AI Paper Atlas

Research map / Point clouds / Registration and matching

PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds

Anh-Quan Cao, Gilles Puy, Alexandre Boulch, Renaud Marlet

ICCV 2021 · Point clouds › Registration and matching · approach: Learned feature-based registration · 70 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
use product of attention matrices instead of last attention matrix only (ablation)PCAM-Soft · 3DMatch val, L=6recall (%): 76.6 -> 82.4Improved
use product of attention matrices instead of last attention matrix only (ablation)PCAM-Sparse · 3DMatch val, L=6recall (%): 78.4 -> 81.6Improved
use product of attention matrices instead of last attention matrix only (ablation)PCAM-Soft · 3DMatch val, L=2recall (%): 37.1 -> 52.3Improved
use product of attention matrices instead of last attention matrix only (ablation)PCAM-Sparse · 3DMatch val, L=2recall (%): 54.1 -> 52.4No clear effect
exchange context between point clouds via intermediate cross-attention matrices (ablation)PCAM-Soft · 3DMatch val, L=6recall (%): 64.4 -> 76.6Improved
exchange context between point clouds via intermediate cross-attention matrices (ablation)PCAM-Sparse · 3DMatch val, L=6recall (%): 68.0 -> 78.4Improved
increase number of layers from 2 to 6 (ablation)PCAM-Soft · 3DMatch valrecall (%): 52.3 -> 82.4Improved
increase number of layers from 2 to 6 (ablation)PCAM-Sparse · 3DMatch valrecall (%): 52.4 -> 81.6Improved
add LGeoConf confidence geometric loss (ablation)PCAM-Soft · 3DMatch val, L=2recall (%): 45.9 -> 52.3Improved
add LGeoConf confidence geometric loss (ablation)PCAM-Sparse · 3DMatch val, L=2recall (%): 48.5 -> 52.4Improved
replace classification attention loss by geometric attention loss (ablation)PCAM-Soft · 3DMatch val, L=2recall (%): 52.3 -> 19.4Hurt
add geometric attention loss to classification attention loss (ablation)PCAM-Soft · 3DMatch val, L=2recall (%): 52.3 -> 53.7Mixed

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