Kornia Models#

Kornia ships a curated set of model architectures, each wrapped as a regular nn.Module. Most of them come with pretrained weights that are downloaded on first use; MobileViT ships as an architecture only, for training or for loading your own checkpoint. Each page below opens with the shortest script that runs the model, shows what it returns on a real input, and ends with the paper. The builders that construct the models are documented on the kornia.models API page, and the local-feature models under kornia.feature.

Model

Task

Venue

Licence

RT-DETR

Object detection

arXiv 2023

Apache-2.0

YuNet

Face detection

IEEE TIP 2021

Apache-2.0

Segment Anything (SAM)

Promptable segmentation

ICCV 2023

Apache-2.0

MobileSAM

Promptable segmentation (lightweight)

arXiv 2023

Apache-2.0

EfficientViT

Segmentation, classification, detection backbone

ICCV 2023

Apache-2.0

Vision Transformer (ViT)

Image classification backbone

ICLR 2021

Apache-2.0

MobileViT

Image classification backbone (mobile)

ICLR 2022

TinyViT

Image classification backbone (small)

ECCV 2022

LoFTR

Image matching (detector-free)

CVPR 2021

Apache-2.0

HardNet

Local feature descriptor

NeurIPS 2017

MIT

AffNet

Affine shape estimation for local features

ECCV 2018

MIT

SOLD2

Line detection and matching

CVPR 2021

MIT

DexiNed

Edge detection

WACV 2020

MIT

DeFMO

Enhancement – video deblurring of fast-moving objects

CVPR 2021

Apache-2.0