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 Added Most Hyped Most Active Needs Verification Most Verified

Showing 201–250 of 447 papers

Title	Date	Tasks	Status	Hype
Rethinking Point Cloud Registration as Masking and Reconstruction	Jan 1, 2023	Point Cloud Registration	CodeCode Available	1
SiLK: Simple Learned Keypoints	Jan 1, 2023	3D ReconstructionCamera Pose Estimation	—Unverified	0
PEAL: Prior-Embedded Explicit Attention Learning for Low-Overlap Point Cloud Registration	Jan 1, 2023	Point Cloud Registration	CodeCode Available	1
BUFFER: Balancing Accuracy, Efficiency, and Generalizability in Point Cloud Registration	Jan 1, 2023	Computational EfficiencyOpen-Ended Question Answering	CodeCode Available	1
Rethinking Rotation Invariance with Point Cloud Registration	Dec 31, 2022	3D Shape ClassificationPoint Cloud Registration	—Unverified	0
Generalized Differentiable RANSAC	Dec 26, 2022	Point Cloud Registration	CodeCode Available	1
Sliced Optimal Partial Transport	Dec 15, 2022	Point Cloud Registration	CodeCode Available	1
DeepMapping2: Self-Supervised Large-Scale LiDAR Map Optimization	Dec 13, 2022	Point Cloud Registration	CodeCode Available	0
Evolutionary Multitasking with Solution Space Cutting for Point Cloud Registration	Dec 12, 2022	Point Cloud RegistrationTransfer Learning	—Unverified	0
Point Cloud Registration of non-rigid objects in sparse 3D Scans with applications in Mixed Reality	Dec 7, 2022	Mixed RealityObject	—Unverified	0
Challenging the Universal Representation of Deep Models for 3D Point Cloud Registration	Nov 29, 2022	Point Cloud Registration	CodeCode Available	1
Learning Implicit Probability Distribution Functions for Symmetric Orientation Estimation from RGB Images Without Pose Labels	Nov 21, 2022	ObjectPoint Cloud Registration	—Unverified	0
Mirror Sinkhorn: Fast Online Optimization on Transport Polytopes	Nov 18, 2022	Point Cloud Registration	—Unverified	0
Robust Point Cloud Registration Framework Based on Deep Graph Matching(TPAMI Version)	Nov 9, 2022	graph constructionGraph Matching	CodeCode Available	1
DNN Filter for Bias Reduction in Distribution-to-Distribution Scan Matching	Nov 8, 2022	Point Cloud Registration	CodeCode Available	1
KSS-ICP: Point Cloud Registration based on Kendall Shape Space	Nov 5, 2022	Point Cloud RegistrationRetrieval	CodeCode Available	1
GMF: General Multimodal Fusion Framework for Correspondence Outlier Rejection	Nov 1, 2022	Point Cloud RegistrationPosition	CodeCode Available	0
Overlap-guided Gaussian Mixture Models for Point Cloud Registration	Oct 17, 2022	Point Cloud Registration	CodeCode Available	1
Spectral Geometric Verification: Re-Ranking Point Cloud Retrieval for Metric Localization	Oct 10, 2022	Point Cloud RegistrationPoint Cloud Retrieval	CodeCode Available	1
PCKRF: Point Cloud Completion and Keypoint Refinement With Fusion Data for 6D Pose Estimation	Oct 7, 2022	6D Pose EstimationPoint Cloud Completion	CodeCode Available	1
Coarse-to-Fine Point Cloud Registration with SE(3)-Equivariant Representations	Oct 5, 2022	Point Cloud Registration	CodeCode Available	0
Centroid Distance Keypoint Detector for Colored Point Clouds	Oct 4, 2022	Computational EfficiencyKeypoint Detection	CodeCode Available	1
Category-Level Global Camera Pose Estimation with Multi-Hypothesis Point Cloud Correspondences	Sep 28, 2022	Camera Pose EstimationPoint Cloud Registration	—Unverified	0
RIGA: Rotation-Invariant and Globally-Aware Descriptors for Point Cloud Registration	Sep 27, 2022	Point Cloud Registration	CodeCode Available	1
