A new semi-supervised GSC framework achieves over 90% foreground classification accuracy on images while compressing data size by 95% via foreground-aware MAE and SSAE components trained with limited labels.
Semantic communication based on large language model for underwater image transmission
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Semi-Supervised Goal-Oriented Semantic Communication Framework for Foreground Classification
A new semi-supervised GSC framework achieves over 90% foreground classification accuracy on images while compressing data size by 95% via foreground-aware MAE and SSAE components trained with limited labels.