EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

FeatAug-DETR: Enriching One-to-Many Matching for DETRs With Feature Augmentation

Rongyao Fang, Peng Gao, Aojun Zhou, Yingjie Cai, Si Liu, Jifeng Dai, Hongsheng Li

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-09 · 46 (9) · 6402-6415

DOI: 10.1109/tpami.2024.3381961 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024-09. Verfasst von Rongyao Fang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 25-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key R&D Program of China, Centre for Perceptual and Interactive Intelligence (CPII) Ltd., General Research Fund of Hong Kong RGC.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

One-to-one matching is a crucial design in DETR-like object detection frameworks. It enables the DETR to perform end-to-end detection. However, it also faces challenges of lacking positive sample supervision and slow convergence speed. Several recent works proposed the one-to-many matching mechanism to accelerate training and boost detection performance. We revisit these methods and model them in a unified format of augmenting the object queries. In this paper, we propose two methods that realize one-to-many matching from a different perspective of augmenting images or image features. The first method is One-to-many Matching via Data Augmentation (denoted as DataAug-DETR ). It spatially transforms the images and includes multiple augmented versions of each image in the same training batch. Such a simple augmentation strategy already achieves one-to-many matching and surprisingly improves DETR's performance. The second method is One-to-many matching via Feature Augmentation (denoted as FeatAug-DETR ). Unlike DataAug-DETR , it augments the image features instead of the original images and includes multiple augmented features in the same batch to realize one-to-many matching. FeatAug-DETR significantly accelerates DETR training and boosts detection performance while keeping the inference speed unchanged. We conduct extensive experiments to evaluate the effectiveness of the proposed approach on DETR variants, including DAB-DETR, Deformable-DETR, and $\mathcal {H}$ H -Deformable-DETR. Without extra training data, FeatAug-DETR shortens the training convergence periods of Deformable-DETR (Zhu et al. 2020) to 24 epochs and achieves 58.3 AP on COCO val2017 set with Swin-L as the backbone.

Förderung

  • National Key R&D Program of China
  • Centre for Perceptual and Interactive Intelligence (CPII) Ltd.
  • General Research Fund of Hong Kong RGC

Zitieren

APA

Fang, R., Gao, P., Zhou, A., Cai, Y., Liu, S., Dai, J., & Li, H. (2024). FeatAug-DETR: Enriching One-to-Many Matching for DETRs With Feature Augmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(9), 6402-6415. https://doi.org/10.1109/tpami.2024.3381961

Vancouver

Fang R, Gao P, Zhou A, Cai Y, Liu S, Dai J, et al. FeatAug-DETR: Enriching One-to-Many Matching for DETRs With Feature Augmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024;46(9):6402-6415. doi:10.1109/tpami.2024.3381961

BibTeX
@article{Fang2024,
  author = {Fang, Rongyao and Gao, Peng and Zhou, Aojun and Cai, Yingjie and Liu, Si and Dai, Jifeng and Li, Hongsheng},
  title = {FeatAug-DETR: Enriching One-to-Many Matching for DETRs With Feature Augmentation},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2024},
  volume = {46},
  number = {9},
  pages = {6402--6415},
  doi = {10.1109/tpami.2024.3381961}
}
Kostenlos starten →