Source-free Domain Adaptation for Video Object Detection Under Adverse Image Conditions

arXiv:2404.15252v1 Announce Type: new
Abstract: When deploying pre-trained video object detectors in real-world scenarios, the domain gap between training and testing data caused by adverse image conditions often leads to performance degradation. Addressing this issue becomes particularly challenging when only the pre-trained model and degraded videos are available. Although various source-free domain adaptation (SFDA) methods have been proposed for single-frame object detectors, SFDA for video object detection (VOD) remains unexplored. Moreover, most unsupervised domain adaptation works for object detection rely …

abstract arxiv cs.cv data detection detectors domain domain adaptation free gap image issue leads object performance pre-trained model testing training type video videos world