/albumentations

Fast and flexible image augmentation library. Paper about the library: https://www.mdpi.com/2078-2489/11/2/125

Primary LanguagePythonMIT LicenseMIT

Albumentations

PyPI version CI PyPI Downloads Conda Downloads Stack Overflow License: MIT

Docs | Discord | Twitter | LinkedIn

Albumentations is a Python library for image augmentation. Image augmentation is used in deep learning and computer vision tasks to increase the quality of trained models. The purpose of image augmentation is to create new training samples from the existing data.

Here is an example of how you can apply some pixel-level augmentations from Albumentations to create new images from the original one: parrot

Why Albumentations

  • Albumentations supports all common computer vision tasks such as classification, semantic segmentation, instance segmentation, object detection, and pose estimation.
  • The library provides a simple unified API to work with all data types: images (RGB-images, grayscale images, multispectral images), segmentation masks, bounding boxes, and keypoints.
  • The library contains more than 70 different augmentations to generate new training samples from the existing data.
  • Albumentations is fast. We benchmark each new release to ensure that augmentations provide maximum speed.
  • It works with popular deep learning frameworks such as PyTorch and TensorFlow. By the way, Albumentations is a part of the PyTorch ecosystem.
  • Written by experts. The authors have experience both working on production computer vision systems and participating in competitive machine learning. Many core team members are Kaggle Masters and Grandmasters.
  • The library is widely used in industry, deep learning research, machine learning competitions, and open source projects.

Sponsors

Table of contents

Authors

Current Maintainer

Vladimir I. Iglovikov | Kaggle Grandmaster

Emeritus Core Team Members

Mikhail Druzhinin | Kaggle Expert

Alex Parinov | Kaggle Master

Alexander Buslaev | Kaggle Master

Eugene Khvedchenya | Kaggle Grandmaster

Installation

Albumentations requires Python 3.9 or higher. To install the latest version from PyPI:

pip install -U albumentations

Other installation options are described in the documentation.

Documentation

The full documentation is available at https://albumentations.ai/docs/.

A simple example

import albumentations as A
import cv2

# Declare an augmentation pipeline
transform = A.Compose([
    A.RandomCrop(width=256, height=256),
    A.HorizontalFlip(p=0.5),
    A.RandomBrightnessContrast(p=0.2),
])

# Read an image with OpenCV and convert it to the RGB colorspace
image = cv2.imread("image.jpg")
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Augment an image
transformed = transform(image=image)
transformed_image = transformed["image"]

Getting started

I am new to image augmentation

Please start with the introduction articles about why image augmentation is important and how it helps to build better models.

I want to use Albumentations for the specific task such as classification or segmentation

If you want to use Albumentations for a specific task such as classification, segmentation, or object detection, refer to the set of articles that has an in-depth description of this task. We also have a list of examples on applying Albumentations for different use cases.

I want to know how to use Albumentations with deep learning frameworks

We have examples of using Albumentations along with PyTorch and TensorFlow.

I want to explore augmentations and see Albumentations in action

Check the online demo of the library. With it, you can apply augmentations to different images and see the result. Also, we have a list of all available augmentations and their targets.

Who is using Albumentations

See also

List of augmentations

Pixel-level transforms

Pixel-level transforms will change just an input image and will leave any additional targets such as masks, bounding boxes, and keypoints unchanged. The list of pixel-level transforms:

Spatial-level transforms

Spatial-level transforms will simultaneously change both an input image as well as additional targets such as masks, bounding boxes, and keypoints. The following table shows which additional targets are supported by each transform.

Transform Image Mask BBoxes Keypoints
Affine
BBoxSafeRandomCrop
CenterCrop
CoarseDropout
Crop
CropAndPad
CropNonEmptyMaskIfExists
D4
ElasticTransform
GridDistortion
GridDropout
GridElasticDeform
HorizontalFlip
Lambda
LongestMaxSize
MaskDropout
Morphological
NoOp
OpticalDistortion
OverlayElements
PadIfNeeded
Perspective
PiecewiseAffine
PixelDropout
RandomCrop
RandomCropFromBorders
RandomGridShuffle
RandomResizedCrop
RandomRotate90
RandomScale
RandomSizedBBoxSafeCrop
RandomSizedCrop
Resize
Rotate
SafeRotate
ShiftScaleRotate
SmallestMaxSize
Transpose
VerticalFlip
XYMasking

