Who leads the COCO Keypoints benchmark?

ViTPose-H currently leads COCO Keypoints with a score of 80.9 on Ap.

What is the state-of-the-art score on COCO Keypoints?

The state-of-the-art result on COCO Keypoints is 80.9 (Ap), achieved by ViTPose-H as of 2023.

How many models are tracked on COCO Keypoints?

Codesota tracks 4 models on COCO Keypoints across 2 metrics.

When was the COCO Keypoints leaderboard last updated?

The COCO Keypoints leaderboard on Codesota includes results through 2023, with the earliest tracked result from 2019.

Codesota · Benchmark · COCO KeypointsHome/Leaderboards/COCO Keypoints

Unknown

COCO Keypoints.

Name: COCO Keypoints Benchmark Results
Creator: Unknown
Published: 2019-01-01
License: https://creativecommons.org/licenses/by/4.0/

Human pose estimation on COCO with 17 body keypoints

Paper ↗Leaderboard ↓

§ 01 · SOTA history

Year over year.

§ 02 · Leaderboard

Results by metric.

Only 4 models on this benchmark

Help build the community leaderboard — submit your model results.

Found a wrong score or missing run?

Use row edits to send a sourced correction into moderation.

Add / edit result ↗Report issue ↗

Ap

Ap is the reported evaluation metric for COCO Keypoints. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Apverifiedpapervendorcommunityunverified

Muted rows were not state of the art when published — an earlier or same-year result already scored better.

Rank	Model	Trust	Score	Year	Links	Fix
01	ViTPose-H ViTPose-H (ViT-Huge backbone). Top-down pose estimation. AP on COCO test-dev. NeurIPS 2022. Table 3.	verified	80.9	2022	Source ↗	Looks wrong?
02	RTMPose-X RTMPose-X. Real-Time Multi-Person Pose Estimation. AP on COCO test-dev. OpenMMLab, 2023. Table 2.	verified	78.8	2023	Source ↗	Looks wrong?
03	HRNet-W48 HRNet-W48 top-down. AP on COCO test-dev. Microsoft Research, CVPR 2019. Table 5.	verified	75.5	2019	Source ↗	Looks wrong?

Map

Map is the reported evaluation metric for COCO Keypoints. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Mapverifiedpapervendorcommunityunverified

Muted rows were not state of the art when published — an earlier or same-year result already scored better.

Rank	Model	Trust	Score	Year	Links	Fix
01	ViTPose-G	unverified	80.9	2022	Paper ↗Code ↗	Looks wrong?

§ 04 · Submit a result

Add to the leaderboard.

← Back to Leaderboards