Point Cloud Registration-Driven Robust Feature Matching for 3D Siamese Object Tracking	Sep 14, 2022	Object LocalizationObject Tracking	—Unverified	0
PointCLM: A Contrastive Learning-based Framework for Multi-instance Point Cloud Registration	Sep 1, 2022	Contrastive LearningPoint Cloud Registration	CodeCode Available	1
Improving RGB-D Point Cloud Registration by Learning Multi-scale Local Linear Transformation	Aug 31, 2022	Point Cloud Registration	CodeCode Available	1
Global Consistent Point Cloud Registration Based on Lie-algebraic Cohomology	Aug 15, 2022	Point Cloud Registration	—Unverified	0
UTOPIC: Uncertainty-aware Overlap Prediction Network for Partial Point Cloud Registration	Aug 4, 2022	DecoderPoint Cloud Registration	CodeCode Available	1
SuperLine3D: Self-supervised Line Segmentation and Description for LiDAR Point Cloud	Aug 3, 2022	Point Cloud RegistrationSegmentation	CodeCode Available	1
Accurate Ground-Truth Depth Image Generation via Overfit Training of Point Cloud Registration using Local Frame Sets	Jul 14, 2022	Image GenerationPoint Cloud Registration	—Unverified	0
CorrI2P: Deep Image-to-Point Cloud Registration via Dense Correspondence	Jul 12, 2022	Image to Point Cloud RegistrationPoint Cloud Registration	CodeCode Available	1
Learning to Register Unbalanced Point Pairs	Jul 9, 2022	Point Cloud Registration	—Unverified	0
ImLoveNet: Misaligned Image-supported Registration Network for Low-overlap Point Cloud Pairs	Jul 2, 2022	Point Cloud Registration	—Unverified	0
Learning Dense Features for Point Cloud Registration Using a Graph Attention Network	Jun 14, 2022	GPUGraph Attention	—Unverified	0
Neural Shape Mating: Self-Supervised Object Assembly with Adversarial Shape Priors	May 30, 2022	ObjectPoint Cloud Registration	—Unverified	0
Spotlights: Probing Shapes from Spherical Viewpoints	May 25, 2022	Point Cloud CompletionPoint Cloud Registration	CodeCode Available	0
Non-rigid Point Cloud Registration with Neural Deformation Pyramid	May 25, 2022	Point Cloud Registration	CodeCode Available	2
RCP: Recurrent Closest Point for Scene Flow Estimation on 3D Point Clouds	May 23, 2022	Motion EstimationPoint Cloud Registration	—Unverified	0
A New Outlier Removal Strategy Based on Reliability of Correspondence Graph for Fast Point Cloud Registration	May 16, 2022	Point Cloud Registration	CodeCode Available	2
A Novel Augmented Reality Ultrasound Framework Using an RGB-D Camera and a 3D-printed Marker	May 9, 2022	Point Cloud Registration	—Unverified	0
Multi-view Point Cloud Registration based on Evolutionary Multitasking with Bi-Channel Knowledge Sharing Mechanism	May 6, 2022	3D ReconstructionPoint Cloud Registration	—Unverified	0
Discrete-Continuous Smoothing and Mapping	Apr 25, 2022	Point Cloud Registration	CodeCode Available	1
Stress-Testing Point Cloud Registration on Automotive LiDAR	Apr 16, 2022	Autonomous DrivingBenchmarking	CodeCode Available	1
Points to Patches: Enabling the Use of Self-Attention for 3D Shape Recognition	Apr 8, 2022	3D Feature Matching3D Point Cloud Classification	CodeCode Available	1
Outlier-Robust Geometric Perception: A Novel Thresholding-Based Estimator with Intra-Class Variance Maximization	Apr 4, 2022	Point Cloud Registration	CodeCode Available	0
MatchNorm: Learning-based Point Cloud Registration for 6D Object Pose Estimation in the Real World	Mar 29, 2022	6D Pose Estimation using RGBObject	—Unverified	0
REGTR: End-to-end Point Cloud Correspondences with Transformers	Mar 28, 2022	Point Cloud RegistrationPose Estimation	CodeCode Available	2
