AllenNeuralDynamics / AllenNeuralDynamics/Rhapso
Add intensity-based candidate matching and RANSAC over full point set for R2R
Open
- Dominant language
- Python
- Stars
- 9
- Forks
- 3
- PR merge metrics
- No merged PRs in 30d
Description
R2R registration does not match reliably using geometry-based candidate descriptors.
Store a local intensity/context region around each detected interest point.
Generate match candidates by context-region correlation within the search radius.
Update RANSAC to seed from these candidates, then iteratively resample/refit against geometric matches from the full point sets.
Integrate a param to allow users to pick between matching approaches
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.