A continual knowledge-distillation method uses unlabeled inference images and a strong teacher localizer to expand the generalization domain of absolute pose regression.
Title resolution pending
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
-
ConDo: Continual Domain Expansion for Absolute Pose Regression
A continual knowledge-distillation method uses unlabeled inference images and a strong teacher localizer to expand the generalization domain of absolute pose regression.