SC^2-PCR: A Second Order Spatial Compatibility for Efficient and Robust Point Cloud Registration	Mar 28, 2022	Point Cloud Registration	CodeCode Available	1
VRNet: Learning the Rectified Virtual Corresponding Points for 3D Point Cloud Registration	Mar 24, 2022	Point Cloud Registration	—Unverified	0

Show:10 25 50

← PrevPage 5 of 9Next →

All datasets ETH (trained on 3DMatch)3DMatch Benchmark 3DMatch (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)FPv1 FP-O-E FP-O-H FP-O-M FP-R-E

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	Exhaustive Grid Search	Recall (30cm, 5 degrees)	94.25	—	Unverified
2	FCGF+SC2-PCR	Recall (30cm, 5 degrees)	92.85	—	Unverified
3	FPFH+SC2-PCR	Recall (30cm, 5 degrees)	85.41	—	Unverified
4	YOHO-C	Recall (30cm, 5 degrees)	84.85	—	Unverified
5	YOHO-O	Recall (30cm, 5 degrees)	79.94	—	Unverified
6	FCGF+PointDSC	Recall (30cm, 5 degrees)	77.42	—	Unverified
7	FPFH+PointDSC	Recall (30cm, 5 degrees)	41.94	—	Unverified
8	GeoTransformer	Recall (30cm, 5 degrees)	4.91	—	Unverified
9	GO-ICP	Recall (30cm, 5 degrees)	1.54	—	Unverified
10	GeDi	Feature Matching Recall	0.98	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	IMFNet	Feature Matching Recall	98.6	—	Unverified
2	MS-SVConv	Feature Matching Recall	98.4	—	Unverified
3	GeDi (no code published as of May 27 2021)	Feature Matching Recall	97.9	—	Unverified
4	SpinNet (no code published as of Dec 15 2020)	Feature Matching Recall	97.6	—	Unverified
5	LMVD	Feature Matching Recall	97.5	—	Unverified
6	DIP + Point-TnT	Feature Matching Recall	96.8	—	Unverified
7	D3Feat-Pred	Feature Matching Recall	95.8	—	Unverified
8	D3Feat-rand	Feature Matching Recall	95.3	—	Unverified
9	DIP	Feature Matching Recall	94.8	—	Unverified
10	3DSmoothNet	Feature Matching Recall	94.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	NgeNet	Recall (0.3m, 15 degrees)	95	—	Unverified
2	GeoTransformer	Recall (0.3m, 15 degrees)	95	—	Unverified
3	PCAM-Sparse (All post-processing)	Recall (0.3m, 15 degrees)	92.4	—	Unverified
4	PCAM-Soft (All post-processing)	Recall (0.3m, 15 degrees)	91.3	—	Unverified
5	DGR (RE (all), TE(all) are reported in PCAM)	Recall (0.3m, 15 degrees)	91.3	—	Unverified
6	Exhaustive Grid Search	Recall (0.3m, 15 degrees)	84.11	—	Unverified
7	RANSAC-2M	Recall (0.3m, 15 degrees)	66.1	—	Unverified
8	FGR	Recall (0.3m, 15 degrees)	42.7	—	Unverified
9	Go-ICP	Recall (0.3m, 15 degrees)	22.9	—	Unverified
10	Super4PCS	Recall (0.3m, 15 degrees)	21.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeDi	Success Rate	98.92	—	Unverified
2	FCGF+SC2-PCR	Success Rate	97.66	—	Unverified
3	FCGF+PointDSC	Success Rate	96.76	—	Unverified
4	Exhaustive Grid Search	Success Rate	94.95	—	Unverified
5	FPFH+PointDSC	Success Rate	94.05	—	Unverified
6	DIP	Success Rate	93.51	—	Unverified
7	Greedy Grid Search	Success Rate	90.27	—	Unverified
8	YOHO-C	Success Rate	82.16	—	Unverified
9	YOHO-O	Success Rate	81.44	—	Unverified
10	SpinNet	Success Rate	81.44	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CoFF	Recall ( correspondence RMSE below 0.2)	81.6	—	Unverified
2	GeoTransformer - P2PNet	Recall ( correspondence RMSE below 0.2)	74	—	Unverified
3	NgeNet	Recall ( correspondence RMSE below 0.2)	71.9	—	Unverified
4	REGTR	Recall ( correspondence RMSE below 0.2)	64.8	—	Unverified
5	Predator-1k	Recall ( correspondence RMSE below 0.2)	62.5	—	Unverified
6	Predator-5k	Recall ( correspondence RMSE below 0.2)	59.8	—	Unverified
7	PCAM (reported in REGTR)	Recall ( correspondence RMSE below 0.2)	54.9	—	Unverified
8	DGR (reported in REGTR)	Recall ( correspondence RMSE below 0.2)	48.7	—	Unverified
