Active domain adaptation for semantic segmentation via dynamically balancing domainness and uncertainty

Siqi Zhang, Lu Zhang, Zhiyong Liu

Abstract: Highlights•We propose a novel active domain adaptation framework for semantic segmentation.•The domainness score is designed to evaluate target samples' representativeness.•The prediction uncertainty is utilized to evaluate target samples' discriminability.•We introduce a dynamic balancing mechanism within the acquisition function.•Extensive experiments validate the superior performance of our method.