State-of-the-art and curated Computer Vision algorithms for AI.

Kornia AI is on the mission to leverage and democratize the next generation of Computer Vision tools and Deep Learning libraries within the context of an Open Source community.

>>> import kornia.geometry as K
>>> registrator = K.ImageRegistrator('similarity')
>>> model = registrator.register(img1, img2)

Ready to use with state-of-the art Deep Learning models:

DexiNed edge detection model.

import kornia
from kornia.contrib import EdgeDetectorBuilder

image = kornia.io.get_sample_images()[0][None]
model = EdgeDetectorBuilder.build()
model.save(image)

RTDETRDetector for object detection.

import kornia
from kornia.contrib import RTDETRDetectorBuilder

image = kornia.io.get_sample_images()[0][None]
model = RTDETRDetectorBuilder.build()
model.save(image)

BoxMotTracker for object tracking.

import kornia
from kornia.contrib import BoxMotTracker

image = kornia.io.get_sample_images()[0][None]
model = BoxMotTracker()
for i in range(4):
   model.update(image)
model.save(image)

Vision Transformer for image classification.

>>> import torch.nn as nn
>>> from kornia.models.vit import VisionTransformer
>>> classifier = nn.Sequential(
...   VisionTransformer(image_size=224, patch_size=16),
...   nn.Linear(768, 1000),  # Example: 768 is the default hidden_dim, 1000 is num_classes
... )
>>> logits = classifier(img)    # BxN
>>> scores = logits.argmax(-1)  # B

Multi-framework support

You can now use Kornia with NumPy, TensorFlow, and JAX.

>>> import kornia
>>> tf_kornia = kornia.to_tensorflow()

Powered by

Join the community

• Join our social network communities with 1.8k+ members:

  • Twitter: we share recent research and news with our community.

  • Slack: come to us and chat with our engineers and mentors to get support and resolve your questions.

• Subscribe to our YouTube channel to get the latest video demos.

---