Data Augmentation Using Random Image Cropping and Patching for Deep CNNs

  title={Data Augmentation Using Random Image Cropping and Patching for Deep CNNs},
  author={Ryo Takahashi and Takashi Matsubara and Kuniaki Uehara},
  journal={IEEE Transactions on Circuits and Systems for Video Technology},
Deep convolutional neural networks (CNNs) have achieved remarkable results in image processing tasks. However, their high expression ability risks overfitting. Consequently, data augmentation techniques have been proposed to prevent overfitting while enriching datasets. Recent CNN architectures with more parameters are rendering traditional data augmentation techniques insufficient. In this study, we propose a new data augmentation technique called random image cropping and patching (RICAP… 

Figures and Tables from this paper

Data Augmentation Using Mixup and Random Erasing

This paper proposes two different combinations of these two methods, namely RSM and RDM, to compensate their respective shortcomings for object detection and image classification on various datasets.

TokenMix: Rethinking Image Mixing for Data Augmentation in Vision Transformers

This paper proposes a novel data augmentation technique TokenMix, which helps vision transformers focus on the foreground area to infer the classes and enhances their robustness to occlusion, with consistent performance gains.

A survey on Image Data Augmentation for Deep Learning

This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing DataAugmentation, a data-space solution to the problem of limited data.

Distribution-preserving data augmentation

A novel distribution-preserving data augmentation method is proposed that creates plausible image variations by shifting pixel colors to another point in the image color distribution by defining a regularized density decreasing direction to create paths from the original pixels’ color to the distribution tails.

Enhancing Performance of Deep Learning Models with different Data Augmentation Techniques: A Survey

  • Cherry KhoslaB. Saini
  • Computer Science
    2020 International Conference on Intelligent Engineering and Management (ICIEM)
  • 2020
This survey paper focuses on different data augmentation techniques based on data warping and oversampling and gives a brief discussion on different solutions of reducing the overfitting problem.

Regional Patch-Based Feature Interpolation Method for Effective Regularization

This paper proposes a regularization method that applies both image manipulation and feature map regularization based on patches that improves the performance of the model and makes it possible to better generalize the model.

Empirical Evaluation of the Effectiveness of Variational Autoencoders on Data Augmentation for the Image Classification Problem

This study evaluates the effectiveness of a generative model, variational autoencoders (VAE), on the image classification problem and observes that the contribution of data augmentation is sensitive to the size of the dataset and VAE augmented is as effective as the transformation based augmentation methods.

PLANET: Improved Convolutional Neural Networks with Image Enhancement for Image Classification

Sufficient experiment results show that PLANET utilizing InnerMove for image enhancement outperforms the comparative algorithms, and InnerMove has a more significant effect than the comparative data enhancement methods for image classification tasks.

Rainwater-Removal Image Conversion Learning with Training Pair Augmentation

An image conversion method that efficiently removes raindrops on a camera lens from an image using a deep learning technique and presents a raindrop-removed image using the Pix2pix generative adversarial network model.



RICAP: Random Image Cropping and Patching Data Augmentation for Deep CNNs

A new data augmentation technique called random image cropping and patching (RICAP), which randomly crops four images and patches them to construct a new training image, enriching the variety of training images.

Random Erasing Data Augmentation

In training, Random Erasing randomly selects a rectangle region in an image and erases its pixels with random values and yields consistent improvement over strong baselines in image classification, object detection and person re-identification.

Improved Regularization of Convolutional Neural Networks with Cutout

This paper shows that the simple regularization technique of randomly masking out square regions of input during training, which is called cutout, can be used to improve the robustness and overall performance of convolutional neural networks.

Striving for Simplicity: The All Convolutional Net

It is found that max-pooling can simply be replaced by a convolutional layer with increased stride without loss in accuracy on several image recognition benchmarks.

Deep Pyramidal Residual Networks

This research gradually increases the feature map dimension at all units to involve as many locations as possible in the network architecture and proposes a novel residual unit capable of further improving the classification accuracy with the new network architecture.

Deeply-Supervised Nets

The proposed deeply-supervised nets (DSN) method simultaneously minimizes classification error while making the learning process of hidden layers direct and transparent, and extends techniques from stochastic gradient methods to analyze the algorithm.

AutoAugment: Learning Augmentation Policies from Data

This paper describes a simple procedure called AutoAugment to automatically search for improved data augmentation policies, which achieves state-of-the-art accuracy on CIFAR-10, CIFar-100, SVHN, and ImageNet (without additional data).

Aggregated Residual Transformations for Deep Neural Networks

On the ImageNet-1K dataset, it is empirically show that even under the restricted condition of maintaining complexity, increasing cardinality is able to improve classification accuracy and is more effective than going deeper or wider when the authors increase the capacity.

Very Deep Convolutional Networks for Large-Scale Image Recognition

This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.

Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

This work proposes a Parametric Rectified Linear Unit (PReLU) that generalizes the traditional rectified unit and derives a robust initialization method that particularly considers the rectifier nonlinearities.