AACL combines uniform-strength random augmentation with adaptive CutMix to improve semi-supervised segmentation of remote sensing images, reporting up to 2% mIoU gains over WSCL.
IEEE Transactions on Geoscience and Remote Sensing (2023)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptively Augmented Consistency Learning: A Semi-supervised Segmentation Framework for Remote Sensing
AACL combines uniform-strength random augmentation with adaptive CutMix to improve semi-supervised segmentation of remote sensing images, reporting up to 2% mIoU gains over WSCL.