-resolution multispectral observations without losing spatial or spectral detail?How can we measure label-preserving adversarial vulnerability separately from ordinary misclassification in high-dimensional classifiers?How can 3D occupancy models learn from noisy 2D pseudo-labels without 3D annotations?