EUVIMEDEUVIMED entdecken
Crossref

Efficient Crowd Counting via Dual Knowledge Distillation

Rui Wang, Yixue Hao, Long Hu, Xianzhi Li, Min Chen, Yiming Miao, Iztok Humar

IEEE Transactions on Image Processing · 2024 · 33 · 569-583

DOI: 10.1109/tip.2023.3343609 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2024. Verfasst von Rui Wang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 24-mal zitiert (Crossref) und nennt 74 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China, Shenzhen Science and Technology Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Most researchers focus on designing accurate crowd counting models with heavy parameters and computations but ignore the resource burden during the model deployment. A real-world scenario demands an efficient counting model with low-latency and high-performance. Knowledge distillation provides an elegant way to transfer knowledge from a complicated teacher model to a compact student model while maintaining accuracy. However, the student model receives the wrong guidance with the supervision of the teacher model due to the inaccurate information understood by the teacher in some cases. In this paper, we propose a dual-knowledge distillation (DKD) framework, which aims to reduce the side effects of the teacher model and transfer hierarchical knowledge to obtain a more efficient counting model. First, the student model is initialized with global information transferred by the teacher model via adaptive perspectives. Then, the self-knowledge distillation forces the student model to learn the knowledge by itself, based on intermediate feature maps and target map. Specifically, the optimal transport distance is utilized to measure the difference of feature maps between the teacher and the student to perform the distribution alignment of the counting area. Extensive experiments are conducted on four challenging datasets, demonstrating the superiority of DKD. When there are only approximately 6% of the parameters and computations from the original models, the student model achieves a faster and more accurate counting performance as the teacher model even surpasses it.

Förderung

  • National Key Research and Development Program of China
  • National Natural Science Foundation of China
  • Shenzhen Science and Technology Program
  • Shenzhen Institute of Artificial Intelligence and Robotics for Society
  • Slovenian Research and Innovation Agency through the Decentralized solutions for the Digitalization of Industry and Smart Cities and Communities under Research…

Zitieren

APA

Wang, R., Hao, Y., Hu, L., Li, X., Chen, M., Miao, Y., & Humar, I. (2024). Efficient Crowd Counting via Dual Knowledge Distillation. IEEE Transactions on Image Processing, 33, 569-583. https://doi.org/10.1109/tip.2023.3343609

Vancouver

Wang R, Hao Y, Hu L, Li X, Chen M, Miao Y, et al. Efficient Crowd Counting via Dual Knowledge Distillation. IEEE Transactions on Image Processing. 2024;33:569-583. doi:10.1109/tip.2023.3343609

BibTeX
@article{Wang2024,
  author = {Wang, Rui and Hao, Yixue and Hu, Long and Li, Xianzhi and Chen, Min and Miao, Yiming and Humar, Iztok},
  title = {Efficient Crowd Counting via Dual Knowledge Distillation},
  journal = {IEEE Transactions on Image Processing},
  year = {2024},
  volume = {33},
  pages = {569--583},
  doi = {10.1109/tip.2023.3343609}
}
Kostenlos starten →