A survey on Image Data Augmentation for Deep Learning (2019-07-06T00:00:00.000000Z)

TL;DR

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.

Abstract

Deep convolutional neural networks have performed remarkably well on many Computer Vision tasks. However, these networks are heavily reliant on big data to avoid overfitting. Overfitting refers to the phenomenon when a network learns a function with very high variance such as to perfectly model the training data. Unfortunately, many application domains do not have access to big data, such as medical image analysis. This survey focuses on Data Augmentation, a data-space solution to the problem of limited data. Data Augmentation encompasses a suite of techniques that enhance the size and quality of training datasets such that better Deep Learning models can be built using them. The image augmentation algorithms discussed in this survey include geometric transformations, color space augmentations, kernel filters, mixing images, random erasing, feature space augmentation, adversarial training, generative adversarial networks, neural style transfer, and meta-learning. The application of augmentation methods based on GANs are heavily covered in this survey. In addition to augmentation techniques, this paper will briefly discuss other characteristics of Data Augmentation such as test-time augmentation, resolution impact, final dataset size, and curriculum learning. This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing Data Augmentation. Readers will understand how Data Augmentation can improve the performance of their models and expand limited datasets to take advantage of the capabilities of big data.

Authors

T. Khoshgoftaar

23 papers

Connor Shorten

3 papers

TL;DR

Abstract

Authors

References142 items

Adversarial Framing for Image and Video Classification

Data Augmentation Using Random Image Cropping and Patching for Deep CNNs

Learning data augmentation policies using augmented random search

Forward Noise Adjustment Scheme for Data Augmentation

A survey on addressing high-class imbalance in big data

Understanding Mixup Training Methods

Generative Adversarial Network in Medical Imaging: A Review

Albumentations: fast and flexible image augmentations

Style Augmentation: Data Augmentation via Style Randomization

Data Augmentation for Skin Lesion Analysis

DADA: Deep Adversarial Data Augmentation for Extremely Low Data Regime Classification

DOPING: Generative Data Augmentation for Unsupervised Anomaly Detection with GAN

Icing on the Cake: An Easy and Quick Post-Learnig Method You Can Try After Deep Learning

3D conditional generative adversarial networks for high-quality PET image estimation at low dose

Automated Image Data Preprocessing with Deep Reinforcement Learning

Emotion Classification with Data Augmentation Using Generative Adversarial Networks

Improved Mixed-Example Data Augmentation

AutoAugment: Learning Augmentation Policies from Data

Data augmentation for improving deep learning in image classification problem

Label Refinery: Improving ImageNet Classification through Label Progression

Taskonomy: Disentangling Task Transfer Learning

Learning More Robust Features with Adversarial Training

Test-time augmentation with uncertainty estimation for deep learning-based medical image segmentation

MelanoGANs: High Resolution Skin Lesion Synthesis with GANs

GAN-based synthetic brain MR image generation

Simple random search provides a competitive approach to reinforcement learning

GAN-based Synthetic Medical Image Augmentation for increased CNN Performance in Liver Lesion Classification

Chest x-ray generation and data augmentation for cardiovascular abnormality classification

Regularized Evolution for Image Classifier Architecture Search

Data Augmentation by Pairing Samples for Images Classification

Learning implicit brain MRI manifolds with deep learning

Data Distillation: Towards Omni-Supervised Learning

A Rotation and a Translation Suffice: Fooling CNNs with Simple Transformations

The Effectiveness of Data Augmentation in Image Classification using Deep Learning

Are GANs Created Equal? A Large-Scale Study

Regularization for Deep Learning: A Taxonomy

Progressive Growing of GANs for Improved Quality, Stability, and Variation

How to fool radiologists with generative adversarial networks? A visual turing test for lung cancer diagnosis

One Pixel Attack for Fooling Deep Neural Networks

Many Paths to Equilibrium: GANs Do Not Need to Decrease a Divergence At Every Step

A systematic study of the class imbalance problem in convolutional neural networks

Retinal Vasculature Segmentation Using Local Saliency Maps and Generative Adversarial Networks For Image Super Resolution

Research on data augmentation for image classification based on convolution neural networks

Accelerated Magnetic Resonance Imaging by Adversarial Neural Network

Biomedical Data Augmentation Using Generative Adversarial Neural Networks

Improving Deep Learning using Generic Data Augmentation

Random Erasing Data Augmentation

Improved Regularization of Convolutional Neural Networks with Cutout

PatchShuffle Regularization

Revisiting Unreasonable Effectiveness of Data in Deep Learning Era

Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly

Super-Resolution via Deep Learning

Generative Adversarial Networks for Noise Reduction in Low-Dose CT

Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

Smart Augmentation Learning an Optimal Data Augmentation Strategy

Domain randomization for transferring deep neural networks from simulation to the real world

Evolution Strategies as a Scalable Alternative to Reinforcement Learning

Image Classification of Melanoma, Nevus and Seborrheic Keratosis by Deep Neural Network Ensemble

Large-Scale Evolution of Image Classifiers

A survey on deep learning in medical image analysis

Dataset Augmentation in Feature Space

Dermatologist-level classification of skin cancer with deep neural networks

Learning from Simulated and Unsupervised Images through Adversarial Training

StackGAN: Text to Photo-Realistic Image Synthesis with Stacked Generative Adversarial Networks

Least Squares Generative Adversarial Networks

Neural Architecture Search with Reinforcement Learning

Understanding Data Augmentation for Classification: When to Warp?

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

Densely Connected Convolutional Networks

Playing for Data: Ground Truth from Computer Games

Instance Normalization: The Missing Ingredient for Fast Stylization

Tutorial on Variational Autoencoders

Improved Techniques for Training GANs

Adversarial Feature Learning

A survey of transfer learning

One-shot Learning with Memory-Augmented Neural Networks