Show simple item record

dc.contributor.authorNikzad, M
dc.contributor.authorGao, Y
dc.contributor.authorZhou, J
dc.description.abstractPooling layers are an important part of convolutional neural networks (CNNs). They reduce the dimensionality of feature maps and pass salient information to subsequent layers. In this paper, we introduce a novel gradient-based feature pooling method that can down-sample feature maps while better preserving key information. This method considers the spatial gradient of the pixels within a pooling region as a key to select the most possible descriptive information in contrast to the current practice of existing methods that mostly rely on the pixel values. Extensive experiments on different benchmark image classification tasks and CNN architectures demonstrate that the proposed method achieves superior results over existing pooling approaches.
dc.relation.ispartofconferencename2019 IEEE Visual Communications and Image Processing (VCIP 2019)
dc.relation.ispartofconferencetitle2019 IEEE International Conference on Visual Communications and Image Processing, VCIP 2019
dc.relation.ispartoflocationSydney, Australia
dc.subject.fieldofresearchArtificial Intelligence and Image Processing
dc.titleGradient-based pooling for convolutional neural networks
dc.typeConference output
dc.type.descriptionE1 - Conferences
dcterms.bibliographicCitationNikzad, M; Gao, Y; Zhou, J, Gradient-based pooling for convolutional neural networks, 2019 IEEE International Conference on Visual Communications and Image Processing, VCIP 2019, 2019
gro.hasfulltextNo Full Text
gro.griffith.authorZhou, Jun
gro.griffith.authorGao, Yongsheng
gro.griffith.authorNikzad Dehaji, Nick

Files in this item


There are no files associated with this item.

This item appears in the following Collection(s)

  • Conference outputs
    Contains papers delivered by Griffith authors at national and international conferences.

Show simple item record