9	FCGF (reported in PREDATOR)	Recall ( correspondence RMSE below 0.2)	40.1	—	Unverified
10	D3Feat (reported in PREDATOR)	Recall ( correspondence RMSE below 0.2)	37.2	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	NgeNet	Recall ( correspondence RMSE below 0.2)	92.9	—	Unverified
2	REGTR	Recall ( correspondence RMSE below 0.2)	92	—	Unverified
3	Predator-1k	Recall ( correspondence RMSE below 0.2)	90.5	—	Unverified
4	Predator-5k	Recall ( correspondence RMSE below 0.2)	89	—	Unverified
5	PCAM (reported in REGTR)	Recall ( correspondence RMSE below 0.2)	85.5	—	Unverified
6	DGR (reported in REGTR)	Recall ( correspondence RMSE below 0.2)	85.3	—	Unverified
7	FCGF (reported in PREDATOR)	Recall ( correspondence RMSE below 0.2)	85.1	—	Unverified
8	D3Feat (reported in PREDATOR)	Recall ( correspondence RMSE below 0.2)	81.6	—	Unverified
9	3DSN (reported in PREDATOR)	Recall ( correspondence RMSE below 0.2)	78.4	—	Unverified
10	Predator-NR	Recall ( correspondence RMSE below 0.2)	62.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeoTransformer	Recall (0.6m, 5 degrees)	99.5	—	Unverified
2	DGR + ICP (RE (all), TE(all) are reported in PCAM)	Recall (0.6m, 5 degrees)	98.2	—	Unverified
3	FCGF	Recall (0.6m, 5 degrees)	98.2	—	Unverified
4	PCAM-soft + ICP	Recall (0.6m, 5 degrees)	98	—	Unverified
5	PCAM-Sparse + ICP	Recall (0.6m, 5 degrees)	97.4	—	Unverified
6	PCAM - Soft	Recall (0.6m, 5 degrees)	97.2	—	Unverified
7	DGR (RE (all), TE(all) are reported in PCAM)	Recall (0.6m, 5 degrees)	96.9	—	Unverified
8	PCAM-Sparse	Recall (0.6m, 5 degrees)	96.5	—	Unverified
9	Exhaustive Grid Search	Recall (0.6m, 5 degrees)	94.59	—	Unverified
10	RANSAC	Recall (0.6m, 5 degrees)	34.2	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Greedy Grid Search	Recall (3cm, 10 degrees)	92.81	—	Unverified
2	GeoTransformer	Recall (3cm, 10 degrees)	56.15	—	Unverified
3	DIP	Recall (3cm, 10 degrees)	54.81	—	Unverified
4	FCGF + PointDSC	Recall (3cm, 10 degrees)	47.85	—	Unverified
5	SpinNet	Recall (3cm, 10 degrees)	42.46	—	Unverified
6	FCGF + YOHO-C	Recall (3cm, 10 degrees)	29.18	—	Unverified
7	FCGF + YOHO-O	Recall (3cm, 10 degrees)	18.91	—	Unverified
8	FPFH-8M	Recall (3cm, 10 degrees)	9.51	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	99.88	—	Unverified
2	GeDi	Recall (3cm, 10 degrees)	99.64	—	Unverified
3	Exhaustive Grid Search	Recall (3cm, 10 degrees)	99.47	—	Unverified
4	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	98.52	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	63.94	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	38.85	—	Unverified
2	Exhaustive Grid Search	Recall (3cm, 10 degrees)	37.06	—	Unverified
3	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	17.8	—	Unverified
4	GeDi	Recall (3cm, 10 degrees)	8.7	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	2.64	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Exhaustive Grid Search	Recall (3cm, 10 degrees)	88.06	—	Unverified
2	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	84.7	—	Unverified
3	GeDi	Recall (3cm, 10 degrees)	75.4	—	Unverified
4	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	63	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	22.07	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeDi	Recall (3cm, 10 degrees)	99.76	—	Unverified
2	Exhaustive Grid Search	Recall (3cm, 10 degrees)	99.64	—	Unverified
3	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	99.64	—	Unverified
4	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	98.46	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	64.12	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeDi	Recall (3cm, 10 degrees)	99.41	—	Unverified
