Point Cloud Registration on KITTI (trained on 3DMatch)

Metric: Success Rate (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Success Rate▼	Extra Data	Paper	Date↕	Code
1	GeDi	98.92	No	Learning general and distinctive 3D local deep d...	2021-05-21	Code
2	FCGF+SC2-PCR	97.66	No	-	-	-
3	FCGF+PointDSC	96.76	No	PointDSC: Robust Point Cloud Registration using ...	2021-03-09	Code
4	Exhaustive Grid Search	94.95	No	-	-	Code
5	FPFH+PointDSC	94.05	No	PointDSC: Robust Point Cloud Registration using ...	2021-03-09	Code
6	DIP	93.51	No	Distinctive 3D local deep descriptors	2020-09-01	Code
7	Greedy Grid Search	90.27	No	Challenging the Universal Representation of Deep...	2022-11-29	Code
8	YOHO-C	82.16	No	You Only Hypothesize Once: Point Cloud Registrat...	2021-09-01	Code
9	SpinNet	81.44	No	SpinNet: Learning a General Surface Descriptor f...	2020-11-24	Code
10	YOHO-O	81.44	No	You Only Hypothesize Once: Point Cloud Registrat...	2021-09-01	Code
11	GeoTransformer	67.93	No	GeoTransformer: Fast and Robust Point Cloud Regi...	2023-07-25	Code
12	Predator	41.2	No	PREDATOR: Registration of 3D Point Clouds with L...	2020-11-25	Code
13	D3Feat-pred	36.76	No	D3Feat: Joint Learning of Dense Detection and De...	2020-03-06	Code
14	FCGF	24.19	No	-	-	Code

#1GeDiSOTA
98.92
Success Rate· 2021-05-21
Learning general and distinctive 3D local deep descriptors for point cloud registration Code
#2FCGF+SC2-PCR
97.66
Success Rate
No paper
#3FCGF+PointDSCSOTA
96.76
Success Rate· 2021-03-09
PointDSC: Robust Point Cloud Registration using Deep Spatial Consistency Code
#4Exhaustive Grid Search
94.95
Success Rate
No paperCode
#5FPFH+PointDSC
94.05
Success Rate· 2021-03-09
PointDSC: Robust Point Cloud Registration using Deep Spatial Consistency Code
#6DIPSOTA
93.51
Success Rate· 2020-09-01
Distinctive 3D local deep descriptors Code
#7Greedy Grid Search
90.27
Success Rate· 2022-11-29
Challenging the Universal Representation of Deep Models for 3D Point Cloud Registration Code
#8YOHO-C
82.16
Success Rate· 2021-09-01
You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptors Code
#9SpinNet
81.44
Success Rate· 2020-11-24
SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration Code
#10YOHO-O
81.44
Success Rate· 2021-09-01
You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptors Code
#11GeoTransformer
67.93
Success Rate· 2023-07-25
GeoTransformer: Fast and Robust Point Cloud Registration with Geometric Transformer Code
#12Predator
41.2
Success Rate· 2020-11-25
PREDATOR: Registration of 3D Point Clouds with Low Overlap Code
#13D3Feat-predSOTA
36.76
Success Rate· 2020-03-06
D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features Code
#14FCGF
24.19
Success Rate
No paperCode