A few more examples of augmentations

Semantic segmentation on the Inria dataset

inria

Medical imaging

medical

Object detection and semantic segmentation on the Mapillary Vistas dataset

vistas

Keypoints augmentation

Benchmarking results

System Information

  • Platform: macOS-15.0.1-arm64-arm-64bit
  • Processor: arm
  • CPU Count: 10
  • Python Version: 3.12.7

Benchmark Parameters

  • Number of images: 1000
  • Runs per transform: 10
  • Max warmup iterations: 1000

Library Versions

  • albumentations: 1.4.20
  • augly: 1.0.0
  • imgaug: 0.4.0
  • kornia: 0.7.3
  • torchvision: 0.20.0

Performance Comparison

Transform albumentations
1.4.20
augly
1.0.0
imgaug
0.4.0
kornia
0.7.3
torchvision
0.20.0
HorizontalFlip 8618 ± 1233 4807 ± 818 6042 ± 788 390 ± 106 914 ± 67
VerticalFlip 22847 ± 2031 9153 ± 1291 10931 ± 1844 1212 ± 402 3198 ± 200
Rotate 1146 ± 79 1119 ± 41 1136 ± 218 143 ± 11 181 ± 11
Affine 682 ± 192 - 774 ± 97 147 ± 9 130 ± 12
Equalize 892 ± 61 - 581 ± 54 152 ± 19 479 ± 12
RandomCrop80 47341 ± 20523 25272 ± 1822 11503 ± 441 1510 ± 230 32109 ± 1241
ShiftRGB 2349 ± 76 - 1582 ± 65 - -
Resize 2316 ± 166 611 ± 78 1806 ± 63 232 ± 24 195 ± 4
RandomGamma 8675 ± 274 - 2318 ± 269 108 ± 13 -
Grayscale 3056 ± 47 2720 ± 932 1681 ± 156 289 ± 75 1838 ± 130
RandomPerspective 412 ± 38 - 554 ± 22 86 ± 11 96 ± 5
GaussianBlur 1728 ± 89 242 ± 4 1090 ± 65 176 ± 18 79 ± 3
MedianBlur 868 ± 60 - 813 ± 30 5 ± 0 -
MotionBlur 4047 ± 67 - 612 ± 18 73 ± 2 -
Posterize 9094 ± 301 - 2097 ± 68 430 ± 49 3196 ± 185
JpegCompression 918 ± 23 778 ± 5 459 ± 35 71 ± 3 625 ± 17
GaussianNoise 166 ± 12 67 ± 2 206 ± 11 75 ± 1 -
Elastic 201 ± 5 - 235 ± 20 1 ± 0 2 ± 0
Clahe 454 ± 22 - 335 ± 43 94 ± 9 -
CoarseDropout 13368 ± 744 - 671 ± 38 536 ± 87 -
Blur 5267 ± 543 246 ± 3 3807 ± 325 - -
ColorJitter 628 ± 55 255 ± 13 - 55 ± 18 46 ± 2
Brightness 8956 ± 300 1163 ± 86 - 472 ± 101 429 ± 20
Contrast 8879 ± 1426 736 ± 79 - 425 ± 52 335 ± 35
RandomResizedCrop 2828 ± 186 - - 287 ± 58 511 ± 10
Normalize 1196 ± 56 - - 626 ± 40 519 ± 12
PlankianJitter 2204 ± 385 - - 813 ± 211 -

Contributing

To create a pull request to the repository, follow the documentation at CONTRIBUTING.md

https://github.com/albuemntations-team/albumentation/graphs/contributors

Community and Support

Comments

In some systems, in the multiple GPU regime, PyTorch may deadlock the DataLoader if OpenCV was compiled with OpenCL optimizations. Adding the following two lines before the library import may help. For more details pytorch/pytorch#1355

cv2.setNumThreads(0)
cv2.ocl.setUseOpenCL(False)

Citing

If you find this library useful for your research, please consider citing Albumentations: Fast and Flexible Image Augmentations:

@Article{info11020125,
    AUTHOR = {Buslaev, Alexander and Iglovikov, Vladimir I. and Khvedchenya, Eugene and Parinov, Alex and Druzhinin, Mikhail and Kalinin, Alexandr A.},
    TITLE = {Albumentations: Fast and Flexible Image Augmentations},
    JOURNAL = {Information},
    VOLUME = {11},
    YEAR = {2020},
    NUMBER = {2},
    ARTICLE-NUMBER = {125},
    URL = {https://www.mdpi.com/2078-2489/11/2/125},
    ISSN = {2078-2489},
    DOI = {10.3390/info11020125}
}