imgaug and image_augmentor
About imgaug
aleju/imgaug
Image augmentation for machine learning experiments.
This tool helps machine learning engineers and researchers expand their image datasets by creating many altered versions of original images. You feed it a set of input images, along with associated data like heatmaps, segmentation maps, keypoints, or bounding boxes. It then outputs a much larger collection of subtly modified images and their corresponding updated annotations, making your models more robust.
About image_augmentor
codebox/image_augmentor
Data augmentation tool for images
This tool helps machine learning practitioners expand their image datasets. You provide a directory of existing image files, and it generates new, modified versions of those images using operations like flips, rotations, and noise. The output is a larger collection of diverse images, which helps improve the performance of neural networks, especially when your initial dataset is small. Researchers and data scientists who train image-based AI models would find this valuable.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work