Deep CNNs Meet Global Covariance Pooling: Better Representation and Generalization

@article{Wang2020DeepCM,
  title={Deep CNNs Meet Global Covariance Pooling: Better Representation and Generalization},
  author={Qilong Wang and Jiangtao Xie and Wangmeng Zuo and Lei Zhang and Peihua Li},
  journal={IEEE transactions on pattern analysis and machine intelligence},
  year={2020}
}
  • Qilong Wang, Jiangtao Xie, +2 authors Peihua Li
  • Published 2020
  • Computer Science, Medicine, Mathematics
  • IEEE transactions on pattern analysis and machine intelligence
  • Compared with global average pooling in existing deep convolutional neural networks (CNNs), global covariance pooling can capture richer statistics of deep features, having potential for improving representation and generalization abilities of deep CNNs. [...] Key Method Furthermore, a global Gaussian embedding method is proposed to incorporate first-order statistics into MPN-COV.Expand Abstract

    Citations

    Publications citing this paper.

    A CNN-Based Tool to Index Emotion on Anime Character Stickers

    VIEW 2 EXCERPTS
    CITES METHODS & BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 68 REFERENCES

    Places: A 10 Million Image Database for Scene Recognition

    VIEW 11 EXCERPTS
    HIGHLY INFLUENTIAL

    Densely Connected Convolutional Networks

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    FASON: First and Second Order Information Fusion Network for Texture Recognition

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    Free-Form Region Description with Second-Order Pooling

    VIEW 16 EXCERPTS
    HIGHLY INFLUENTIAL

    Matrix Backpropagation for Deep Networks with Structured Layers

    VIEW 20 EXCERPTS
    HIGHLY INFLUENTIAL

    Image Classification with the Fisher Vector: Theory and Practice

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    Aggregated Residual Transformations for Deep Neural Networks

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Improved Bilinear Pooling with CNNs

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Local Log-Euclidean Multivariate Gaussian Descriptor and Its Application to Image Classification.

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL