الرئيسية الأخبار أحدث الأوراق البحثية الدروس مجموعات البيانات الموسوعة SOTA نماذج LLM لوحة الأداء GPU الفعاليات

العربية

2D Human Pose Estimation On Coco Wholebody 1

المقاييس

WB

body

face

foot

hand

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

اسم النموذج	WB	body	face	foot	hand	Paper Title	Repository
Sapiens-2B	74.4	79.2	91.2	84.1	70.4	Sapiens: Foundation for Human Vision Models	-
ZoomNet (V1.0 data)	63.0	74.5	88.0	60.9	57.9	ZoomNAS: Searching for Whole-body Human Pose Estimation in the Wild	-
RTMW-x	70.2	76.3	88.4	79.6	66.4	RTMW: Real-Time Multi-Person 2D and 3D Whole-body Pose Estimation	-
RTMW-m	58	67.6	78.3	67.1	49.1	RTMW: Real-Time Multi-Person 2D and 3D Whole-body Pose Estimation	-
HPRNet	34.8	59.4	75.4	53.0	50.4	HPRNet: Hierarchical Point Regression for Whole-Body Human Pose Estimation	-
AE	27.4	40.5	47.7	7.7	34.1	Associative Embedding: End-to-End Learning for Joint Detection and Grouping	-
ZoomNAS (V1.0 data)	65.4	74.0	88.9	61.7	62.5	ZoomNAS: Searching for Whole-body Human Pose Estimation in the Wild	-
Sapiens-0.3B	62.0	66.4	87.1	67.3	58.1	Sapiens: Foundation for Human Vision Models	-
PCNet	66.4	76.9	89.7	65.3	65.6	-	-
ZoomNet (V0.5 data)	54.1	74.3	62.3	79.8	40.1	Whole-Body Human Pose Estimation in the Wild	-
Zauss et al.	60.4	69.6	85.0	63.4	52.9	Keypoint Communities	-
RTMPose	65.3	71.4	88.9	69.2	59.0	RTMPose: Real-Time Multi-Person Pose Estimation based on MMPose	-
TCFormer	64.2	71.8	79.0	74.4	61.4	Not All Tokens Are Equal: Human-centric Visual Analysis via Token Clustering Transformer	-
ViTPose+-H	61.2	75.9	63.3	77.9	54.7	ViTPose++: Vision Transformer for Generic Body Pose Estimation	-
OpenPose	33.8	56.3	48.2	53.2	19.8	Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields	-
HRNet	43.2	65.9	52.3	31.4	30.0	Deep High-Resolution Representation Learning for Human Pose Estimation	-

0 of 16 row(s) selected.

2D Human Pose Estimation On Coco Wholebody 1 | SOTA | HyperAI