HyperVision is the first channel-adaptive pre-trained backbone for ground-based hyperspectral vision, trained on 15k images via pseudo-labeling and cross-modal distillation, achieving SOTA on segmentation, tracking, and detection with head-only fine-tuning.
Deep Residual Learning for Image Recognition
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
HyperVision: A Channel-Adaptive Ground-Based Hyperspectral Vision Pre-trained Backbone
HyperVision is the first channel-adaptive pre-trained backbone for ground-based hyperspectral vision, trained on 15k images via pseudo-labeling and cross-modal distillation, achieving SOTA on segmentation, tracking, and detection with head-only fine-tuning.