SoDa² is a single-stage framework with decoupled spectral-spatial alignment and GMM-based open-set recognition that outperforms prior methods on cross-scene HSI classification tasks.
Open-set domain adaptation for scene classification using multi-adversarial learning,
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
-
SoDa2: Single-Stage Open-Set Domain Adaptation via Decoupled Alignment for Cross-Scene Hyperspectral Image Classification
SoDa² is a single-stage framework with decoupled spectral-spatial alignment and GMM-based open-set recognition that outperforms prior methods on cross-scene HSI classification tasks.