Skip to content
In GSoC 2026

Kornia

Kornia AI is a non-profit organization advancing Spatial Artificial Intelligence through open collaboration. Founded in 2018 and registered as a non-profit in 2023, Kornia AI is supported by a global contributor community. The organization began with Kornia, a widely adopted Python library for Differentiable Computer Vision (>2M downloads/month, 400+ Google Scholar citations). In recent years, the core maintainers have shifted focus to kornia-rs, a native Rust stack for high‑performance 3D computer vision and spatial AI. Our direction emphasizes robotics and edge deployment: safe, memory‑efficient, low‑latency pipelines that run on constrained devices (e.g., embedded/Jetson‑class hardware). kornia-rs is gaining traction in the open-source community and is used by leading companies such as Farm-ng, Rerun.io, and Copper Robotics, who have featured the project on their platforms.

Contributors in 2026

3

+1 from 2025

Projects since 2016

5

2.5 a cycle on average

Cycles

2 of 11

2-cycle streak

Rank in 2026

#119

of 183, 0.3% of slots

Mentors

5

for 2 projects in 2025

Contributors per cycle

Accepted projects each year, against the median organization that year

  1. 3
View as table
Kornia contributors per cycle
KorniaMedian organization
201604
201704
201803
201904
202004
202104
202203
202304
202404
202525
202634

Quick answers

Is Kornia in GSoC 2026?
Yes, with 3 contributors.
How long has it taken part?
2 of 11 cycles since 2016, first in 2025, with a 2-cycle streak running.
How many contributors does it usually take?
2.5 a cycle on average, 5 in total. Its largest cycle was 2026, with 3.
What does it build with?
CUDA, Rust, Deep Learning, Data Science, Spatial AI.

Accepted projects

3 projects in 2026 · All 5 projects since 2016

  1. Kornia-SLAM: Robust visual-inertial SLAM

    Astik Srivastava

    This project intends to extend the existing kornia-slam repository to create a visual + inertial measurements based simultaneous localization and...

  2. GPU Acceleration for kornia-tensor and kornia-imgproc

    Inchara Jayaram

    kornia-rs currently supports only CPU-based computation for image processing operations like resize, warp, and color transforms. This project...

  3. GPU-Accelerated VLM/VLA infrastructure on Bubbaloop

    Siddharth Rambhia

    This project builds GPU-accelerated image processing kernels for kornia-rs and integrates them into a real-time vision-language model (VLM) pipeline...

No accepted proposals or progress posts from Kornia contributors yet. Contributed here? Share yours.

Claim your project

Data: Google Summer of Code public archive. Contributor counts are accepted projects.