A weakly-supervised endoscopic lesion segmentation method that transfers semantic knowledge via class-balanced pseudo-labels, adversarial alignment, and feature centroids obtains higher mIoU than prior methods on a new dataset and on urban benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Semantic-Transferable Weakly-Supervised Endoscopic Lesions Segmentation
A weakly-supervised endoscopic lesion segmentation method that transfers semantic knowledge via class-balanced pseudo-labels, adversarial alignment, and feature centroids obtains higher mIoU than prior methods on a new dataset and on urban benchmarks.