UM-Adapt uses cross-task consistency and contour preservation to adapt multi-task depth, segmentation, and normal prediction from synthetic to real images without target labels.
No more discrimi- nation: Cross city adaptation of road scene segmenters
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
-
UM-Adapt: Unsupervised Multi-Task Adaptation Using Adversarial Cross-Task Distillation
UM-Adapt uses cross-task consistency and contour preservation to adapt multi-task depth, segmentation, and normal prediction from synthetic to real images without target labels.