ML Research Wiki / Benchmarks / Pose Estimation / CrowdPose

CrowdPose

Pose Estimation Benchmark

Performance Over Time

📊 Showing 12 results | 📏 Metric: AP

Top Performing Models

Rank	Model	Paper	AP	Date	Code
1	BUCTD-W48 (w/cond. input from PETR, and generative sampling)	Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity	78.50	2023-06-13	📦 amathislab/BUCTD
2	ViTPose-G	ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation	78.30	2022-04-26	📦 huggingface/transformers 📦 vitae-transformer/vitpose 📦 vitae-transformer/qformer
3	BUCTD-W48 (w/cond. input from PETR)	Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity	76.70	2023-06-13	📦 amathislab/BUCTD
4	SwinV2-L 1K-MIM	Revealing the Dark Secrets of Masked Image Modeling	75.50	2022-05-26	📦 SwinTransformer/MIM-Depth-Estimation
5	SwinV2-B 1K-MIM	Revealing the Dark Secrets of Masked Image Modeling	74.90	2022-05-26	📦 SwinTransformer/MIM-Depth-Estimation
6	BUCTD-W48	Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity	72.90	2023-06-13	📦 amathislab/BUCTD
7	OpenPifPaf	OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association	70.50	2021-03-03	📦 openpifpaf/openpifpaf 📦 openpifpaf/openpifpafwebdemo 📦 vita-epfl/openpifpaf
8	MIPNet (HRNet-W48)	Multi-Instance Pose Networks: Rethinking Top-Down Pose Estimation	70.00	2021-01-27	📦 rawalkhirodkar/MIPNet
9	KAPAO-L	Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation	68.90	2021-11-16	📦 wmcnally/kapao
10	KAPAO-M	Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation	67.10	2021-11-16	📦 wmcnally/kapao

All Papers (12)

Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity

2023

BUCTD-W48 (w/cond. input from PETR, and generative sampling)

amathislab/BUCTD

ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation

2022

ViTPose-G

huggingface/transformers vitae-transformer/vitpose

Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity

2023

BUCTD-W48 (w/cond. input from PETR)

amathislab/BUCTD

Revealing the Dark Secrets of Masked Image Modeling

2022

SwinV2-L 1K-MIM

SwinTransformer/MIM-Depth-Estimation

Revealing the Dark Secrets of Masked Image Modeling

2022

SwinV2-B 1K-MIM

SwinTransformer/MIM-Depth-Estimation

Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity

2023

BUCTD-W48

amathislab/BUCTD

OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association

2021

OpenPifPaf

openpifpaf/openpifpaf openpifpaf/openpifpafwebdemo

Multi-Instance Pose Networks: Rethinking Top-Down Pose Estimation

2021

MIPNet (HRNet-W48)

rawalkhirodkar/MIPNet

Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation

2021

KAPAO-L

wmcnally/kapao

Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation

2021

KAPAO-M

wmcnally/kapao

Greedy Offset-Guided Keypoint Grouping for Human Pose Estimation

2021

Hourglass-104

hellojialee/OffsetGuided

Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation

2021

KAPAO-S

wmcnally/kapao

CrowdPose

Performance Over Time

Edit Benchmark Results

Edit Result

Top Performing Models

All Papers (12)

Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity

ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation

Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity

Revealing the Dark Secrets of Masked Image Modeling

Revealing the Dark Secrets of Masked Image Modeling

Rethinking pose estimation in crowds: overcoming the detection information-bottleneck and ambiguity

OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association

Multi-Instance Pose Networks: Rethinking Top-Down Pose Estimation

Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation

Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation

Greedy Offset-Guided Keypoint Grouping for Human Pose Estimation

Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation

Model	Paper	AP	Date
BUCTD-W48 (w/cond. input from PETR, and generative sampling)	Rethinking pose estimation in crowds: overcoming …	78.50	2023-06-13
ViTPose-G	ViTPose: Simple Vision Transformer Baselines for …	78.30	2022-04-26
BUCTD-W48 (w/cond. input from PETR)	Rethinking pose estimation in crowds: overcoming …	76.70	2023-06-13
SwinV2-L 1K-MIM	Revealing the Dark Secrets of Masked Image Modeli…	75.50	2022-05-26
SwinV2-B 1K-MIM	Revealing the Dark Secrets of Masked Image Modeli…	74.90	2022-05-26
BUCTD-W48	Rethinking pose estimation in crowds: overcoming …	72.90	2023-06-13
OpenPifPaf	OpenPifPaf: Composite Fields for Semantic Keypoin…	70.50	2021-03-03
MIPNet (HRNet-W48)	Multi-Instance Pose Networks: Rethinking Top-Down…	70.00	2021-01-27
KAPAO-L	Rethinking Keypoint Representations: Modeling Key…	68.90	2021-11-16
KAPAO-M	Rethinking Keypoint Representations: Modeling Key…	67.10	2021-11-16
Hourglass-104	Greedy Offset-Guided Keypoint Grouping for Human …	65.20	2021-07-07
KAPAO-S	Rethinking Keypoint Representations: Modeling Key…	63.80	2021-11-16