Prototypical Contrast Adaptation for Domain Adaptive Semantic Segmentation

TL;DR

ProCA introduces a contrastive learning approach that leverages class prototypes and inter-class relationships for improved unsupervised domain adaptation in semantic segmentation, achieving state-of-the-art results.

Contribution

It proposes a novel class-centered distribution alignment method using contrastive learning with prototypes, enhancing domain adaptation performance.

Findings

Achieves state-of-the-art results on GTA5 to Cityscapes and SYNTHIA to Cityscapes tasks.

Effectively incorporates inter-class relationships into domain adaptation.

Demonstrates superior discrimination in target domain segmentation.

Abstract

Unsupervised Domain Adaptation (UDA) aims to adapt the model trained on the labeled source domain to an unlabeled target domain. In this paper, we present Prototypical Contrast Adaptation (ProCA), a simple and efficient contrastive learning method for unsupervised domain adaptive semantic segmentation. Previous domain adaptation methods merely consider the alignment of the intra-class representational distributions across various domains, while the inter-class structural relationship is insufficiently explored, resulting in the aligned representations on the target domain might not be as easily discriminated as done on the source domain anymore. Instead, ProCA incorporates inter-class information into class-wise prototypes, and adopts the class-centered distribution alignment for adaptation. By considering the same class prototypes as positives and other class prototypes as negatives to achieve class-centered distribution alignment, ProCA achieves state-of-the-art performance on classical domain adaptation tasks, {\em i.e., GTA5 $\to$ Cityscapes \text{and} SYNTHIA $\to$ Cityscapes}. Code is available at \href{https://github.com/jiangzhengkai/ProCA}{ProCA}