Dilated affinity is jointly predicted with segmentation labels to strengthen features and support efficient label propagation refinement on benchmark datasets.
Group normalization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Improving Semantic Segmentation via Dilated Affinity
Dilated affinity is jointly predicted with segmentation labels to strengthen features and support efficient label propagation refinement on benchmark datasets.