Parallel Grid Pooling for Data Augmentation
arXiv:1803.11370
Abstract
Convolutional neural network (CNN) architectures utilize downsampling layers, which restrict the subsequent layers to learn spatially invariant features while reducing computational costs. However, such a downsampling operation makes it impossible to use the full spectrum of input features. Motivated by this observation, we propose a novel layer called parallel grid pooling (PGP) which is applicable to various CNN models. PGP performs downsampling without discarding any intermediate feature. It works as data augmentation and is complementary to commonly used data augmentation techniques. Furthermore, we demonstrate that a dilated convolution can naturally be represented using PGP operations, which suggests that the dilated convolution can also be regarded as a type of data augmentation technique. Experimental results based on popular image classification benchmarks demonstrate the effectiveness of the proposed method. Code is available at: https://github.com/akitotakeki
References in corpus (17)
- SMOTE: Synthetic Minority Over-sampling Technique
- mixup: Beyond Empirical Risk Minimization
- WaveNet: A Generative Model for Raw Audio
- Improved Regularization of Convolutional Neural Networks with Cutout
- Wide Residual Networks
- SGDR: Stochastic Gradient Descent with Warm Restarts
- Focal Loss for Dense Object Detection
- Random Erasing Data Augmentation
- Stochastic Pooling for Regularization of Deep Convolutional Neural Networks
- Smart Augmentation - Learning an Optimal Data Augmentation Strategy
- Fractional Max-Pooling
- Neural Machine Translation in Linear Time
- ShakeDrop Regularization for Deep Residual Learning
- Dilated Residual Networks
- Dual Path Networks
- ChainerCV: a Library for Deep Learning in Computer Vision
- Feature Space Transfer for Data Augmentation