Two data selection techniques (GMM visual similarity and bounding-box diversity) reduce required weakly labeled images by up to 100x on Open Images and 20x on Cityscapes while maintaining semantic segmentation performance.
The cityscapes dataset for semantic urban scene understanding
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
-
Data Selection for training Semantic Segmentation CNNs with cross-dataset weak supervision
Two data selection techniques (GMM visual similarity and bounding-box diversity) reduce required weakly labeled images by up to 100x on Open Images and 20x on Cityscapes while maintaining semantic segmentation performance.