kornia 0.7.2

Open Source Differentiable Computer Vision Library for PyTorch

Stars: 9298, Watchers: 9298, Forks: 924, Open Issues: 257

The kornia/kornia repo was created 5 years ago and the last code push was 3 hours ago.
The project is extremely popular with a mindblowing 9298 github stars!

How to Install kornia

You can install kornia using pip

pip install kornia

or add it to a project with poetry

poetry add kornia

Package Details

Author

License

Apache-2.0

Homepage

PyPi:

https://pypi.org/project/kornia/

Documentation:

https://kornia.readthedocs.io/en/latest

GitHub Repo:

https://github.com/kornia/kornia

Classifiers

• Scientific/Engineering/Artificial Intelligence
• Scientific/Engineering/Image Processing
• Software Development/Libraries

Related Packages

Errors

A list of common kornia errors.

Code Examples

Here are some kornia code examples and snippets.

GitHub Issues

The kornia package has 257 open issues on GitHub

• RandomMotionBlur fails when kernel_size is a Tuple[int, int]
• AugmentationSequential manipulates BBoxes input
• update the histogram_matching function
• The RandomPosterize class raises RuntimeError because of conflicting devices.
• torchgeometry old code migration
• Support list input
• Add antialias to warp_perspective
• Fix bounding box return type in AugmentationSequential
• [feat] Adding cropping option for SSIM
• Support for object Detection pipeline and Augmentations
• New feature: SOLD2 line detector and descriptor
• Some Gaussian filter unit tests do not properly test equivalence of functionality
• LoFTR JIT Support
• WIP: Store GaussianBlur2d kernel as buffer
• Deprecate return_transform in augmentations

See more issues on GitHub