Unlabeled data augmentation with diffusion model for semi-supervised object detection
The Visual Computer · 2025-11 · 41 (14) · 11767-11782
DOI: 10.1007/s00371-025-04103-4 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Visual Computer, online erschienen am 25.08.2025. Verfasst von Zhanyun Lu und 8 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 64 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
In the field of medical object detection, acquiring abundant high-quality data remains a critical challenge. To address this issue, we propose DABDM (data augmentation based on diffusion model), a semi-supervised object detection framework that exploits diffusion models to enhance medical image datasets. Our approach generates a large volume of synthetic images, which are utilized for the unsupervised training phase in semi-supervised learning, thereby mitigating the reliance on extensive labeled data. Experimental results demonstrate that DABDM significantly improves the performance of medical object detection model, showcasing its potential to advance the field by providing a robust solution to the data scarcity problem. By incorporating the generated unlabeled training data into the semi-supervised framework, we observed a notable improvement in model accuracy. Specifically, our experiments showed an increase of up to 6.92% after adding the generated images.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Lu, Z., Gu, R., Cheng, H., Xu, M., Xu, P., Kinoshita, Y., Ye, J., Jia, G., & Wu, Q. (2025). Unlabeled data augmentation with diffusion model for semi-supervised object detection. The Visual Computer, 41(14), 11767-11782. https://doi.org/10.1007/s00371-025-04103-4
Vancouver
Lu Z, Gu R, Cheng H, Xu M, Xu P, Kinoshita Y, et al. Unlabeled data augmentation with diffusion model for semi-supervised object detection. The Visual Computer. 2025;41(14):11767-11782. doi:10.1007/s00371-025-04103-4
BibTeX
@article{Lu2025,
author = {Lu, Zhanyun and Gu, Renshu and Cheng, Huimin and Xu, Mingyu and Xu, Peifang and Kinoshita, Yuichiro and Ye, Juan and Jia, Gangyong and Wu, Qing},
title = {Unlabeled data augmentation with diffusion model for semi-supervised object detection},
journal = {The Visual Computer},
year = {2025},
volume = {41},
number = {14},
pages = {11767--11782},
doi = {10.1007/s00371-025-04103-4}
}
EUVIMED