D-Know: Disentangled Domain Knowledge-Aided Learning for Open-Domain Continual Object Detection
Applied Sciences · 01.12.2025 · 15 (23) · 12723
DOI: 10.3390/app152312723 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 01.12.2025. Verfasst von Bintao He und 6 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Major Project, NSFC, State Key Laboratory of Communication Content Cognition.
Abstract (Originalsprache)
Continual learning for open-vocabulary object detection aims to enable pretrained vision–language detectors to adapt to diverse specialized domains while preserving their zero-shot generalization capabilities. However, existing methods primarily focus on mitigating catastrophic forgetting, often neglecting the substantial domain shifts commonly encountered in real-world applications. To address this critical oversight, we pioneer Open-Domain Continual Object Detection (OD-COD), a new paradigm that requires detectors to continually adapt across domains with significant stylistic gaps. We propose Disentangled Domain Knowledge-Aided Learning (D-Know) to tackle this challenge. This framework explicitly disentangles domain-general priors from category-specific adaptation, managing them dynamically in a scalable domain knowledge base. Specifically, D-Know first learns domain priors in a self-supervised manner and then leverages these priors to facilitate category-specific adaptation within each domain. To rigorously evaluate this task, we construct OD-CODB, the first dedicated benchmark spanning six domains with substantial visual variations. Extensive experiments demonstrate that D-Know achieves superior performance, surpassing current state-of-the-art methods by an average of 4.2% mAP under open-domain continual settings while maintaining strong zero-shot generalization. Furthermore, experiments under the few-shot setting confirm D-Know’s superior data efficiency.
Förderung
- National Science and Technology Major Project
- NSFC
- State Key Laboratory of Communication Content Cognition
- Project of China Knowledge Centre for Engineering Science and Technology
Zitieren
APA
He, B., Yan, C., Kou, Y., Wang, Y., Lv, X., Du, H., & Xie, Y. (2025). D-Know: Disentangled Domain Knowledge-Aided Learning for Open-Domain Continual Object Detection. Applied Sciences, 15(23), 12723. https://doi.org/10.3390/app152312723
Vancouver
He B, Yan C, Kou Y, Wang Y, Lv X, Du H, et al. D-Know: Disentangled Domain Knowledge-Aided Learning for Open-Domain Continual Object Detection. Applied Sciences. 2025;15(23):12723. doi:10.3390/app152312723
BibTeX
@article{He2025,
author = {He, Bintao and Yan, Caixia and Kou, Yan and Wang, Yinghao and Lv, Xin and Du, Haipeng and Xie, Yugui},
title = {D-Know: Disentangled Domain Knowledge-Aided Learning for Open-Domain Continual Object Detection},
journal = {Applied Sciences},
year = {2025},
volume = {15},
number = {23},
pages = {12723},
doi = {10.3390/app152312723}
}
EUVIMED