EDRL: Entropy-guided disentangled representation learning for unsupervised domain adaptation in semantic segmentation

Runze Wang, Qin Zhou, Guoyan Zheng

Abstract: Highlights•A method synergistically integrating disentangled representation learning and entropy-based adversarial learning.•A soft gating strategy to dynamically select features for further performance improvement.•The proposed method achieved better results than other state-of-the-art methods when validated on two public datasets.