A Pairwise DomMix Attentive Adversarial Network for Unsupervised Domain Adaptive Object Detection

TL;DR

This paper introduces a novel unsupervised domain adaptive object detection method using a pairwise attentive adversarial network with a Domain Mixup module, effectively handling large domain shifts by constructing intermediate features and focusing on disparate regions.

Contribution

The paper proposes a pairwise attentive adversarial network with a Domain Mixup module for improved domain alignment in unsupervised object detection, addressing large domain shifts.

Findings

Outperforms existing methods on benchmark datasets.

Effectively mitigates large domain shifts.

Enhances feature sharing between domains.

Abstract

Unsupervised Domain Adaptive Object Detection (DAOD) could adapt a model trained on a source domain to an unlabeled target domain for object detection. Existing unsupervised DAOD methods usually perform feature alignments from the target to the source. Unidirectional domain transfer would omit information about the target samples and result in suboptimal adaptation when there are large domain shifts. Therefore, we propose a pairwise attentive adversarial network with a Domain Mixup (DomMix) module to mitigate the aforementioned challenges. Specifically, a deep-level mixup is employed to construct an intermediate domain that allows features from both domains to share their differences. Then a pairwise attentive adversarial network is applied with attentive encoding on both image-level and instance-level features at different scales and optimizes domain alignment by adversarial learning. This allows the network to focus on regions with disparate contextual information and learn their similarities between different domains. Extensive experiments are conducted on several benchmark datasets, demonstrating the superiority of our proposed method.