3D Hand Pose Estimation On Freihand

Metrics

PA-F@15mm

PA-F@5mm

PA-MPJPE

PA-MPVPE

Results

Performance results of various models on this benchmark

Model Name	PA-F@15mm	PA-F@5mm	PA-MPJPE	PA-MPVPE	Paper Title	Repository
YoutubeHand	0.966	0.614	8.4	8.6	Weakly-Supervised Mesh-Convolutional Hand Reconstruction in the Wild
I2UV-HandNet	0.977	0.707	6.7	6.9	I2UV-HandNet: Image-to-UV Prediction Network for Accurate and High-fidelity 3D Hand Mesh Modeling	-
SAR	0.981	0.724	6.3	6.5	-	-
Deformer	0.984	0.743	6.2	6.4	Deformable Mesh Transformer for 3D Human Mesh Recovery
FastViT-MA36	0.981	0.722	6.6	6.7	FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization
MANO CNN	0.934	0.516	11.0	10.9	FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from Single RGB Images	-
METRO	0.984	0.731	6.5	6.3	End-to-End Human Pose and Mesh Reconstruction with Transformers
ExPose (hand sub-network h)	0.918	0.484	12.2	11.8	Monocular Expressive Body Regression through Body-Driven Attention
Hamba	0.992	0.806	5.7	5.3	Hamba: Single-view 3D Hand Reconstruction with Graph-guided Bi-Scanning Mamba
WiLoR	0.993	0.825	5.5	5.1	WiLoR: End-to-end 3D Hand Localization and Reconstruction in-the-wild	-
MaskHand	0.991	0.801	5.5	5.4	MMHMR: Generative Masked Modeling for Hand Mesh Recovery	-
HIU-DMTL	0.974	0.699	7.1	7.3	Hand Image Understanding via Deep Multi-Task Learning
Zimmermann et al.	0.935	0.529	-	10.7	FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from Single RGB Images	-
MobRecon	0.986	0.784	5.7	5.8	MobRecon: Mobile-Friendly Hand Mesh Reconstruction from Monocular Image
Tang et al.	0.981	0.724	6.7	6.7	Towards Accurate Alignment in Real-time 3D Hand-Mesh Reconstruction	-
PeCLR	-	-	6.6	-	PeCLR: Self-Supervised 3D Hand Pose Estimation from monocular RGB via Equivariant Contrastive Learning
DetNet	-	-	15.7	-	Monocular Real-time Full Body Capture with Inter-part Correlations	-
Hand4Whole	0.971	0.664	7.7	7.7	Accurate 3D Hand Pose Estimation for Whole-Body 3D Human Mesh Estimation
Pose2Mesh	0.969	0.674	7.7	7.8	Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose
AMVUR	0.987	0.767	6.2	6.1	A Probabilistic Attention Model with Occlusion-aware Texture Regression for 3D Hand Reconstruction from a Single RGB Image

0 of 32 row(s) selected.