Flip-Rotate-Pooling Convolution and Split Dropout on Convolution Neural Networks for Image Classification

Abstract

This paper presents a new version of Dropout called Split Dropout (sDropout) and rotational convolution techniques to improve CNNs' performance on image classification. The widely used standard Dropout has advantage of preventing deep neural networks from overfitting by randomly dropping units during training. Our sDropout randomly splits the data into two subsets and keeps both rather than discards one subset. We also introduce two rotational convolution techniques, i.e. rotate-pooling convolution (RPC) and flip-rotate-pooling convolution (FRPC) to boost CNNs' performance on the robustness for rotation transformation. These two techniques encode rotation invariance into the network without adding extra parameters. Experimental evaluations on ImageNet2012 classification task demonstrate that sDropout not only enhances the performance but also converges faster. Additionally, RPC and FRPC make CNNs more robust for rotation transformations. Overall, FRPC together with sDropout bring $1.18\%$ (model of Zeiler and Fergus~\cite{zeiler2013visualizing}, 10-view, top-1) accuracy increase in ImageNet 2012 classification task compared to the original network.

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Improving neural networks by preventing co-adaptation of feature detectors
• Going Deeper with Convolutions
• One weird trick for parallelizing convolutional neural networks
• Visualizing and Understanding Convolutional Networks
• Some Improvements on Deep Convolutional Neural Network Based Image Classification
• Unsupervised feature learning by augmenting single images

Cited by in corpus (6)

• Exploiting Cyclic Symmetry in Convolutional Neural Networks
• LGN-CNN: a biologically inspired CNN architecture
• Scale-invariant scale-channel networks: Deep networks that generalise to previously unseen scales
• Deep Rotation Equivariant Network
• Non-Parametric Transformation Networks
• More Is More -- Narrowing the Generalization Gap by Adding Classification Heads