Albumentations: Fast and Flexible Image Augmentations
Albumentations: быстрые и гибкие преобразования изображений
2020-02-24
SCID: 54.1/bmd2yat9
Discuss with AI
Albumentationscomputer visiondeep learning regularizationimage augmentationimage processing speed
Figures from the paper
Abstract (AI)
Data augmentation is a commonly used technique for increasing both the size and the diversity of labeled training sets by leveraging input transformations that preserve corresponding output labels. In computer vision, image augmentations have become a common implicit regularization technique to combat overfitting in deep learning models and are ubiquitously used to improve performance. While most deep learning frameworks implement basic image transformations, the list is typically limited to some variations of flipping, rotating, scaling, and cropping. Moreover, image processing speed varies in existing image augmentation libraries. We present Albumentations, a fast and flexible open source library for image augmentation with many various image transform operations available that is also an easy-to-use wrapper around other augmentation libraries. We discuss the design principles that drove the implementation of Albumentations and give an overview of the key features and distinct capabilities. Finally, we provide examples of image augmentations for different computer vision tasks and demonstrate that Albumentations is faster than other commonly used image augmentation tools on most image transform operations.
Key Findings
1
Albumentations is designed for flexibility and supports augmentation examples across multiple computer vision tasks.
2
Albumentations is introduced as an open-source image augmentation library offering a broad range of transformation operations for computer vision.
3
Benchmarking indicates that Albumentations is faster than commonly used image augmentation tools for most image transformation operations.
4
The library provides an easy-to-use wrapper around other image augmentation libraries, extending augmentation functionality beyond standard framework transformations.
5
The work presents implementation design principles and key capabilities intended to improve augmentation usability and processing speed.
Research Object
Albumentations, an open-source image augmentation library
Research Subject
the library’s speed, flexibility, transform-operation coverage, design principles, and comparative performance for computer vision image augmentation
Publication Details
Publication Date
2020-02-24
Journal
Publisher
ISSN
Cited by
2693
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai6
ImageNet classification with deep convolutional neural networks2017
ImageNet: A large-scale hierarchical image database2009
Gradient-based learning applied to document recognition1998
PyTorch: An Imperative Style, High-Performance Deep Learning Library2019
A survey on Image Data Augmentation for Deep Learning2019
Improving Deep Learning with Generic Data Augmentation2018
Cited by5
A survey on Image Data Augmentation for Deep Learning2019
Brain Tumor Detection Based on Deep Learning Approaches and Magnetic Resonance Imaging2023
Deep convolutional neural network for automatic fault recognition from 3D seismic datasets2021
EmbryoNet: using deep learning to link embryonic phenotypes to signaling pathways2023
VEPL Dataset: A Vegetation Encroachment in Power Line Corridors Dataset for Semantic Segmentation of Drone Aerial Orthomosaics2023