Understand the project
Practical overview
When it fits
Use Kornia when a PyTorch-based vision system needs reusable differentiable image operations, augmentation or geometric-vision components that can be integrated into an application or training pipeline.
Who it is for
- Python developers building computer-vision or image-analysis applications
- PyTorch practitioners adding trainable image operations to machine-learning pipelines
- Robotics and spatial-AI teams working with geometric vision
- Researchers and educators prototyping vision workflows
Documented uses
- Build differentiable image-processing pipelines
- Add geometric transformations and augmentation to model training
- Implement camera, stereo, depth and 3D-vision components
- Use feature detectors, descriptors and matchers in vision systems
- Combine documented vision operations with custom PyTorch neural networks
Before you choose
Requirements & limits
The repository is public, open source and licensed under Apache-2.0. The documented technical context identifies Python and PyTorch, with Linux, Windows, macOS, cloud and self-hosted environments listed.
- PyTorch integration is central to the library’s documented design.
- Operations are differentiable and support batch transformations, automatic differentiation and GPU acceleration.
- The documented scope spans basic image processing, augmentation, geometric vision and selected AI models.
- The repository links to documentation, tutorials, examples and a browser-based playground for further exploration.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Evidence and transparency
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 2 October 2026
- Original website title
- kornia/kornia · Official homepage title · retained in its source language