3D Human Pose Estimation On Human36M

Métriques

Average MPJPE (mm)

PA-MPJPE

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Average MPJPE (mm)	PA-MPJPE	Paper Title	Repository
GTRS	64.3	45.4	A Lightweight Graph Transformer Network for Human Mesh Reconstruction from 2D Human Pose
SRNet (T=243)	44.8	-	SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach
MHFormer	43	-	MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimation
Rogez et al.	-	87.3	MoCap-guided Data Augmentation for 3D Pose Estimation in the Wild	-
Probabilistic Monocular (T=1)	61.8	-	Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flows
Modulated-GCN	49.4	-	Modulated Graph Convolutional Network for 3D Human Pose Estimation
KAMA	-	40.2	KAMA: 3D Keypoint Aware Body Mesh Articulation	-
GLA-GCN (T=243, GT)	21.0	17.6	GLA-GCN: Global-local Adaptive Graph Convolutional Network for 3D Human Pose Estimation from Monocular Video
Sequence-to-sequence network	58.5	-	Exploiting temporal information for 3D pose estimation
WSGAN	81.1	-	Weakly Supervised Generative Network for Multiple 3D Human Pose Hypotheses
Regular Splitting Graph Network	47	38.6	Regular Splitting Graph Network for 3D Human Pose Estimation
Pavlakos et al.	-	75.9	Learning to Estimate 3D Human Pose and Shape from a Single Color Image	-
HEMlets Pose	45.1	-	HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimation	-
ONS LCN	52.7	-	Optimizing Network Structure for 3D Human Pose Estimation	-
3DHPM	63.67	-	3D Human Pose Machines with Self-supervised Learning
LWCDR (extra train data)	21.0	-	Lightweight Multi-View 3D Pose Estimation through Camera-Disentangled Representation	-
SmartEdgeSensor (H36M+COCO)	23.5	-	Real-Time Multi-View 3D Human Pose Estimation using Semantic Feedback to Smart Edge Sensors
RemoCap	52.8	33.4	RemoCap: Disentangled Representation Learning for Motion Capture	-
Sparseness Meets Deepness	113.01	-	Sparseness Meets Deepness: 3D Human Pose Estimation from Monocular Video
UniHPE-w48	50.5	36.2	UniHPE: Towards Unified Human Pose Estimation via Contrastive Learning	-

0 of 357 row(s) selected.