kornia
Kornia is a derivative computerized vision library based on PyTorch. Mass conversions integrate into existing AI workflows with automatic deriving and GPU acceleration support. It offers geometric vision algorithms, image processing operators, and data boosting tools.
KO
kornia
100kornia · Image & Photo Processing
STARS11.5K
FORKS1.4K
COMMITS3.6K
CONTRIBUTORS403
VERSIONv0.8.3
UPDATEDtoday
PythonPyTorchComputer VisionImage ProcessingApache-2.0
Common commercial products in this category: Adobe Photoshop, Adobe Lightroom, Cloudinary
What can it do?
- You incorporate derivative image processing operators such as filtering, conversion, and edge detection into deep learning lines.
- To train AI models, you run data boost pipelines that contain color, noise, and geometric distortions.
- You use pre-trained AI models for a variety of vision tasks, such as facial detection, feature matching, and segmentation.