L-CNN: A Lattice cross-fusion strategy for multistream convolutional neural networks

L-CNN: A Lattice cross-fusion strategy for multistream convolutional neural networks

Electronics Letters, vol. 55, no. 22, pp. 1180-1182, 2029 This paper proposes a fusion strategy for multistream convolutional networks, the Lattice Cross Fusion. This approach crosses signals from convolution layers performing mathematical operation-based fusions right before pooling layers. Results...

Full description

Saved in:
Bibliographic Details
Main Authors: de Almeida, Ana Paula G. S, Vidal, Flavio de Barros
Format: Journal Article
Language:English
Published: 01-08-2020
Subjects:
Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Electronics Letters, vol. 55, no. 22, pp. 1180-1182, 2029 This paper proposes a fusion strategy for multistream convolutional networks, the Lattice Cross Fusion. This approach crosses signals from convolution layers performing mathematical operation-based fusions right before pooling layers. Results on a purposely worsened CIFAR-10, a popular image classification data set, with a modified AlexNet-LCNN version show that this novel method outperforms by 46% the baseline single stream network, with faster convergence, stability, and robustness.
DOI:10.48550/arxiv.2008.00157