Albumentations
Fast and flexible image augmentation library with 70+ transforms for computer vision model training.
Albumentations
Albumentations is the most widely used image augmentation library in computer vision. It provides a composable, framework-agnostic API for 70+ pixel-level and spatial transforms, all implemented in highly optimised C++/SIMD under the hood for maximum throughput during training.
Key Features
- 70+ transforms: flips, rotations, crops, colour jitter, blur, noise, elastic distortions, perspective, and domain-specific weather/medical effects
- Multi-target: augment images, masks, bounding boxes, and keypoints in sync with a single pipeline call
- Framework agnostic: integrates with PyTorch, Keras/TF, JAX, FastAI, and plain NumPy
- Blazing fast: built on OpenCV C++ backend; 40-50× faster than torchvision for many transforms
- Serialisable: save and load augmentation pipelines as JSON/YAML for reproducibility
- AutoAugment & RandAugment: policy-based augmentation search strategies included
Quick Start
pip install albumentations
import albumentations as A
import cv2
transform = A.Compose([
A.HorizontalFlip(p=0.5),
A.RandomBrightnessContrast(p=0.2),
A.ShiftScaleRotate(shift_limit=0.05, scale_limit=0.1, rotate_limit=15, p=0.5),
A.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
])
image = cv2.imread("image.jpg")
augmented = transform(image=image)["image"]
npx ai-supply add albumentations-image-augmentation
Curated mirror of the open-source Albumentations (MIT). Get it from the source.
Compromise signals — malicious or tampered code (leaked secrets, backdoors, a dropped executable) — reduce the score, and known dependency CVEs carry a bounded penalty (they warrant review but never QUARANTINE — update the dependency to clear). Other dangerous-by-capability traits are risk surface, expected for some capabilities. Every finding is mapped to its OWASP control below.
Findings mapped to the OWASP Top 10 for LLM Applications (2025) and the OWASP Machine Learning Security Top 10. Expand any flagged control for the exact findings — compromise reduces the score; expected/risk-surface do not, except a known CVE, which carries a small bounded penalty (high/critical → Review).
The same gate an agent runs before installing (POST /api/v1/trust/albumentations-image-augmentation/check). Click a policy:
Consume Albumentations programmatically. Authenticate with an API key or session — see Authorize an agent.
# Agents: CHECK BEFORE YOU INSTALL (no auth) — score, grade, level, capability manifest
curl https://ai-supply.store/api/v1/trust/albumentations-image-augmentation
# Gate against your org policy (returns { pass, violations })
curl -X POST https://ai-supply.store/api/v1/trust/albumentations-image-augmentation/check \
-H "Content-Type: application/json" \
-d '{"minGrade":"B","denyPermissions":["shell"],"denyUnknownEgress":true}'
# CLI
npx ai-supply add albumentations-image-augmentation
# REST (install → download)
curl -X POST https://ai-supply.store/api/v1/listings/albumentations-image-augmentation/install \
-H "Authorization: Bearer $AIM_KEY"
# MCP tool
install_listing({ "slug": "albumentations-image-augmentation" })OpenAPI spec →Curated mirror — latest upstream source. See the repository for tagged releases.