Unsupervised Domain Adaptation with Implicit Pseudo Supervision for Semantic Segmentation
This work addresses domain shift issues in semantic segmentation for applications like autonomous driving, but it is incremental as it builds on existing pseudo-labelling techniques.
The paper tackles the problem of noisy pseudo labels in unsupervised domain adaptation for semantic segmentation by introducing a tri-learning architecture that implicitly generates and aligns pseudo labels, resulting in considerable improvements on GTA5 to Cityscapes and SYNTHIA to Cityscapes tasks.
Pseudo-labelling is a popular technique in unsuper-vised domain adaptation for semantic segmentation. However, pseudo labels are noisy and inevitably have confirmation bias due to the discrepancy between source and target domains and training process. In this paper, we train the model by the pseudo labels which are implicitly produced by itself to learn new complementary knowledge about target domain. Specifically, we propose a tri-learning architecture, where every two branches produce the pseudo labels to train the third one. And we align the pseudo labels based on the similarity of the probability distributions for each two branches. To further implicitly utilize the pseudo labels, we maximize the distances of features for different classes and minimize the distances for the same classes by triplet loss. Extensive experiments on GTA5 to Cityscapes and SYNTHIA to Cityscapes tasks show that the proposed method has considerable improvements.