Albumentations: fast and flexible image augmentations

Abstract

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 output labels. In computer vision domain, image augmentations have become a common implicit regularization technique to combat overfitting in deep convolutional neural networks and are ubiquitously used to improve performance. While most deep learning frameworks implement basic image transformations, the list is typically limited to some variations and combinations of flipping, rotating, scaling, and cropping. Moreover, the image processing speed varies in existing tools for image augmentation. We present Albumentations, a fast and flexible library for image augmentations with many various image transform operations available, that is also an easy-to-use wrapper around other augmentation libraries. We provide examples of image augmentations for different computer vision tasks and show that Albumentations is faster than other commonly used image augmentation tools on the most of commonly used image transformations. The source code for Albumentations is made publicly available online at https://github.com/albu/albumentations

References in corpus (12)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• A Survey on Deep Learning in Medical Image Analysis
• The NumPy array: a structure for efficient numerical computation
• CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
• AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
• Smart Augmentation - Learning an Optimal Data Augmentation Strategy
• Automatic Instrument Segmentation in Robot-Assisted Surgery Using Deep Learning
• Learning to Compose Domain-Specific Transformations for Data Augmentation
• Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules
• Further advantages of data augmentation on convolutional neural networks
• A Bayesian Data Augmentation Approach for Learning Deep Models
• Angiodysplasia Detection and Localization Using Deep Convolutional Neural Networks

Cited by in corpus (88)

