MADAv2: Advanced Multi-Anchor Based Active Domain Adaptation Segmentation

TL;DR

This paper introduces MADAv2, a novel multi-anchor active domain adaptation method for semantic segmentation that effectively selects informative target samples and employs semi-supervised strategies to improve performance, outperforming existing methods.

Contribution

The paper proposes a multi-anchor based active sample selection and semi-supervised adaptation strategy, significantly enhancing domain adaptation for segmentation tasks.

Findings

Achieves 71.4% mIoU on GTA5 dataset

Outperforms state-of-the-art methods by large margins

Effectiveness confirmed through ablation studies

Abstract

Unsupervised domain adaption has been widely adopted in tasks with scarce annotated data. Unfortunately, mapping the target-domain distribution to the source-domain unconditionally may distort the essential structural information of the target-domain data, leading to inferior performance. To address this issue, we firstly propose to introduce active sample selection to assist domain adaptation regarding the semantic segmentation task. By innovatively adopting multiple anchors instead of a single centroid, both source and target domains can be better characterized as multimodal distributions, in which way more complementary and informative samples are selected from the target domain. With only a little workload to manually annotate these active samples, the distortion of the target-domain distribution can be effectively alleviated, achieving a large performance gain. In addition, a powerful semi-supervised domain adaptation strategy is proposed to alleviate the long-tail distribution problem and further improve the segmentation performance. Extensive experiments are conducted on public datasets, and the results demonstrate that the proposed approach outperforms state-of-the-art methods by large margins and achieves similar performance to the fully-supervised upperbound, i.e., 71.4% mIoU on GTA5 and 71.8% mIoU on SYNTHIA. The effectiveness of each component is also verified by thorough ablation studies.