WUDA: Unsupervised Domain Adaptation Based on Weak Source Domain Labels

TL;DR

This paper introduces WUDA, a new task combining weak source labels with unsupervised domain adaptation for semantic segmentation, and explores two frameworks with extensive experiments and domain shift analysis.

Contribution

It defines the WUDA task and proposes two frameworks for weakly supervised domain adaptation in semantic segmentation, along with dataset construction and domain shift measurement.

Findings

Different frameworks perform variably across datasets with varying domain shifts.

The paper constructs diverse dataset pairs to analyze domain shift effects.

Introduces the use of representation shift metric for urban landscape segmentation.

Abstract

Unsupervised domain adaptation (UDA) for semantic segmentation addresses the cross-domain problem with fine source domain labels. However, the acquisition of semantic labels has always been a difficult step, many scenarios only have weak labels (e.g. bounding boxes). For scenarios where weak supervision and cross-domain problems coexist, this paper defines a new task: unsupervised domain adaptation based on weak source domain labels (WUDA). To explore solutions for this task, this paper proposes two intuitive frameworks: 1) Perform weakly supervised semantic segmentation in the source domain, and then implement unsupervised domain adaptation; 2) Train an object detection model using source domain data, then detect objects in the target domain and implement weakly supervised semantic segmentation. We observe that the two frameworks behave differently when the datasets change. Therefore, we construct dataset pairs with a wide range of domain shifts and conduct extended experiments to analyze the impact of different domain shifts on the two frameworks. In addition, to measure domain shift, we apply the metric representation shift to urban landscape image segmentation for the first time. The source code and constructed datasets are available at \url{https://github.com/bupt-ai-cz/WUDA}.