• A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS
• TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning
• Deepfake Detection by Human Crowds, Machines, and Machine-informed Crowds
• A Survey on Deep Learning for Skin Lesion Segmentation
• Using DUCK-Net for Polyp Image Segmentation
• Fracture Detection in Wrist X-ray Images Using Deep Learning-Based Object Detection Models
• BS-Net: learning COVID-19 pneumonia severity on a large Chest X-Ray dataset
• SinGAN-Seg: Synthetic training data generation for medical image segmentation
• Robust Multi-Task Learning and Online Refinement for Spacecraft Pose Estimation across Domain Gap
• Deep Learning-based Spacecraft Relative Navigation Methods: A Survey
• 3D Convolutional Neural Networks for Stalled Brain Capillary Detection
• Robust deep learning-based semantic organ segmentation in hyperspectral images
• Segmentation of cell-level anomalies in electroluminescence images of photovoltaic modules
• Exchanging Dual Encoder-Decoder: A New Strategy for Change Detection with Semantic Guidance and Spatial Localization
• A Convolutional-Transformer Network for Crack Segmentation with Boundary Awareness
• DAugNet: Unsupervised, Multi-source, Multi-target, and Life-long Domain Adaptation for Semantic Segmentation of Satellite Images
• InsPLAD: A Dataset and Benchmark for Power Line Asset Inspection in UAV Images
• Vision Transformer based COVID-19 Detection using Chest X-rays
• Image Compositing for Segmentation of Surgical Tools without Manual Annotations
• Densely-Populated Traffic Detection using YOLOv5 and Non-Maximum Suppression Ensembling
• Ten Quick Tips for Deep Learning in Biology
• Self-supervised Audiovisual Representation Learning for Remote Sensing Data
• Deep Learning Methods for Screening Pulmonary Tuberculosis Using Chest X-rays
• Season-invariant GNSS-denied visual localization for UAVs
• Vehicle-Rear: A New Dataset to Explore Feature Fusion for Vehicle Identification Using Convolutional Neural Networks
• Danish Fungi 2020 -- Not Just Another Image Recognition Dataset
• Mutually improved endoscopic image synthesis and landmark detection in unpaired image-to-image translation
• GaNDLF: A Generally Nuanced Deep Learning Framework for Scalable End-to-End Clinical Workflows in Medical Imaging
• Multi-Label Contrastive Learning for Abstract Visual Reasoning
• An original framework for Wheat Head Detection using Deep, Semi-supervised and Ensemble Learning within Global Wheat Head Detection (GWHD) Dataset
• Advanced computer vision for extracting georeferenced vehicle trajectories from drone imagery
• Image-based Automatic Dial Meter Reading in Unconstrained Scenarios
• Domain-guided data augmentation for deep learning on medical imaging
• Forensic Analysis of Synthetically Generated Western Blot Images
• Segmenting Medical Images with Limited Data
• Multi-modal land cover mapping of remote sensing images using pyramid attention and gated fusion networks
• Autonomous Coordinated Control of the Light Guide for Positioning in Vitreoretinal Surgery
• Navigation-Oriented Scene Understanding for Robotic Autonomy: Learning to Segment Driveability in Egocentric Images
• A First Look at Dataset Bias in License Plate Recognition
• Esophageal virtual disease landscape using mechanics-informed machine learning
• Evaluating Generic Auto-ML Tools for Computational Pathology
• Macroscale fracture surface segmentation via semi-supervised learning considering the structural similarity
• Boosting EfficientNets Ensemble Performance via Pseudo-Labels and Synthetic Images by pix2pixHD for Infection and Ischaemia Classification in Diabetic Foot Ulcers
• DFGC 2021: A DeepFake Game Competition
• SCSS-Net: Solar Corona Structures Segmentation by Deep Learning
• Anatomically aware dual-hop learning for pulmonary embolism detection in CT pulmonary angiograms
• Rethinking Cooking State Recognition with Vision Transformers
• Online Supervised Training of Spaceborne Vision during Proximity Operations using Adaptive Kalman Filtering
• MedAugment: Universal Automatic Data Augmentation Plug-in for Medical Image Analysis
• Spectral decoupling allows training transferable neural networks in medical imaging
• ODOR: The ICPR2022 ODeuropa Challenge on Olfactory Object Recognition
• Multi-scale attention-based instance segmentation for measuring crystals with large size variation
• ObjectMix: Data Augmentation by Copy-Pasting Objects in Videos for Action Recognition
• A Multimodal Hybrid Late-Cascade Fusion Network for Enhanced 3D Object Detection
• Image segmentation of treated and untreated tumor spheroids by Fully Convolutional Networks
• Virtual Underwater Datasets for Autonomous Inspections
• Improving Small Drone Detection Through Multi-Scale Processing and Data Augmentation
• AngioMoCo: Learning-based Motion Correction in Cerebral Digital Subtraction Angiography
• VascX Models: Model Ensembles for Retinal Vascular Analysis from Color Fundus Images
• Advancing Multinational License Plate Recognition Through Synthetic and Real Data Fusion: A Comprehensive Evaluation
• AGILE: Approach-based Grasp Inference Learned from Element Decomposition
• Light-weight Document Image Cleanup using Perceptual Loss
• Prototype Memory for Large-scale Face Representation Learning
• Cannabis Seed Variant Detection using Faster R-CNN
• Weakly Supervised Object Detection in Chest X-Rays with Differentiable ROI Proposal Networks and Soft ROI Pooling
• DualStreamFoveaNet: A Dual Stream Fusion Architecture with Anatomical Awareness for Robust Fovea Localization
• UL-VIO: Ultra-lightweight Visual-Inertial Odometry with Noise Robust Test-time Adaptation
• Simulation of Charge Stability Diagrams for Automated Tuning Solutions (SimCATS)
• Enhancing kelp forest detection in remote sensing images using crowdsourced labels with Mixed Vision Transformers and ConvNeXt segmentation models
• An Enhanced Harmonic Densely Connected Hybrid Transformer Network Architecture for Chronic Wound Segmentation Utilising Multi-Colour Space Tensor Merging
• Self-Supervised Representation Learning for Nerve Fiber Distribution Patterns in 3D-PLI
• Multiclass Semantic Segmentation to Identify Anatomical Sub-Regions of Brain and Measure Neuronal Health in Parkinson's Disease
• Synthetic Data Generation for Augmenting Small Samples
• Lacking Data? No worries! How synthetic images can alleviate image scarcity in wildlife surveys: a case study with muskox (Ovibos moschatus)
• Landmarks Augmentation with Manifold-Barycentric Oversampling
• Clarifying Myths About the Relationship Between Shape Bias, Accuracy, and Robustness
• SwiftSRGAN -- Rethinking Super-Resolution for Efficient and Real-time Inference
• Comprehensive Evaluation of Prototype Neural Networks
• DACov: A Deeper Analysis of Data Augmentation on the Computed Tomography Segmentation Problem
• What augmentations are sensitive to hyper-parameters and why?
• Employing similarity to highlight differences: On the impact of anatomical assumptions in chest X-ray registration methods
• Leveraging Modified Ex Situ Tomography Data for Segmentation of In Situ Synchrotron X-Ray Computed Tomography
• Collaborative real-time vision-based device for olive oil production monitoring
• RedDino: A foundation model for red blood cell analysis
• Contrastive meta-domain adaptation for robust skin lesion classification across clinical and acquisition conditions
• deepNIR: Datasets for generating synthetic NIR images and improved fruit detection system using deep learning techniques
• Embedding Shift Dissection on CLIP: Effects of Augmentations on VLM's Representation Learning
• Measurement of Hybrid Rocket Solid Fuel Regression Rate for a Slab Burner using Deep Learning