mahmoudnafifi/WB_color_augmenter

WB color augmenter improves the accuracy of image classification and image semantic segmentation methods by emulating different WB effects (ICCV 2019) [Python & Matlab].

47
/ 100
Emerging

This tool helps improve the accuracy of image classification and semantic segmentation systems by generating variations of input images that realistically emulate different camera white balance settings. It takes original images and outputs multiple augmented versions, each with a distinct but plausible color cast. This is useful for anyone training deep learning models on image data who wants to make their models more robust to real-world lighting and camera conditions.

175 stars. No commits in the last 6 months.

Use this if you are training image recognition models and want to make them less sensitive to color shifts caused by incorrect white balance settings in cameras.

Not ideal if your image analysis task does not involve color-sensitive deep learning models or if you need to augment images with unrealistic or extreme color transformations.

image-classification computer-vision deep-learning-training image-processing color-correction
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

175

Forks

36

Language

MATLAB

License

MIT

Category

image-inpainting

Last pushed

Sep 20, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/mahmoudnafifi/WB_color_augmenter"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.