SOTAVerified

Point Cloud Registration

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.

Source: Iterative Global Similarity Points : A robust coarse-to-fine integration solution for pairwise 3D point cloud registration

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 101–150 of 447 papers

TitleStatusHype
Addressing the generalization of 3D registration methods with a featureless baseline and an unbiased benchmarkCode1
VRHCF: Cross-Source Point Cloud Registration via Voxel Representation and Hierarchical Correspondence FilteringCode1
MEDPNet: Achieving High-Precision Adaptive Registration for Complex Die Castings—0
FastMAC: Stochastic Spectral Sampling of Correspondence GraphCode2
NeRF-Supervised Feature Point Detection and DescriptionCode0
PSS-BA: LiDAR Bundle Adjustment with Progressive Spatial Smoothing—0
Extend Your Own Correspondences: Unsupervised Distant Point Cloud Registration by Progressive Distance ExtensionCode2
PCR-99: A Practical Method for Point Cloud Registration with 99 Percent OutliersCode1
CLIPPER+: A Fast Maximal Clique Algorithm for Robust Global RegistrationCode1
CLIPPER: Robust Data Association without an Initial Guess—0
Registration of algebraic varieties using Riemannian optimizationCode0
Iterative Feedback Network for Unsupervised Point Cloud RegistrationCode1
PosDiffNet: Positional Neural Diffusion for Point Cloud Registration in a Large Field of View with PerturbationsCode0
OptFlow: Fast Optimization-based Scene Flow Estimation without Supervision—0
ColorPCR: Color Point Cloud Registration with Multi-Stage Geometric-Color Fusion—0
Dynamic Cues-Assisted Transformer for Robust Point Cloud Registration—0
Inlier Confidence Calibration for Point Cloud Registration—0
Diff-PCR: Diffusion-Based Correspondence Searching in Doubly Stochastic Matrix Space for Point Cloud Registration—0
On Partial Optimal Transport: Revising the Infeasibility of Sinkhorn and Efficient Gradient MethodsCode0
D3Former: Jointly Learning Repeatable Dense Detectors and Feature-enhanced Descriptors via Saliency-guided Transformer—0
SPEAL: Skeletal Prior Embedded Attention Learning for Cross-Source Point Cloud Registration—0
PCRDiffusion: Diffusion Probabilistic Models for Point Cloud Registration—0
Adaptive Multi-step Refinement Network for Robust Point Cloud RegistrationCode0
ZeroReg: Zero-Shot Point Cloud Registration with Foundation Models—0
DIPR: Efficient Point Cloud Registration via Dynamic Iteration—0
E2PNet: Event to Point Cloud Registration with Spatio-Temporal Representation LearningCode1
Nothing Stands Still: A Spatiotemporal Benchmark on 3D Point Cloud Registration Under Large Geometric and Temporal Change—0
Cross-Modal Information-Guided Network using Contrastive Learning for Point Cloud RegistrationCode1
Transformation Decoupling Strategy based on Screw Theory for Deterministic Point Cloud Registration with Gravity PriorCode0
HDMNet: A Hierarchical Matching Network with Double Attention for Large-scale Outdoor LiDAR Point Cloud Registration—0
SE(3) Diffusion Model-based Point Cloud Registration for Robust 6D Object Pose EstimationCode1
DBDNet:Partial-to-Partial Point Cloud Registration with Dual Branches Decoupling—0
OAAFormer: Robust and Efficient Point Cloud Registration Through Overlapping-Aware Attention in Transformer—0
Colmap-PCD: An Open-source Tool for Fine Image-to-point cloud RegistrationCode2
Lie Neurons: Adjoint-Equivariant Neural Networks for Semisimple Lie AlgebrasCode1
FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth EstimatorsCode2
