AI Paper Atlas

Research map / Point clouds / Registration and matching

SIRA-PCR: Sim-to-Real Adaptation for 3D Point Cloud Registration

Suyi Chen, Hao Xu, Ru Li, Guanghui Liu, Chi-Wing Fu, Shuaicheng Liu

ICCV 2023 · Point clouds › Registration and matching · approach: Learned feature-based registration · 28 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
insert ShapeNet objects into 3D-FRONT scenes (ablation)FlyingShapes · 3DLoMatchRegistration Recall: 33.5 -> 55.1Improved
delete flat planes with 50% probability (ablation)FlyingShapes · 3DLoMatchRegistration Recall: 55.1 -> 57.4Improved
add Structured3D to FlyingShapes (ablation)FlyingShapes · 3DLoMatchRegistration Recall: 57.4 -> 63.8Improved
add uniform or Gaussian noise to synthetic data (ablation)GeoTrans pretrained on FlyingShapes · 3DLoMatchRegistration Recall: 63.8 -> 62.7 (uniform), 63.8 (Gaussian)No clear effect
apply TSDF fusion fine-tuning (ablation)GeoTrans pretrained on FlyingShapes · 3DLoMatchRegistration Recall: 63.8 -> 63.9No clear effect
add SIRA sim-to-real adaptation (ablation)GeoTrans pretrained on FlyingShapes · 3DLoMatchRegistration Recall: 63.8 -> 67.4Improved
add adaptive re-sample module (ARM) (ablation)SIRA · 3DLoMatchRegistration Recall: 65.3 -> 67.4Improved
add multi-scale discriminator (ablation)SIRA · 3DLoMatchRegistration Recall: 66.6 -> 67.4Improved
fine-tune synthetic-trained model on 3DMatchSIRA-PCR · ETH (RANSAC, 250 samples)Registration Recall: 95.4 -> 82.2Hurt
train only on synthetic data (SIRA-PCR) versus GeoTrans trained on real dataGeoTrans · 3DLoMatchRegistration Recall: 74.0 -> 67.4Hurt
fine-tune SIRA-PCR on 3DMatch, using RANSACSIRA-PCR · 3DLoMatch RR (5000 samples)Registration Recall: GeoTrans 75.0 -> SIRA-PCR 73.5Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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