3D Hand Pose Estimation On Ho 3D

评估指标

AUC_J

AUC_V

F@15mm

F@5mm

PA-MPJPE (mm)

PA-MPVPE

评测结果

各个模型在此基准测试上的表现结果

模型名称	AUC_J	AUC_V	F@15mm	F@5mm	PA-MPJPE (mm)	PA-MPVPE	Paper Title	Repository
Pose2Mesh	0.754	0.749	0.909	0.441	12.5	12.7	Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose
KPT-Transformer	0.786	-	-	-	10.8	-	Keypoint Transformer: Solving Joint Identification in Challenging Hands and Object Interactions for Accurate 3D Pose Estimation
ArtiBoost	0.773	0.782	0.944	0.488	11.4	10.9	ArtiBoost: Boosting Articulated 3D Hand-Object Pose Estimation via Online Exploration and Synthesis
HFL-Net	-	-	0.965	0.575	8.9	8.7	Harmonious Feature Learning for Interactive Hand-Object Pose Estimation
WiLoR	0.851	0.846	0.983	0.646	7.5	7.7	WiLoR: End-to-end 3D Hand Localization and Reconstruction in-the-wild	-
AMVUR	0.835	0.836	0.965	0.608	8.3	8.2	A Probabilistic Attention Model with Occlusion-aware Texture Regression for 3D Hand Reconstruction from a Single RGB Image
METRO	0.792	0.779	0.946	0.484	10.4	11.1	End-to-End Human Pose and Mesh Reconstruction with Transformers
MobRecon	-	-	0.957	0.538	9.2	9.4	MobRecon: Mobile-Friendly Hand Mesh Reconstruction from Monocular Image
Tse et al.	-	-	0.943	0.485	-	10.9	Collaborative Learning for Hand and Object Reconstruction with Attention-guided Graph Convolution	-
HOISDF	-	-	-	-	9.2	-	HOISDF: Constraining 3D Hand-Object Pose Estimation with Global Signed Distance Fields
Hampali et al.	0.788	0.790	0.942	0.506	10.7	10.6	HOnnotate: A method for 3D Annotation of Hand and Object Poses
HOR	-	-	-	-	12.0	-	Towards unconstrained joint hand-object reconstruction from RGB videos	-
HandOccNet	0.819	0.819	0.963	0.564	9.1	8.8	HandOccNet: Occlusion-Robust 3D Hand Mesh Estimation Network	-
HaMeR	0.846	0.841	0.980	0.635	7.7	7.9	Reconstructing Hands in 3D with Transformers
I2UV-HandNet	0.804	0.799	0.943	0.500	9.9	10.1	I2UV-HandNet: Image-to-UV Prediction Network for Accurate and High-fidelity 3D Hand Mesh Modeling	-
Hamba	0.850	0.846	0.982	0.648	7.5	7.7	Hamba: Single-view 3D Hand Reconstruction with Graph-guided Bi-Scanning Mamba
S2Hand	0.773	0.777	0.930	0.450	11.4	11.2	Model-based 3D Hand Reconstruction via Self-Supervised Learning
H2ONet	0.829	0.828	0.966	0.570	8.5	8.6	H2ONet: Hand-Occlusion-and-Orientation-Aware Network for Real-Time 3D Hand Mesh Reconstruction
DMA	-	-	-	-	10.1	-	Interacting Hand-Object Pose Estimation via Dense Mutual Attention
HandBooster	0.836	0.832	0.972	0.585	8.2	8.4	HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions

0 of 23 row(s) selected.