AI Paper Atlas

Research map / Point clouds / Registration and matching

(Just) A Spoonful of Refinements Helps the Registration Error Go Down

Sérgio Agostinho, Aljoša Ošep, Alessio Del Bue, Laura Leal-Taixé

ICCV 2021 · Point clouds › Registration and matching · approach: Robust inlier selection for registration · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add iterative linearized refinement layer during trainingDCP-v2 · ModelNet40, DCP, noise on source only (unseen instances)RMSE(R) deg: 12.974750 -> 8.938098Improved
add iterative linearized refinement layer during trainingDCP-v2 · ModelNet40, DCP, noise on source only (unseen instances)RMSE(t): 0.003941 -> 0.004221Hurt
add iterative linearized refinement layer during trainingDCP-v2 · ModelNet40, DCP, unseen categoriesRMSE(R) deg: 3.150191 -> 2.051713Improved
add iterative linearized refinement layer during trainingDCP-v2 · ModelNet40, DCP, unseen categoriesRMSE(t): 0.005039 -> 0.004543Improved
add iterative linearized refinement layer during trainingDCP-v2 · 3DMatch, DCP, 30% min overlapRMSE(R) deg: 11.211800 -> 10.232300Improved
add iterative linearized refinement layer during trainingDCP-v2 · 3DMatch, DCP, 30% min overlapRMSE(t): 0.137432 -> 0.128846Improved
add iterative linearized refinement layer during trainingRPM-Net · ModelNet40, RPM-Net, partial visible data with noiseAnisotropic rot. err (deg): 0.893 -> 0.826Improved
add iterative linearized refinement layer during trainingRPM-Net · ModelNet40, RPM-Net, partial visible data with noiseAnisotropic trans. err: 0.0087 -> 0.0081Improved
add inlier loss term to RPM-Net training (ablation)RPM-Net · ModelNet40, RPM-Net, partial visible data with noiseAnisotropic rot. err (deg): 0.993 -> 0.893Improved
add inlier loss term to RPM-Net + Ours training (ablation)RPM-Net + Ours · ModelNet40, RPM-Net, partial visible data with noiseAnisotropic rot. err (deg): 0.872 -> 0.826Mixed
add iterative refinement layer, 26 training runs (unseen categories)DCP-v2 · ModelNet40, DCP, unseen categories, 26 runsRMSE(R) deg: 4.264873 -> 3.311812Improved
add iterative refinement layer, 26 training runs (unseen categories)DCP-v2 · ModelNet40, DCP, unseen categories, 26 runsRMSE(t): 0.006250 -> 0.006532Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy