Kornia

Kornia is an open-source differentiable computer vision library built on top of PyTorch. It enables developers, researchers, and engineers to apply traditional computer vision algorithms as fully differentiable modules, making it easy to integrate with deep learning pipelines. Kornia provides a wide range of tools for image processing, geometry, feature detection, and augmentation, supporting both academic research and industrial applications.

• Built on top of PyTorch, enabling GPU acceleration and seamless integration with deep learning workflows.
• Differentiable implementations of classical computer vision algorithms.
• Modules for image filtering, geometry, morphology, and feature detection.
• Powerful data augmentation toolkit for training robust deep learning models.
• Easy to use, modular, and extensible for both research and production environments.
• Supports batch processing and parallelism for large-scale datasets.
• Open-source and actively maintained by the community.
• Lightweight, flexible, and scalable for small projects as well as enterprise-level systems.

To check the version of Kornia run the following commands:
# sudo su
# sudo apt update
# cd /opt
# source /opt/kornia-env/bin/activate
# pip show kornia | grep Version