TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Computer Vision/Point Cloud Registration

Point Cloud Registration

56 benchmarks447 papers

Point Cloud Registration is a fundamental problem in 3D computer vision and photogrammetry. Given several sets of points in different coordinate systems, the aim of registration is to find the transformation that best aligns all of them into a common coordinate system. Point Cloud Registration plays a significant role in many vision applications such as 3D model reconstruction, cultural heritage management, landslide monitoring and solar energy analysis.

<span class="description-source">Source: Iterative Global Similarity Points : A robust coarse-to-fine integration solution for pairwise 3D point cloud registration </span>

Benchmarks

Point Cloud Registration on 3DMatch Benchmark

Feature Matching Recall

Point Cloud Registration on 3DMatch (at least 30% overlapped - FCGF setting)

Recall (0.3m, 15 degrees)RE (all)TE (all)

Point Cloud Registration on KITTI (trained on 3DMatch)

Success Rate

Point Cloud Registration on 3DLoMatch (10-30% overlap)

Recall ( correspondence RMSE below 0.2)

Point Cloud Registration on ETH (trained on 3DMatch)

Recall (30cm, 5 degrees)Feature Matching Recall

Point Cloud Registration on 3DMatch (at least 30% overlapped - sample 5k interest points)

Recall ( correspondence RMSE below 0.2)

Point Cloud Registration on KITTI (FCGF setting)

Recall (0.6m, 5 degrees)RE (all)TE (all)

Point Cloud Registration on FPv1

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-O-E

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-O-H

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-O-M

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-R-E

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-R-H

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-R-M

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-T-E

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-T-H

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on FP-T-M

Recall (3cm, 10 degrees)RRE (degrees)RTE (cm)

Point Cloud Registration on KITTI

Success RateRRERTE

Point Cloud Registration on RotKITTI Registration Benchmark

RR@(1.5,0.3)RR@(1,0.1)

Point Cloud Registration on 3DMatch (trained on KITTI)

Recall

Point Cloud Registration on KITTI (Distant PCR)

mRR @ Normal Criterion (1.5°&0.3m)RR @ Loose Criterion (5°&2m), on LoKITTI

Point Cloud Registration on nuScenes (Distant PCR)

mRR @ Normal Criterion (1.5°&0.3m)RR @ Loose Criterion (5°&2m), on LoNuScenes

Point Cloud Registration on ScanNet++ (trained on 3DMatch)

Recall ( correspondence RMSE below 0.2)

Point Cloud Registration on 3RScan

CDRRERTE