Kornia — Geometric Computer Vision Library
Differentiable computer vision library built on PyTorch: geometry, augmentation, colour, filtering, feature extraction, and more.
Kornia
Kornia is a differentiable computer vision library on top of PyTorch. Every operation — from homography estimation to image augmentation — is implemented as a differentiable module, making it ideal for learning-based vision pipelines and end-to-end training.
Key Features
- Geometry: camera models, homographies, epipolar geometry, pose estimation
- Image processing: filtering, colour space conversion, morphology, edge detection
- Augmentation: photometric and geometric transforms for training data
- Feature extraction: SIFT, LoFTR, DISK, KeyNet, SOLD2 descriptors
- 3D vision: depth, point clouds, quaternions, rotation groups (SO3/SE3)
- Fully differentiable — gradients flow through every op
Quick Start
import torch
import kornia
import kornia.augmentation as K
# Differentiable augmentation pipeline
aug = K.AugmentationSequential(
K.RandomHorizontalFlip(p=0.5),
K.ColorJitter(0.2, 0.2, 0.2, 0.2),
K.RandomGaussianBlur((3, 3), (0.1, 2.0)),
data_keys=["input", "bbox"],
)
tensor = torch.rand(2, 3, 256, 256)
out = aug(tensor)
Install via ai-supply
npx ai-supply add kornia-geometric-vision-library
Curated mirror of the open-source Kornia (Apache-2.0). Get it from the source.
Compromise signals — malicious or tampered code (leaked secrets, backdoors, a dropped executable) — reduce the score, and known dependency CVEs carry a bounded penalty (they warrant review but never QUARANTINE — update the dependency to clear). Other dangerous-by-capability traits are risk surface, expected for some capabilities. Every finding is mapped to its OWASP control below.
Findings mapped to the OWASP Top 10 for LLM Applications (2025) and the OWASP Machine Learning Security Top 10. Expand any flagged control for the exact findings — compromise reduces the score; expected/risk-surface do not, except a known CVE, which carries a small bounded penalty (high/critical → Review).
The same gate an agent runs before installing (POST /api/v1/trust/kornia-geometric-vision-library/check). Click a policy:
Consume Kornia — Geometric Computer Vision Library programmatically. Authenticate with an API key or session — see Authorize an agent.
# Agents: CHECK BEFORE YOU INSTALL (no auth) — score, grade, level, capability manifest
curl https://ai-supply.store/api/v1/trust/kornia-geometric-vision-library
# Gate against your org policy (returns { pass, violations })
curl -X POST https://ai-supply.store/api/v1/trust/kornia-geometric-vision-library/check \
-H "Content-Type: application/json" \
-d '{"minGrade":"B","denyPermissions":["shell"],"denyUnknownEgress":true}'
# CLI
npx ai-supply add kornia-geometric-vision-library
# REST (install → download)
curl -X POST https://ai-supply.store/api/v1/listings/kornia-geometric-vision-library/install \
-H "Authorization: Bearer $AIM_KEY"
# MCP tool
install_listing({ "slug": "kornia-geometric-vision-library" })OpenAPI spec →Curated mirror — latest upstream source. See the repository for tagged releases.