DL-CapsNet: A Deep and Light Capsule Network

被引:1
作者
Shiri, Pouya [1 ]
Baniasadi, Amirali [1 ]
机构
[1] Univ Victoria, Victoria, BC, Canada
来源
DESIGN AND ARCHITECTURE FOR SIGNAL AND IMAGE PROCESSING, DASIP 2022 | 2022年 / 13425卷
基金
加拿大自然科学与工程研究理事会;
关键词
Capsule Networks; Deep CapsNet; Fast CapsNet;
D O I
10.1007/978-3-031-12748-9_5
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Capsule Network (CapsNet) is among the promising classifiers and a possible successor of the classifiers built based on Convolutional Neural Network (CNN). CapsNet is more accurate than CNNs in detecting images with overlapping categories and those with applied affine transformations. In this work, we propose a deep variant of CapsNet consisting of several capsule layers. In addition, we design the Capsule Summarization layer to reduce the complexity by reducing the number of parameters. DL-CapsNet, while being highly accurate, employs a small number of parameters and delivers faster training and inference. DL-CapsNet can process complex datasets with a high number of categories.
引用
收藏
页码:57 / 68
页数:12
相关论文
共 17 条
  • [1] Gabor capsule network with preprocessing blocks for the recognition of complex images
    Abra Ayidzoe, Mighty
    Yu, Yongbin
    Mensah, Patrick Kwabena
    Cai, Jingye
    Adu, Kwabena
    Tang, Yifan
    [J]. MACHINE VISION AND APPLICATIONS, 2021, 32 (04)
  • [2] Mask Dynamic Routing to Combined Model of Deep Capsule Network and U-Net
    Chen, Junying
    Liu, Zhan
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2020, 31 (07) : 2653 - 2664
  • [3] Cv-CapsNet: Complex-Valued Capsule Network
    Cheng, Xinming
    He, Jiangnan
    He, Jianbiao
    Xu, Honglei
    [J]. IEEE ACCESS, 2019, 7 : 85492 - 85499
  • [4] Deli A., 2018, HITNET NEURAL NETWOR, P1
  • [5] Huang GL, 2017, IEEE ICC
  • [6] DA-CapsNet: dual attention mechanism capsule network
    Huang, Wenkai
    Zhou, Fobao
    [J]. SCIENTIFIC REPORTS, 2020, 10 (01)
  • [7] Kolesnikov Alexander, 2020, Computer Vision - ECCV 2020 16th European Conference. Proceedings. Lecture Notes in Computer Science (LNCS 12350), P491, DOI 10.1007/978-3-030-58558-7_29
  • [8] Krizhevsky A., 2009, Cifar-100 (canadian institute for advanced research)
  • [9] LeCun Y., MNIST DATABASE HANDW
  • [10] DeepCaps: Going Deeper with Capsule Networks
    Rajasegaran, Jathushan
    Jayasundara, Vinoj
    Jayasekara, Sandaru
    Jayasekara, Hirunima
    Seneviratne, Suranga
    Rodrigo, Ranga
    [J]. 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 10717 - 10725