kornia/kornia-rs

[Feature]: Huber kernel based robustification of PnP pose refinement

Open

#1,099 opened on Aug 12, 2026

View on GitHubContributor guide
 (1 comment) (0 reactions) (0 assignees)Rust (188 forks)auto 404
enhancementhelp wantedtriage

Repository metrics

Stars
 (675 stars)
PR merge metrics
 (PR metrics pending)

Description

🚀 Feature Description

Current PnP from kornia-rs is not robust to fast jerky motion seen in Euroc Vicon room datasets. So a fix is added by adding a huber kernel in the pnp pose refinement cost function. The resulting scheme is under in kornia-slam here: https://github.com/kornia/kornia-slam/pull/54

📂 Feature Category

Rust Core Library

💡 Motivation

To robustify per frame tracking in Kornia-SLAM

💭 Proposed Solution

Adding huber kernel in the refine_pose cost function

📚 Library Reference

lightweight VIO (https://github.com/93won/lightweight_vio)

🔄 Alternatives Considered

No response

🎯 Use Cases

No response

📝 Additional Context

No response

🤝 Contribution Intent

  • I plan to submit a PR to implement this feature
  • I'm requesting this feature but not planning to implement it

Contributor guide

Research direction
Start by exploring the current PnP implementation in kornia rs, focusing on the pose refinement cost function. Review the referenced PR in kornia slam to see how the Huber kernel is applied, and understand how it improves robustness to jerky motion. Consider testing the existing code against Euroc Vicon room datasets to identify failure cases before implementing changes.
Tech stack
rust
Domain
computer vision
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
GitRust
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60