List your company
ResearchFormerly WizWand

3D Hand Pose Estimation on NYU (test)

7.06Mean Error (mm)HandMCM
6.309611.374816.4421.5052Nov 20, 2017Apr 3, 2019Aug 14, 2020Dec 27, 2021May 10, 2023Sep 20, 2024Feb 2, 2026
Updated 8mo ago

Evaluation results

MethodLinks
HandMCMInput=Depth2026.027.06
HandDAGTInput=Depth2026.027.12
IPNetInput=Depth2026.027.17
HandR2N2Input=Point2026.027.27
SSRNInput=Depth2026.027.37
HandDiffInput=Depth2026.027.38
PHGInput=Depth2026.027.39
HandGCNFormerInput=Point2026.027.43
OursLabel Usage=100%, Augmented Set=No2023.038.01
Ourstraining_annotations=100%2023.038.01
OursLabel Usage=75%, Augmented Set=No2023.038.06
OursLabel Usage=50%, Augmented Set=No2023.038.11
OursLabel Usage=25%, Augmented Set=No2023.038.14
Ourstraining_annotations=25%2023.038.14
OursNumber of Labeled Examples=10,0002023.038.16
JGR-P2O2023.038.29
Fang et al.FPS(s)=111.202023.048.29
JGR-P2OInput=Depth2026.028.29
HambaInput=Depth2026.028.38
V2V-PoseNet2017.118.42
HandFoldingNet2023.038.42
V2VInput=Voxel2026.028.42
A2J-TransformerFPS(s)=24.812023.048.43
V2V-PoseNet2023.038.58
HandFoldingInput=Point2026.028.58
OursNumber of Labeled Examples=1,0002023.038.6
A2JFPS=105.062019.088.61
A2J2023.038.61
Xiong et al.FPS(s)=105.062023.048.61
OursLabel Usage=1%, Augmented Set=No2023.038.71
P2PFPS=41.82019.089.045
Point-to-Point2023.039.1
Point-to-PointInput=Point2026.029.1
V2VFPS=352019.089.22
Moon et al.FPS(s)=352023.049.22
NARHT2023.039.8
MURAUERNumber of Labeled Examples=10,0002023.039.9
CrossInfoNet2023.0310.08
CrossInfoNetInput=Depth2026.0210.08
DenseRegFPS=27.82019.0810.2
DenseReg2023.0310.2
DenseRegInput=Depth2026.0210.2
HandPointNetFPS=482019.0810.5
HandPointNet2023.0310.54
HandPointNetInput=Point2026.0210.54
SHPR-NetInput=Point2026.0210.78
Generalized-Feedback2023.0310.89
MURAUERNumber of Labeled Examples=1,0002023.0310.9
SO-HandNetLabel Usage=100%, Augmented Set=No2023.0311.2
Pose-REN2017.1111.81
Pose-REN2019.0811.81
Pose-REN2023.0311.81
Pose-RenInput=Depth2026.0211.81
OursNumber of Labeled Examples=1002023.0312.11
MURAUERNumber of Labeled Examples=1002023.0312.2
DeepPrior++2017.1112.24
DeepPrior++FPS=302019.0812.24
DeepPrior++2023.0312.24
DeepPrior++Input=Depth2026.0212.24
REN-9x6x62017.1112.69
REN-9x6x62019.0812.69
REN-9x6x62023.0312.69
SO-HandNetLabel Usage=75%, Augmented Set=No2023.0312.8
REN-4x6x62017.1113.39
REN-4x6x62019.0813.39
REN-4x6x62023.0313.39
3DCNN2017.1114.1
Baek et al.Label Usage=100%, Augmented Set=Yes, 10 times2023.0314.1
SO-HandNetLabel Usage=50%, Augmented Set=No2023.0314.1
3DCNN2023.0314.1
Lie-X2017.1114.51
Lie-X2019.0814.51
Baek et al.(w/o refine)Label Usage=100%, Augmented Set=Yes, 10 times2023.0314.9
SO-HandNetLabel Usage=25%, Augmented Set=No2023.0314.9
LSPSLabel Usage=100%, Augmented Set=No2023.0315.4
LSPSLabel Usage=50%, Augmented Set=No2023.0315.45
LSPSLabel Usage=75%, Augmented Set=No2023.0315.45
Crossing NetLabel Usage=100%, Augmented Set=No2023.0315.5
Global-to-Local2017.1115.6
Global-to-LocalFPS=502019.0815.6
LSPSLabel Usage=25%, Augmented Set=No2023.0315.7
Crossing NetLabel Usage=75%, Augmented Set=No2023.0315.9
Feedback2017.1115.97
Crossing NetLabel Usage=50%, Augmented Set=No2023.0316
Crossing NetLabel Usage=25%, Augmented Set=No2023.0316.1
Baek et al.(w/o aug.; refine)Label Usage=100%, Augmented Set=No2023.0316.4
MURAUERNumber of Labeled Examples=102023.0316.4
JTSC2017.1116.8
JTSC2019.0816.8
DeepModel2017.1117.04
DeepModel2019.0817.04
DeepModel2023.0317.04
Baek et al.(baseline)Label Usage=100%, Augmented Set=No2023.0317.3
Hand3D2017.1117.6
Hand3DFPS=302019.0817.6
DeepPrior2017.1119.73
DeepPrior2023.0319.73
DISCO2017.1120.7
DISCO2019.0820.7
OursNumber of Labeled Examples=102023.0325.82