3D Human Pose Estimation On Human36M

Metrics

Average MPJPE (mm)

PA-MPJPE

Results

Performance results of various models on this benchmark

			Paper Title
Pyramid of 3D HOG features	125.28	-	Predicting people’s 3D poses from short sequences
Structured Prediction	125.0	-	Structured Prediction of 3D Human Pose with Deep Neural Networks
StructNet-Avg(500)-APF	121.31	-	Maximum-Margin Structured Learning with Deep Networks for 3D Human Pose Estimation
3DCNN	119	-	Human Pose Estimation in Space and Time using 3D CNN
CNN Lifter	117.34	-	3D Human Pose Estimation Using Convolutional Neural Networks with 2D Pose Information
Events	116.4	-	Lifting Monocular Events to 3D Human Poses
Sparseness Meets Deepness	113.01	-	Sparseness Meets Deepness: 3D Human Pose Estimation from Monocular Video
Deep Kinematic Pose	107.26	-	Deep Kinematic Pose Regression
Embodied Scene-aware	103.4	73.7	Embodied Scene-aware Human Pose Estimation
Dual-source approach	97.39	108.3	A Dual-Source Approach for 3D Pose Estimation from a Single Image
CHPR	92.4	67.5	Compositional Human Pose Regression
HUND (SS)	91.8	66	Neural Descent for Visual 3D Human Pose and Shape
RepNet	89.9	-	RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation
Projected-pose belief maps + 2D fusion layers	88.39	-	Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image
HMR	87.97	58.1	End-to-end Recovery of Human Shape and Pose
LCR-Net	87.7	71.6	LCR-Net: Localization-Classification-Regression for Human Pose
THUNDR (WS)	87	62.2	THUNDR: Transformer-based 3D HUmaN Reconstruction with Markers
Ray3D (T=9 CPN H36M+HEva+3DHP)	84.4	-	Ray3D: ray-based 3D human pose estimation for monocular absolute 3D localization
HUND+SO+GT + Dynamics	84	56	Trajectory Optimization for Physics-Based Reconstruction of 3d Human Pose from Monocular Video
HMMR (T=20)	83.7	56.9	Learning 3D Human Dynamics from Video

0 of 357 row(s) selected.

Command Palette

3D Human Pose Estimation On Human36M

Metrics

Results