EUVIMEDEUVIMED entdecken
Crossref

Representative Feature Alignment for Adaptive Object Detection

Shan Xu, Huaidong Zhang, Xuemiao Xu, Xiaowei Hu, Yangyang Xu, Liangui Dai, Kup-Sze Choi, Pheng-Ann Heng

IEEE Transactions on Circuits and Systems for Video Technology · 2023-02 · 33 (2) · 689-700

DOI: 10.1109/tcsvt.2022.3202094 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2023-02. Verfasst von Shan Xu und 7 weiteren Autor:innen. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 72 Literaturangaben. Gefördert von: Key-Area Research and Development Program of Guangdong Province, China, NSFC, Hong Kong Innovation and Technology Fund.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unsupervised domain adaptation for object detection aims to generalize the object detector trained on the label-rich source domain to the unlabeled target domain. Recently, existing works adopt the instance-level alignment or pixel-level alignment to perform domain transfer, which can effectively avoid the negative transfer due to the diverse background between domains. However, we find that they treat all the regions of an instance feature equally without suppressing background area. They do not segment the specific texture and discriminative regions of objects, which are transferable during adaptation. We call the features that combine the local structure feature and semantic discriminant features as representative features. We propose a novel Representative Feature Alignment (RFA) model to align the features extracted from representative patterns of objects, i.e. representative features, for domain adaptation. Specifically, the representative features are extracted by the Representative Feature Extraction (RFE) submodules. The RFE submodules take the features extracted from different intermediate layers of the detector as input, and filter out the representative features layer-by-layer via integrating class weighting generator, category selection and class activation mapping. Then the representative features from multi-layers are further adaptively aggregated to obtain the final representative features, which are utilized to conduct feature alignment in a class-aware manner. Our representative features are free of untransferable regions and background areas, which leads to better feature alignment. Extensive experimental results show that the proposed model outperforms state-of-the-art methods on a few benchmark datasets.

Förderung

  • Key-Area Research and Development Program of Guangdong Province, China
  • NSFC
  • Hong Kong Innovation and Technology Fund
  • Hong Kong Research Grants Council

Zitieren

APA

Xu, S., Zhang, H., Xu, X., Hu, X., Xu, Y., Dai, L., Choi, K., & Heng, P. (2023). Representative Feature Alignment for Adaptive Object Detection. IEEE Transactions on Circuits and Systems for Video Technology, 33(2), 689-700. https://doi.org/10.1109/tcsvt.2022.3202094

Vancouver

Xu S, Zhang H, Xu X, Hu X, Xu Y, Dai L, et al. Representative Feature Alignment for Adaptive Object Detection. IEEE Transactions on Circuits and Systems for Video Technology. 2023;33(2):689-700. doi:10.1109/tcsvt.2022.3202094

BibTeX
@article{Xu2023,
  author = {Xu, Shan and Zhang, Huaidong and Xu, Xuemiao and Hu, Xiaowei and Xu, Yangyang and Dai, Liangui and Choi, Kup-Sze and Heng, Pheng-Ann},
  title = {Representative Feature Alignment for Adaptive Object Detection},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2023},
  volume = {33},
  number = {2},
  pages = {689--700},
  doi = {10.1109/tcsvt.2022.3202094}
}
Kostenlos starten →