Q-REG: End-to-End Trainable Point Cloud Registration with Surface Curvature—0
Partial Transport for Point-Cloud Registration—0
KDD-LOAM: Jointly Learned Keypoint Detector and Descriptors Assisted LiDAR Odometry and Mapping—0
CoFiI2P: Coarse-to-Fine Correspondences for Image-to-Point Cloud Registration—0
AutoSynth: Learning to Generate 3D Training Data for Object Point Cloud Registration—0
FF-LOGO: Cross-Modality Point Cloud Registration with Feature Filtering and Local to Global OptimizationCode0
A Ground Segmentation Method Based on Point Cloud Map for Unstructured Roads—0
Fast and Accurate Deep Loop Closing and Relocalization for Reliable LiDAR SLAM—0
EP2P-Loc: End-to-End 3D Point to 2D Pixel Localization for Large-Scale Visual LocalizationCode1
SGNet: Salient Geometric Network for Point Cloud Registration—0
Point-TTA: Test-Time Adaptation for Point Cloud Registration Using Multitask Meta-Auxiliary Learning—0
In-Rack Test Tube Pose Estimation Using RGB-D Data—0
Overlap Bias Matching is Necessary for Point Cloud Registration—0
DReg-NeRF: Deep Registration for Neural Radiance FieldsCode1
Show:102550
← PrevPage 3 of 9Next →
All datasetsETH (trained on 3DMatch)3DMatch Benchmark3DMatch (at least 30% overlapped - FCGF setting)KITTI (trained on 3DMatch)3DLoMatch (10-30% overlap)3DMatch (at least 30% overlapped - sample 5k interest points)KITTI (FCGF setting)FPv1FP-O-EFP-O-HFP-O-MFP-R-E

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Exhaustive Grid SearchRecall (30cm, 5 degrees)94.25—Unverified
2FCGF+SC2-PCRRecall (30cm, 5 degrees)92.85—Unverified
3FPFH+SC2-PCRRecall (30cm, 5 degrees)85.41—Unverified
4YOHO-CRecall (30cm, 5 degrees)84.85—Unverified
5YOHO-ORecall (30cm, 5 degrees)79.94—Unverified
6FCGF+PointDSCRecall (30cm, 5 degrees)77.42—Unverified
7FPFH+PointDSCRecall (30cm, 5 degrees)41.94—Unverified
8GeoTransformerRecall (30cm, 5 degrees)4.91—Unverified
9GO-ICPRecall (30cm, 5 degrees)1.54—Unverified
10GeDiFeature Matching Recall0.98—Unverified
#ModelMetricClaimedVerifiedStatus
1IMFNetFeature Matching Recall98.6—Unverified
2MS-SVConvFeature Matching Recall98.4—Unverified
3GeDi (no code published as of May 27 2021)Feature Matching Recall97.9—Unverified
4SpinNet (no code published as of Dec 15 2020)Feature Matching Recall97.6—Unverified
5LMVDFeature Matching Recall97.5—Unverified
6DIP + Point-TnTFeature Matching Recall96.8—Unverified
7D3Feat-PredFeature Matching Recall95.8—Unverified
8D3Feat-randFeature Matching Recall95.3—Unverified
9DIPFeature Matching Recall94.8—Unverified
103DSmoothNetFeature Matching Recall94.7—Unverified
#ModelMetricClaimedVerifiedStatus
1GeoTransformerRecall (0.3m, 15 degrees)95—Unverified
2NgeNetRecall (0.3m, 15 degrees)95—Unverified
3PCAM-Sparse (All post-processing)Recall (0.3m, 15 degrees)92.4—Unverified
4DGR (RE (all), TE(all) are reported in PCAM)Recall (0.3m, 15 degrees)91.3—Unverified
5PCAM-Soft (All post-processing)Recall (0.3m, 15 degrees)91.3—Unverified
6Exhaustive Grid SearchRecall (0.3m, 15 degrees)84.11—Unverified
7RANSAC-2MRecall (0.3m, 15 degrees)66.1—Unverified
8FGRRecall (0.3m, 15 degrees)42.7—Unverified
9Go-ICPRecall (0.3m, 15 degrees)22.9—Unverified
10Super4PCSRecall (0.3m, 15 degrees)21.6—Unverified
#ModelMetricClaimedVerifiedStatus
1GeDiSuccess Rate98.92—Unverified
2FCGF+SC2-PCRSuccess Rate97.66—Unverified
3FCGF+PointDSCSuccess Rate96.76—Unverified
4Exhaustive Grid SearchSuccess Rate94.95—Unverified
5FPFH+PointDSCSuccess Rate94.05—Unverified
6DIPSuccess Rate93.51—Unverified
7Greedy Grid SearchSuccess Rate90.27—Unverified
8YOHO-CSuccess Rate82.16—Unverified
9SpinNetSuccess Rate81.44—Unverified
