EUVIMEDEUVIMED entdecken
Crossref

Conv2Former: A Simple Transformer-Style ConvNet for Visual Recognition

Qibin Hou, Cheng-Ze Lu, Ming-Ming Cheng, Jiashi Feng

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-12 · 46 (12) · 8274-8283

DOI: 10.1109/tpami.2024.3401450 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024-12. Verfasst von Qibin Hou und 3 weiteren Autor:innen. Die Arbeit wurde bisher 179-mal zitiert (Crossref) und nennt 87 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities Nankai University, CAST.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Vision Transformers have been the most popular network architecture in visual recognition recently due to the strong ability of encode global information. However, its high computational cost when processing high-resolution images limits the applications in downstream tasks. In this paper, we take a deep look at the internal structure of self-attention and present a simple Transformer style convolutional neural network (ConvNet) for visual recognition. By comparing the design principles of the recent ConvNets and Vision Transformers, we propose to simplify the self-attention by leveraging a convolutional modulation operation. We show that such a simple approach can better take advantage of the large kernels ( ≥ 7×7) nested in convolutional layers and we observe a consistent performance improvement when gradually increasing the kernel size from 5×5 to 21×21. We build a family of hierarchical ConvNets using the proposed convolutional modulation, termed Conv2Former. Our network is simple and easy to follow. Experiments show that our Conv2Former outperforms existent popular ConvNets and vision Transformers, like Swin Transformer and ConvNeXt in all ImageNet classification, COCO object detection and ADE20 k semantic segmentation.

Förderung

  • National Natural Science Foundation of China
  • Fundamental Research Funds for the Central Universities Nankai University
  • CAST
  • Young Elite Scientist Sponsorship
  • Nankai University

Zitieren

APA

Hou, Q., Lu, C. Z., Cheng, M. M., & Feng, J. (2024). Conv2Former: A Simple Transformer-Style ConvNet for Visual Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12), 8274-8283. https://doi.org/10.1109/tpami.2024.3401450

Vancouver

Hou Q, Lu CZ, Cheng MM, Feng J. Conv2Former: A Simple Transformer-Style ConvNet for Visual Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024;46(12):8274-8283. doi:10.1109/tpami.2024.3401450

BibTeX
@article{Hou2024,
  author = {Hou, Qibin and Lu, Cheng-Ze and Cheng, Ming-Ming and Feng, Jiashi},
  title = {Conv2Former: A Simple Transformer-Style ConvNet for Visual Recognition},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2024},
  volume = {46},
  number = {12},
  pages = {8274--8283},
  doi = {10.1109/tpami.2024.3401450}
}
Kostenlos starten →