Unsupervised Domain Adaptation for One-stage Object Detector using Offsets to Bounding Box

TL;DR

This paper introduces OADA, a novel unsupervised domain adaptation method for one-stage object detectors that aligns features based on bounding box offset values, improving domain invariance and detection performance.

Contribution

It proposes a simple, effective conditioning approach to align features conditioned on offsets, addressing negative transfer in domain adaptive object detection.

Findings

Achieves state-of-the-art performance in various settings.

Enhances both discriminability and transferability of features.

Addresses negative transfer caused by feature distribution variations.

Abstract

Most existing domain adaptive object detection methods exploit adversarial feature alignment to adapt the model to a new domain. Recent advances in adversarial feature alignment strives to reduce the negative effect of alignment, or negative transfer, that occurs because the distribution of features varies depending on the category of objects. However, by analyzing the features of the anchor-free one-stage detector, in this paper, we find that negative transfer may occur because the feature distribution varies depending on the regression value for the offset to the bounding box as well as the category. To obtain domain invariance by addressing this issue, we align the feature conditioned on the offset value, considering the modality of the feature distribution. With a very simple and effective conditioning method, we propose OADA (Offset-Aware Domain Adaptive object detector) that achieves state-of-the-art performances in various experimental settings. In addition, by analyzing through singular value decomposition, we find that our model enhances both discriminability and transferability.