10YOHO-OSuccess Rate81.44—Unverified
#ModelMetricClaimedVerifiedStatus
1CoFFRecall ( correspondence RMSE below 0.2)81.6—Unverified
2GeoTransformer - P2PNetRecall ( correspondence RMSE below 0.2)74—Unverified
3NgeNetRecall ( correspondence RMSE below 0.2)71.9—Unverified
4REGTRRecall ( correspondence RMSE below 0.2)64.8—Unverified
5Predator-1kRecall ( correspondence RMSE below 0.2)62.5—Unverified
6Predator-5kRecall ( correspondence RMSE below 0.2)59.8—Unverified
7PCAM (reported in REGTR)Recall ( correspondence RMSE below 0.2)54.9—Unverified
8DGR (reported in REGTR)Recall ( correspondence RMSE below 0.2)48.7—Unverified
9FCGF (reported in PREDATOR)Recall ( correspondence RMSE below 0.2)40.1—Unverified
10D3Feat (reported in PREDATOR)Recall ( correspondence RMSE below 0.2)37.2—Unverified
#ModelMetricClaimedVerifiedStatus
1NgeNetRecall ( correspondence RMSE below 0.2)92.9—Unverified
2REGTRRecall ( correspondence RMSE below 0.2)92—Unverified
3Predator-1kRecall ( correspondence RMSE below 0.2)90.5—Unverified
4Predator-5kRecall ( correspondence RMSE below 0.2)89—Unverified
5PCAM (reported in REGTR)Recall ( correspondence RMSE below 0.2)85.5—Unverified
6DGR (reported in REGTR)Recall ( correspondence RMSE below 0.2)85.3—Unverified
7FCGF (reported in PREDATOR)Recall ( correspondence RMSE below 0.2)85.1—Unverified
8D3Feat (reported in PREDATOR)Recall ( correspondence RMSE below 0.2)81.6—Unverified
93DSN (reported in PREDATOR)Recall ( correspondence RMSE below 0.2)78.4—Unverified
10Predator-NRRecall ( correspondence RMSE below 0.2)62.7—Unverified
#ModelMetricClaimedVerifiedStatus
1GeoTransformerRecall (0.6m, 5 degrees)99.5—Unverified
2FCGFRecall (0.6m, 5 degrees)98.2—Unverified
3DGR + ICP (RE (all), TE(all) are reported in PCAM)Recall (0.6m, 5 degrees)98.2—Unverified
4PCAM-soft + ICPRecall (0.6m, 5 degrees)98—Unverified
5PCAM-Sparse + ICPRecall (0.6m, 5 degrees)97.4—Unverified
6PCAM - SoftRecall (0.6m, 5 degrees)97.2—Unverified
7DGR (RE (all), TE(all) are reported in PCAM)Recall (0.6m, 5 degrees)96.9—Unverified
8PCAM-SparseRecall (0.6m, 5 degrees)96.5—Unverified
9Exhaustive Grid SearchRecall (0.6m, 5 degrees)94.59—Unverified
10RANSACRecall (0.6m, 5 degrees)34.2—Unverified
#ModelMetricClaimedVerifiedStatus
1Greedy Grid SearchRecall (3cm, 10 degrees)92.81—Unverified
2GeoTransformerRecall (3cm, 10 degrees)56.15—Unverified
3DIPRecall (3cm, 10 degrees)54.81—Unverified
4FCGF + PointDSCRecall (3cm, 10 degrees)47.85—Unverified
5SpinNetRecall (3cm, 10 degrees)42.46—Unverified
6FCGF + YOHO-CRecall (3cm, 10 degrees)29.18—Unverified
7FCGF + YOHO-ORecall (3cm, 10 degrees)18.91—Unverified
8FPFH-8MRecall (3cm, 10 degrees)9.51—Unverified
#ModelMetricClaimedVerifiedStatus
1FPFH+SC2-PCR Recall (3cm, 10 degrees)99.88—Unverified
2GeDiRecall (3cm, 10 degrees)99.64—Unverified
3Exhaustive Grid SearchRecall (3cm, 10 degrees)99.47—Unverified
4FCGF+SC2-PCRRecall (3cm, 10 degrees)98.52—Unverified
5GeoTransformerRecall (3cm, 10 degrees)63.94—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1FPFH+SC2-PCRRecall (3cm, 10 degrees)38.85—Unverified
2Exhaustive Grid SearchRecall (3cm, 10 degrees)37.06—Unverified
3FCGF+SC2-PCRRecall (3cm, 10 degrees)17.8—Unverified
4GeDiRecall (3cm, 10 degrees)8.7—Unverified
5GeoTransformerRecall (3cm, 10 degrees)2.64—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1Exhaustive Grid SearchRecall (3cm, 10 degrees)88.06—Unverified
2FPFH+SC2-PCRRecall (3cm, 10 degrees)84.7—Unverified
3GeDiRecall (3cm, 10 degrees)75.4—Unverified
4FCGF+SC2-PCRRecall (3cm, 10 degrees)63—Unverified
