AI Paper Atlas

Research map / Point clouds / Registration and matching

GraphCPD: Coherent Point Drift for Point Cloud Registration via Graph Signal Processing

Yingcheng Lai, Xingjian Wang, Li Chai

ECCV 2026 · Point clouds › Registration and matching · approach: Robust inlier selection for registration · 0 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
replace isotropic covariance with local graph Laplacian density model (ablation)GraphCPD · DragonAngErr (deg): 0.1542 -> 0.091Improved
replace isotropic covariance with local graph Laplacian density model (ablation)GraphCPD · ArmadilloAngErr (deg): 0.3836 -> 0.095Improved
use weighted global graph instead of unweighted for high-pass filtering (ablation)GraphCPD · DragonAngErr (deg): 0.232 -> 0.091Improved
use weighted global graph instead of unweighted for high-pass filtering (ablation)GraphCPD · ArmadilloAngErr (deg): 0.308 -> 0.095Improved
increase kmatch from 50 to 250 (ablation)GraphCPD · DragonAngErr (deg): 0.242 -> 0.073Improved
increase kmatch from 50 to 250 (ablation)GraphCPD · Dragonruntime (s): 2.559 -> 10.045Hurt
increase kmatch from 50 to 250 (ablation)GraphCPD · ArmadilloAngErr (deg): 0.215 -> 0.071Improved
increase kmatch from 50 to 250 (ablation)GraphCPD · Armadilloruntime (s): 2.354 -> 11.592Hurt
increase local graph size k from 10 to 20 (ablation)GraphCPD · DragonAngErr (deg): 0.091 -> 0.362Hurt
decrease local graph size k from 10 to 5 (ablation)GraphCPD · DragonAngErr (deg): 0.091 -> 0.103Hurt
increase local graph size k from 10 to 20 (ablation)GraphCPD · ArmadilloAngErr (deg): 0.095 -> 0.244Hurt
increase local graph size k from 10 to 20 (ablation)GraphCPD · Dragonruntime (s): 2.950 -> 2.683Improved

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