2	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	85.77	—	Unverified
3	Exhaustive Grid Search	Recall (3cm, 10 degrees)	78	—	Unverified
4	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	75.21	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	47.75	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeDi	Recall (3cm, 10 degrees)	99.94	—	Unverified
2	Exhaustive Grid Search	Recall (3cm, 10 degrees)	97.92	—	Unverified
3	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	94.54	—	Unverified
4	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	91.93	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	55.93	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0.06	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	99.76	—	Unverified
2	Exhaustive Grid Search	Recall (3cm, 10 degrees)	99.7	—	Unverified
3	GeDi	Recall (3cm, 10 degrees)	99.47	—	Unverified
4	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	98.34	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	66.25	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeDi	Recall (3cm, 10 degrees)	99.7	—	Unverified
2	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	99.58	—	Unverified
3	Exhaustive Grid Search	Recall (3cm, 10 degrees)	98.81	—	Unverified
4	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	98.22	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	64.18	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Exhaustive Grid Search	Recall (3cm, 10 degrees)	99.82	—	Unverified
2	GeDi	Recall (3cm, 10 degrees)	99.7	—	Unverified
3	FPFH+SC2-PCR	Recall (3cm, 10 degrees)	99.53	—	Unverified
4	FCGF+SC2-PCR	Recall (3cm, 10 degrees)	98.34	—	Unverified
5	GeoTransformer	Recall (3cm, 10 degrees)	64.29	—	Unverified
6	GO-ICP	Recall (3cm, 10 degrees)	0	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeDi	Success Rate	99.82	—	Unverified
2	D3Feat-pred	Success Rate	99.81	—	Unverified
3	SpinNet	Success Rate	99.1	—	Unverified
4	DIP	Success Rate	97.3	—	Unverified
5	FCGF	Success Rate	96.57	—	Unverified
6	3DFeat-Net	Success Rate	95.97	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UMERegRobust	RR@(1.5,0.3)	81.1	—	Unverified
2	GeoTransformer	RR@(1.5,0.3)	78.5	—	Unverified
3	CoFiNet	RR@(1.5,0.3)	62.5	—	Unverified
4	PREDATOR	RR@(1.5,0.3)	41.6	—	Unverified
5	GCL	RR@(1.5,0.3)	40.1	—	Unverified
6	FCGF	RR@(1.5,0.3)	11.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GeDi	Recall	0.92	—	Unverified
2	SpinNet	Recall	0.85	—	Unverified
3	D3Feat-pred	Recall	0.63	—	Unverified
4	FCGF	Recall	0.33	—	Unverified
5	FPFH	Recall	0.14	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GCL+KPConv	mRR @ Normal Criterion (1.5°&0.3m)	88.8	—	Unverified
2	GCL+Conv	mRR @ Normal Criterion (1.5°&0.3m)	83.5	—	Unverified
3	Predator+APR(a)	mRR @ Normal Criterion (1.5°&0.3m)	78.2	—	Unverified
4	FCGF+APR(s)	mRR @ Normal Criterion (1.5°&0.3m)	62.9	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	GCL+KPConv	mRR @ Normal Criterion (1.5°&0.3m)	71.5	—	Unverified
2	GCL+Conv	mRR @ Normal Criterion (1.5°&0.3m)	70.2	—	Unverified
3	FCGF+APR(s)	mRR @ Normal Criterion (1.5°&0.3m)	62.9	—	Unverified
4	Predator+APR(a)	mRR @ Normal Criterion (1.5°&0.3m)	52.2	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CoFF	Recall ( correspondence RMSE below 0.2)	78.7	—	Unverified
2	GeoTransformer	Recall ( correspondence RMSE below 0.2)	73.4	—	Unverified
3	Predator	Recall ( correspondence RMSE below 0.2)	63.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SGAligner	CD	0.01	—	Unverified
2	SG-PGM	CD	0.01	—	Unverified