5GeoTransformerRecall (3cm, 10 degrees)22.07—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1GeDiRecall (3cm, 10 degrees)99.76—Unverified
2Exhaustive Grid SearchRecall (3cm, 10 degrees)99.64—Unverified
3FPFH+SC2-PCRRecall (3cm, 10 degrees)99.64—Unverified
4FCGF+SC2-PCRRecall (3cm, 10 degrees)98.46—Unverified
5GeoTransformerRecall (3cm, 10 degrees)64.12—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1GeDiRecall (3cm, 10 degrees)99.41—Unverified
2FCGF+SC2-PCR Recall (3cm, 10 degrees)85.77—Unverified
3Exhaustive Grid SearchRecall (3cm, 10 degrees)78—Unverified
4FPFH+SC2-PCRRecall (3cm, 10 degrees)75.21—Unverified
5GeoTransformerRecall (3cm, 10 degrees)47.75—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1GeDiRecall (3cm, 10 degrees)99.94—Unverified
2Exhaustive Grid SearchRecall (3cm, 10 degrees)97.92—Unverified
3FPFH+SC2-PCRRecall (3cm, 10 degrees)94.54—Unverified
4FCGF+SC2-PCRRecall (3cm, 10 degrees)91.93—Unverified
5GeoTransformerRecall (3cm, 10 degrees)55.93—Unverified
6GO-ICPRecall (3cm, 10 degrees)0.06—Unverified
#ModelMetricClaimedVerifiedStatus
1FPFH+SC2-PCRRecall (3cm, 10 degrees)99.76—Unverified
2Exhaustive Grid SearchRecall (3cm, 10 degrees)99.7—Unverified
3GeDiRecall (3cm, 10 degrees)99.47—Unverified
4FCGF+SC2-PCRRecall (3cm, 10 degrees)98.34—Unverified
5GeoTransformerRecall (3cm, 10 degrees)66.25—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1GeDiRecall (3cm, 10 degrees)99.7—Unverified
2FPFH+SC2-PCRRecall (3cm, 10 degrees)99.58—Unverified
3Exhaustive Grid SearchRecall (3cm, 10 degrees)98.81—Unverified
4FCGF+SC2-PCRRecall (3cm, 10 degrees)98.22—Unverified
5GeoTransformerRecall (3cm, 10 degrees)64.18—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1Exhaustive Grid SearchRecall (3cm, 10 degrees)99.82—Unverified
2GeDiRecall (3cm, 10 degrees)99.7—Unverified
3FPFH+SC2-PCRRecall (3cm, 10 degrees)99.53—Unverified
4FCGF+SC2-PCRRecall (3cm, 10 degrees)98.34—Unverified
5GeoTransformerRecall (3cm, 10 degrees)64.29—Unverified
6GO-ICPRecall (3cm, 10 degrees)0—Unverified
#ModelMetricClaimedVerifiedStatus
1GeDiSuccess Rate99.82—Unverified
2D3Feat-predSuccess Rate99.81—Unverified
3SpinNetSuccess Rate99.1—Unverified
4DIPSuccess Rate97.3—Unverified
5FCGFSuccess Rate96.57—Unverified
63DFeat-NetSuccess Rate95.97—Unverified
#ModelMetricClaimedVerifiedStatus
1UMERegRobustRR@(1.5,0.3)81.1—Unverified
2GeoTransformerRR@(1.5,0.3)78.5—Unverified
3CoFiNetRR@(1.5,0.3)62.5—Unverified
4PREDATORRR@(1.5,0.3)41.6—Unverified
5GCLRR@(1.5,0.3)40.1—Unverified
6FCGFRR@(1.5,0.3)11.6—Unverified
#ModelMetricClaimedVerifiedStatus
1GeDiRecall0.92—Unverified
2SpinNetRecall0.85—Unverified
3D3Feat-predRecall0.63—Unverified
4FCGFRecall0.33—Unverified
5FPFHRecall0.14—Unverified
#ModelMetricClaimedVerifiedStatus
1GCL+KPConvmRR @ Normal Criterion (1.5°&0.3m)88.8—Unverified
2GCL+ConvmRR @ Normal Criterion (1.5°&0.3m)83.5—Unverified
3Predator+APR(a)mRR @ Normal Criterion (1.5°&0.3m)78.2—Unverified
4FCGF+APR(s)mRR @ Normal Criterion (1.5°&0.3m)62.9—Unverified
#ModelMetricClaimedVerifiedStatus
1GCL+KPConvmRR @ Normal Criterion (1.5°&0.3m)71.5—Unverified
2GCL+ConvmRR @ Normal Criterion (1.5°&0.3m)70.2—Unverified
3FCGF+APR(s)mRR @ Normal Criterion (1.5°&0.3m)62.9—Unverified
4Predator+APR(a)mRR @ Normal Criterion (1.5°&0.3m)52.2—Unverified
#ModelMetricClaimedVerifiedStatus
1CoFFRecall ( correspondence RMSE below 0.2)78.7—Unverified
2GeoTransformerRecall ( correspondence RMSE below 0.2)73.4—Unverified
3PredatorRecall ( correspondence RMSE below 0.2)63.7—Unverified
#ModelMetricClaimedVerifiedStatus
1SGAlignerCD0.01—Unverified
2SG-PGMCD0.01—Unverified