Joint Distribution Alignment via Adversarial Learning for Domain Adaptive Object Detection

TL;DR

This paper introduces JADF, a novel adversarial learning framework for domain adaptive object detection that aligns both marginal and conditional distributions and considers class-wise transferability, improving performance in unsupervised and few-shot settings.

Contribution

The paper proposes a joint adversarial adaptation framework that aligns distributions without extra hyperparameters and incorporates class transferability assessment, advancing domain adaptive detection methods.

Findings

JADF outperforms existing methods in UDA and UFDA settings.

Effective alignment of marginal and conditional distributions improves detection accuracy.

Class-wise transferability assessment enhances domain adaptation performance.

Abstract

Unsupervised domain adaptive object detection aims to adapt a well-trained detector from its original source domain with rich labeled data to a new target domain with unlabeled data. Recently, mainstream approaches perform this task through adversarial learning, yet still suffer from two limitations. First, they mainly align marginal distribution by unsupervised cross-domain feature matching, and ignore each feature's categorical and positional information that can be exploited for conditional alignment; Second, they treat all classes as equally important for transferring cross-domain knowledge and ignore that different classes usually have different transferability. In this paper, we propose a joint adaptive detection framework (JADF) to address the above challenges. First, an end-to-end joint adversarial adaptation framework for object detection is proposed, which aligns both marginal and conditional distributions between domains without introducing any extra hyperparameter. Next, to consider the transferability of each object class, a metric for class-wise transferability assessment is proposed, which is incorporated into the JADF objective for domain adaptation. Further, an extended study from unsupervised domain adaptation (UDA) to unsupervised few-shot domain adaptation (UFDA) is conducted, where only a few unlabeled training images are available in unlabeled target domain. Extensive experiments validate that JADF is effective in both the UDA and UFDA settings, achieving significant performance gains over existing state-of-the-art cross-domain detection methods.