EUVIMEDEUVIMED entdecken
Crossref

Adapting Object Size Variance and Class Imbalance for Semi-supervised Object Detection

Yuxiang Nie, Chaowei Fang, Lechao Cheng, Liang Lin, Guanbin Li

Proceedings of the AAAI Conference on Artificial Intelligence · 26.06.2023 · 37 (2) · 1966-1974

DOI: 10.1609/aaai.v37i2.25288 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 26.06.2023. Verfasst von Yuxiang Nie und 4 weiteren Autor:innen. Die Arbeit wurde bisher 18-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Semi-supervised object detection (SSOD) attracts extensive research interest due to its great significance in reducing the data annotation effort. Collecting high-quality and category-balanced pseudo labels for unlabeled images is critical to addressing the SSOD problem. However, most of the existing pseudo-labeling-based methods depend on a large and fixed threshold to select high-quality pseudo labels from the predictions of a teacher model. Considering different object classes usually have different detection difficulty levels due to scale variance and data distribution imbalance, conventional pseudo-labeling-based methods are arduous to explore the value of unlabeled data sufficiently. To address these issues, we propose an adaptive pseudo labeling strategy, which can assign thresholds to classes with respect to their “hardness”. This is beneficial for ensuring the high quality of easier classes and increasing the quantity of harder classes simultaneously. Besides, label refinement modules are set up based on box jittering for guaranteeing the localization quality of pseudo labels. To further improve the algorithm’s robustness against scale variance and make the most of pseudo labels, we devise a joint feature-level and prediction-level consistency learning pipeline for transferring the information of the teacher model to the student model. Extensive experiments on COCO and VOC datasets indicate that our method achieves state-of-the-art performance. Especially, it brings mean average precision gains of 2.08 and 1.28 on MS-COCO dataset with 5% and 10% labeled images, respectively.

Zitieren

APA

Nie, Y., Fang, C., Cheng, L., Lin, L., & Li, G. (2023). Adapting Object Size Variance and Class Imbalance for Semi-supervised Object Detection. Proceedings of the AAAI Conference on Artificial Intelligence, 37(2), 1966-1974. https://doi.org/10.1609/aaai.v37i2.25288

Vancouver

Nie Y, Fang C, Cheng L, Lin L, Li G. Adapting Object Size Variance and Class Imbalance for Semi-supervised Object Detection. Proceedings of the AAAI Conference on Artificial Intelligence. 2023;37(2):1966-1974. doi:10.1609/aaai.v37i2.25288

BibTeX
@article{Nie2023,
  author = {Nie, Yuxiang and Fang, Chaowei and Cheng, Lechao and Lin, Liang and Li, Guanbin},
  title = {Adapting Object Size Variance and Class Imbalance for Semi-supervised Object Detection},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2023},
  volume = {37},
  number = {2},
  pages = {1966--1974},
  doi = {10.1609/aaai.v37i2.25288}
}
Kostenlos starten →