Albumentations to PIL/Pillow Transform Mapping

On this page

• Channel and Target Caveats
• Direct Pillow Benchmark Mappings
• No Direct Pillow Benchmark Implementation
• Migration Pattern
• Benchmark Pages

This page maps Albumentations transforms to Pillow operations where Pillow has a direct benchmark implementation. The source of truth for supported Pillow benchmark transforms is benchmark/transforms/pillow_impl.py; shared parameter names come from benchmark/transforms/specs.py.

If the Pillow column is -, Pillow has no direct analogue in the benchmark. The benchmark intentionally does not recreate Albumentations behavior with custom NumPy, ImageDraw, or handwritten target-propagation glue.

Channel and Target Caveats

Pillow operations are built around image modes such as L, RGB, and RGBA. Many color operations either convert to grayscale/RGB or assume RGB-style channels. Albumentations operates on NumPy arrays, so many image-only transforms can preserve arbitrary channel counts. RGB/color-space transforms still have channel-specific constraints.

Pillow transforms do not update masks, bounding boxes, keypoints, rotated boxes, volumes, or videos. Albumentations transforms are pipeline operations and can update supported targets in the same call.

Pillow also does not provide augmentation-policy containers comparable to Albumentations Compose, OneOf, SomeOf, or RandomOrder. Combining crop, flip, normalization, and optional branches means writing the policy as Python control flow. In Albumentations, the policy is a serializable pipeline object.

Direct Pillow Benchmark Mappings

No Direct Pillow Benchmark Implementation

Migration Pattern

from PIL import Image
import numpy as np
import albumentations as A

image = np.array(Image.open("image.jpg").convert("RGB"))

pipeline = A.Compose([
    A.HorizontalFlip(p=0.5),
    A.Rotate(angle_range=(-45, 45), border_mode=0, fill=0, p=0.5),
    A.OneOf([
        A.GaussianBlur(blur_range=(3, 7), p=1.0),
        A.GaussNoise(std_range=(0.05, 0.2), p=1.0),
    ], p=0.3),
    A.RandomBrightnessContrast(brightness_range=(-0.2, 0.2), contrast_range=(-0.2, 0.2), p=0.5),
])

augmented = pipeline(image=image)["image"]

For non-RGB data, keep the array shape you actually train on and avoid RGB-only transforms unless the conversion is intentional.

Because this is an Albumentations pipeline object, you can serialize it and reuse the same augmentation definition outside the notebook or training script.

Benchmark Pages

• PIL/Pillow benchmark route
• Benchmark transform implementations