Keypoint Detection On Coco Test Dev

Métriques

AP

APL

APM

Résultats

Résultats de performance de divers modèles sur ce benchmark

				Paper Title
HRNet*	-	83.1	73.4	Deep High-Resolution Representation Learning for Human Pose Estimation
Simple Base+*	-	82.7	73.0	Simple Baselines for Human Pose Estimation and Tracking
HRNet	-	81.5	71.9	Deep High-Resolution Representation Learning for Human Pose Estimation
AlphaPose	-	81.5	-	RMPE: Regional Multi-person Pose Estimation
MSPN	76.1	81.5	72.3	Rethinking on Multi-Stage Networks for Human Pose Estimation
Simple Base	-	80.0	70.3	Simple Baselines for Human Pose Estimation and Tracking
CPN+	-	78.1	69.5	Cascaded Pyramid Network for Multi-Person Pose Estimation
CPN	-	77.2	68.7	Cascaded Pyramid Network for Multi-Person Pose Estimation
OpenPifPaf	70.9	76.8	67.1	OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association
AE	-	72.6	60.6	Associative Embedding: End-to-End Learning for Joint Detection and Grouping
PifPaf (single-scale)	66.4	72.1	62.6	PifPaf: Composite Fields for Human Pose Estimation
DirectPose (ResNet-101)	64.8	71.5	60.4	DirectPose: Direct End-to-End Multi-Person Pose Estimation
Mask R-CNN	-	71.4	57.8	Mask R-CNN
Simple Pose	68.1	70.5	66.8	Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation
G-RMI	-	70.0	62.3	Towards Accurate Multi-person Pose Estimation in the Wild
CMU Pose	-	68.2	57.1	Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields

0 of 16 row(s) selected.