Improving Online Source-free Domain Adaptation for Object Detection by Unsupervised Data Acquisition

TL;DR

This paper proposes an unsupervised data acquisition method to improve online source-free domain adaptation for object detection, effectively selecting informative frames to enhance model adaptation in autonomous vehicle scenarios.

Contribution

It introduces a novel approach that prioritizes informative unlabeled frames for online training, advancing the effectiveness of source-free domain adaptation in object detection.

Findings

Outperforms existing state-of-the-art O-SFDA methods

Demonstrates improved detection accuracy in real-world datasets

Validates the effectiveness of unsupervised data acquisition

Abstract

Effective object detection in autonomous vehicles is challenged by deployment in diverse and unfamiliar environments. Online Source-Free Domain Adaptation (O-SFDA) offers model adaptation using a stream of unlabeled data from a target domain in an online manner. However, not all captured frames contain information beneficial for adaptation, especially in the presence of redundant data and class imbalance issues. This paper introduces a novel approach to enhance O-SFDA for adaptive object detection through unsupervised data acquisition. Our methodology prioritizes the most informative unlabeled frames for inclusion in the online training process. Empirical evaluation on a real-world dataset reveals that our method outperforms existing state-of-the-art O-SFDA techniques, demonstrating the viability of unsupervised data acquisition for improving the adaptive object detector.