Convolutional Fully-Connected Capsule Network (CFC-CapsNet): A Novel and Fast Capsule Network

被引:6
|
作者
Shiri, Pouya [1 ]
Baniasadi, Amirali [1 ]
机构
[1] Univ Victoria, Elect & Comp Engn Fac, Victoria, BC, Canada
来源
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY | 2022年 / 94卷 / 07期
基金
加拿大自然科学与工程研究理事会;
关键词
Capsule Networks; CapsNet; Deep Learning; Fast CapsNet;
D O I
10.1007/s11265-021-01731-6
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
A Capsule Network (CapsNet) is a relatively new classifier and one of the possible successors of Convolutional Neural Networks (CNNs). CapsNet maintains the spatial hierarchies between the features and outperforms CNNs at classifying images including overlapping categories. Even though CapsNet works well on small-scale datasets such as MNIST, it fails to achieve a similar level of performance on more complicated datasets and real applications. In addition, CapsNet is slow compared to CNNs when performing the same task and relies on a higher number of parameters. In this work, we introduce Convolutional Fully-Connected Capsule Network (CFC-CapsNet) to address the shortcomings of CapsNet by creating capsules using a different method. We introduce a new layer (CFC layer) as an alternative solution to creating capsules. CFC-CapsNet produces fewer, yet more powerful capsules resulting in higher network accuracy. Our experiments show that CFC-CapsNet achieves competitive accuracy, faster training and inference and uses less number of parameters on the CIFAR-10, SVHN and Fashion-MNIST datasets compared to conventional CapsNet.
引用
收藏
页码:645 / 658
页数:14
相关论文
共 50 条
  • [11] DA-CapsNet: dual attention mechanism capsule network
    Wenkai Huang
    Fobao Zhou
    Scientific Reports, 10
  • [12] Cv-CapsNet: Complex-Valued Capsule Network
    Cheng, Xinming
    He, Jiangnan
    He, Jianbiao
    Xu, Honglei
    IEEE ACCESS, 2019, 7 : 85492 - 85499
  • [13] Res-CapsNet: Residual Capsule Network for Data Classification
    Xiaofen Jia
    Jianqiao Li
    Baiting Zhao
    Yongcun Guo
    Yourui Huang
    Neural Processing Letters, 2022, 54 : 4229 - 4245
  • [14] Capsule Network is Not More Robust than Convolutional Network
    Gu, Jindong
    Tresp, Volker
    Hu, Han
    2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 14304 - 14312
  • [15] DA-CapsNet: dual attention mechanism capsule network
    Huang, Wenkai
    Zhou, Fobao
    SCIENTIFIC REPORTS, 2020, 10 (01)
  • [16] A Comparative Study of a Fully-Connected Artificial Neural Network and a Convolutional Neural Network in Predicting Bridge Maintenance Costs
    Wang, Chongjiao
    Yao, Changrong
    Zhao, Siguang
    Zhao, Shida
    Li, Yadong
    APPLIED SCIENCES-BASEL, 2022, 12 (07):
  • [17] A Novel Fully-Connected Beamforming Network Architecture With the QR Factorization Synthesis Method
    Sun, Li
    Gu, Jun-Yi
    Wan, Yang-Ming
    Zhou, Shi-Gang
    IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, 2024, 72 (11) : 8772 - 8785
  • [18] Efficient-CapsNet: capsule network with self-attention routing
    Mazzia, Vittorio
    Salvetti, Francesco
    Chiaberge, Marcello
    SCIENTIFIC REPORTS, 2021, 11 (01)
  • [19] Efficient-CapsNet: capsule network with self-attention routing
    Vittorio Mazzia
    Francesco Salvetti
    Marcello Chiaberge
    Scientific Reports, 11
  • [20] R-CapsNet: An Improvement of Capsule Network for More Complex Data
    Luo, Lu
    Duan, Shukai
    Wang, Lidan
    2019 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI 2019), 2019, : 2124 - 2129