Albumentations

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:

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 (RBG-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.

Authors

Vladimir I. Iglovikov | Kaggle Grandmaster

Mikhail Druzhinin | Kaggle Expert

Alex Parinov | Kaggle Master

Alexander Buslaev — Computer Vision Engineer at Mapbox | Kaggle Master

Evegene Khvedchenya — Computer Vision Research Engineer at Piñata Farms | Kaggle Grandmaster

Installation

Albumentations requires Python 3.8 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

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	✓	✓	✓	✓
ElasticTransform	✓	✓	✓
Flip	✓	✓	✓	✓
GridDistortion	✓	✓	✓
GridDropout	✓	✓
HorizontalFlip	✓	✓	✓	✓
Lambda	✓	✓	✓	✓
LongestMaxSize	✓	✓	✓	✓
MaskDropout	✓	✓
NoOp	✓	✓	✓	✓
OpticalDistortion	✓	✓	✓
PadIfNeeded	✓	✓	✓	✓
Perspective	✓	✓	✓	✓
PiecewiseAffine	✓	✓	✓	✓
PixelDropout	✓	✓
RandomCrop	✓	✓	✓	✓
RandomCropFromBorders	✓	✓	✓	✓
RandomCropNearBBox	✓	✓	✓	✓
RandomGridShuffle	✓	✓		✓
RandomResizedCrop	✓	✓	✓	✓
RandomRotate90	✓	✓	✓	✓
RandomScale	✓	✓	✓	✓
RandomSizedBBoxSafeCrop	✓	✓	✓
RandomSizedCrop	✓	✓	✓	✓
Resize	✓	✓	✓	✓
Rotate	✓	✓	✓	✓
SafeRotate	✓	✓	✓	✓
ShiftScaleRotate	✓	✓	✓	✓
SmallestMaxSize	✓	✓	✓	✓
Transpose	✓	✓	✓	✓
VerticalFlip	✓	✓	✓	✓
XYMasking	✓	✓		✓

Mixing-level transforms

Transforms that mix several images into one

Transform	Image	Mask	BBoxes	Keypoints	Global Label
MixUp	✓	✓			✓

A few more examples of augmentations

Semantic segmentation on the Inria dataset

Medical imaging

Object detection and semantic segmentation on the Mapillary Vistas dataset

Keypoints augmentation

Benchmarking results

To run the benchmark yourself, follow the instructions in benchmark/README.md

Results for running the benchmark on the first 2000 images from the ImageNet validation set using an AMD Ryzen Threadripper 3970X CPU. The table shows how many images per second can be processed on a single core; higher is better.

Library	Version
Python	3.10.13 (main, Sep 11 2023, 13:44:35) [GCC 11.2.0]
albumentations	1.4.1
imgaug	0.4.0
torchvision	0.17.1+rocm5.7
numpy	1.26.4
opencv-python-headless	4.9.0.80
scikit-image	0.22.0
scipy	1.12.0
pillow	10.2.0
kornia	0.7.2
augly	1.0.0

	albumentations 1.4.0	torchvision 0.17.1+rocm5.7	kornia 0.7.2	augly 1.0.0	imgaug 0.4.0
HorizontalFlip	9843 ± 2135	2436 ± 29	1014 ± 3	3663 ± 18	4884 ± 51
VerticalFlip	9898 ± 18	2570 ± 37	1024 ± 4	5325 ± 13	8683 ± 5
Rotate	610 ± 4	153 ± 2	204 ± 1	626 ± 3	499 ± 5
Affine	1705 ± 67	159 ± 1	200 ± 1	-	663 ± 24
Equalize	1061 ± 14	337 ± 1	77 ± 1	-	845 ± 33
RandomCrop64	203197 ± 2105	15931 ± 27	837 ± 2	21858 ± 362	5681 ± 96
RandomResizedCrop	2998 ± 30	1160 ± 4	190 ± 1	-	-
ShiftRGB	1400 ± 3	-	435 ± 1	-	1528 ± 6
Resize	2581 ± 3	1239 ± 1	197 ± 1	431 ± 1	1728 ± 1
RandomGamma	4556 ± 3	230 ± 1	205 ± 1	-	2282 ± 110
Grayscale	7234 ± 4	1539 ± 7	444 ± 3	2606 ± 2	918 ± 42
ColorJitter	452 ± 43	51 ± 1	50 ± 1	221 ± 1	-
RandomPerspective	465 ± 1	121 ± 1	115 ± 1	-	433 ± 16
GaussianBlur	2315 ± 9	106 ± 2	72 ± 1	161 ± 1	1213 ± 3
MedianBlur	3711 ± 2	-	2 ± 1	-	566 ± 3
MotionBlur	2763 ± 25	-	101 ± 4	-	508 ± 2
Posterize	4238 ± 51	2581 ± 20	284 ± 4	-	1893 ± 9
JpegCompression	208 ± 1	-	-	692 ± 4	435 ± 1
GaussianNoise	64 ± 9	-	-	67 ± 1	212 ± 16
Elastic	129 ± 1	3 ± 1	1 ± 1	-	128 ± 1

Contributing

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

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}
}

joshkotrous/albumentations