---
id: albumentations
version: "2.0.8"
license: MIT License Copyright (c) 2017 Vladimir Iglovikov, Alexander Buslaev, Alexander Parinov, Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated… (full text in the JSON record)
license_treatment: permissive
maintenance: aging
---

# albumentations — Fast, flexible, and advanced augmentation library for deep learning, computer vision, and medical imaging. Albumentations offers a wide range of transformations for both 2D (images, masks, bboxes, keypoints) and 3D (volumes, volumetric masks, keypoints) data, with optimized performance and seamless integration into ML workflows.

License: permissive · Maintenance: aging · Downloads: 5.5M/mo

## What it is and what it does

Albumentations is a Python library for augmenting training data in computer vision and deep learning tasks. It provides a unified API to apply transformations to images, masks, bounding boxes, and keypoints—supporting classification, semantic and instance segmentation, object detection, and pose estimation. The library separates pixel-level transforms from spatial transforms and applies them consistently across all data types in a single pipeline.

The package depends on numpy, scipy, PyYAML, pydantic, albucore, opencv-python-headless, typing-extensions, and eval-type-backport. It is classified as Production/Stable and requires Python 3.9 or higher. The last release was 444 days ago.

Use it for:

- Augment training datasets for image classification to improve model generalization without collecting new labeled data.
- Apply consistent transformations to images and bounding boxes for object detection model training.
- Generate varied training samples for semantic segmentation by transforming both images and pixel-level masks.
- Augment keypoint annotations alongside images for pose estimation and human pose recognition tasks.
- Preprocess medical imaging data with specialized augmentations for anomaly detection and diagnostic tasks.
- Create synthetic training variations for autonomous driving datasets with coordinated image and bounding-box transforms.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Albumentations applies image transformations to training data, supporting classification, segmentation, object detection, and pose estimation with a unified API for images, masks, bounding boxes, and keypoints.

Yes. Albumentations is a mature, permissively licensed library with low install friction and no known vulnerabilities. The aging maintenance status (444 days since release) is a minor concern but does not block use for stable augmentation pipelines. Install it if you need a unified, production-grade augmentation API for computer vision tasks.

## Install

pip install albumentations
uv add albumentations
poetry add albumentations

## Installing albumentations

Before you install: Low friction install with 8 runtime dependencies including numpy, scipy, and opencv-python-headless. Maintenance status is aging (444 days since last release), though the package is marked Production/Stable and supports Python 3.9, 3.10, 3.11, 3.12, 3.13.

License in practice: MIT License permits unrestricted use, modification, and distribution with minimal restrictions—suitable for both open-source and commercial projects.

Quickstart:

pip install albumentations

import albumentations as A

transform = A.Compose([
    A.RandomCrop(width=256, height=256),
    A.HorizontalFlip(p=0.5),
])

transformed = transform(image=image)
transformed_image = transformed["image"]

Requires Python 3.9 or higher.

Verify before relying:

- Exact number of augmentation transforms available (description mentions '70+' but version-specific count unknown)
- Performance benchmarking details and comparison methodology against other libraries
- Whether all advertised framework integrations are actively maintained in version 2.0.8

## Package facts

- License: MIT License Copyright (c) 2017 Vladimir Iglovikov, Alexander Buslaev, Alexander Parinov, Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated… (full text in the JSON record) (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: aging
- Downloads: 5.5M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

image augmentation library, data augmentation deep learning, computer vision transforms, image preprocessing pipeline, bounding box augmentation, semantic segmentation augmentation, keypoint detection transforms, image-augmentation, computer-vision, deep-learning

[View on SkillFed](https://skillfed.io/packages/albumentations) · [View on PyPI](https://pypi.org/project/albumentations/)