Browse › Computer Vision › Pose Estimation › MPII Human Pose

Pose Estimation archive 2025-07-28

MPII Human Pose Benchmark (Pose Estimation)

46 rows 37 with code listed 1 metric Dataset page

Pose Estimation is a computer vision task where the goal is to detect the position and orientation of a person or an object. Usually, this is done by predicting the location of specific keypoints like hands, head, elbows, etc. in case of Human Pose Estimation.

A common benchmark for this task is MPII Human Pose

The archive carries no text for this table; the description above is the archive's text for the task Pose Estimation. archive 2025-07-28

Over time archive 2025-07-28

The chart needs JavaScript; the table below carries every value.

Not inferred: PCKh-0.5. Points are placed at the row's paper date; 46 of 46 rows carry one.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 PCT (swin-l, test set) 94.3 – Paper Code 2023 2 of 2 ran · 0 unverified report
2 Soft-gated Skip Connections 94.1 ✓ Paper Code 2020 linked, not harvested report
3 Cascade Feature Aggregation 93.9 ✓ Paper – 2019 no code linked report
4 PCT (swin-b, test set) 93.8 – Paper Code 2023 2 of 2 ran · 0 unverified report
5 TransPose 93.5 ✓ Paper Code 2020 linked, not harvested report
6 UniHCP (FT) 93.2 ✓ Paper Code 2023 7 of 13 ran · 6 unverified report
7 4xRSN-50 93.0 – Paper Code 2020 2 of 3 ran · 1 unverified report
8 UniPose 92.7 – Paper Code 2020 linked, not harvested report
9 MSPN 92.6 – Paper Code 2019 linked, not harvested report
10 Spatial Context 92.5 – Paper – 2019 no code linked report
11 HRNet-W32 92.3 – Paper Code 2019 8 of 25 ran · 17 unverified report
12 DLCM 92.3 – Paper – 2018 no code linked report
13 Multi-Scale Structure-Aware Network 92.1 – Paper – 2018 no code linked report
14 Pyramid Residual Modules (PRMs) 92.0 – Paper Code 2017 linked, not harvested report
15 Chen et al. ICCV'17 91.9 – Paper Code 2017 linked, not harvested report
16 Chou et al. arXiv'17 91.8 – Paper Code 2017 linked, not harvested report
17 Multi-Context Attention 91.5 – Paper Code 2017 linked, not harvested report
18 Residual Hourglass +ASR+AHO 91.5 – Paper – 2018 no code linked report
19 DU-Net 91.2 – Paper Code 2018 linked, not harvested report
20 Stacked hourglass + Inception-resnet 91.2 – Paper Code 2017 linked, not harvested report
21 EfficientPose IV 91.2 – Paper Code 2020 linked, not harvested report
22 FPD 91.1 – Paper Code 2018 linked, not harvested report
23 Integral Regression 91.0 – Paper Code 2017 linked, not harvested report
24 Stacked Hourglass Networks 90.9 – Paper Code 2016 2 of 25 ran · 23 unverified report
25 DarkPose 90.6 – Paper Code 2019 5 of 9 ran · 4 unverified report
26 TFPose(ResNet-50) 90.4 ✓ Paper – 2021 no code linked report
27 Human Pose Estimation 89.7 – Paper Code 2016 linked, not harvested report
28 Part heatmap regression (ResNet-152) 89.7 – Paper Code 2016 linked, not harvested report
29 DSNTr ResNet-50@28px 89.5 – Paper Code 2018 linked, not harvested report
30 CU-Net 89.4 – Paper Code 2018 linked, not harvested report
31 OpenPose 88.8 – Paper Code 2020 linked, not harvested report
32 Convolutional Pose Machines 88.52 – Paper Code 2016 1 of 4 ran · 3 unverified report
33 ResNet-152 + intermediate supervision 88.52 – Paper Code 2016 linked, not harvested report
34 Dite-HRNet-30 87.6 – Paper Code 2022 linked, not harvested report
35 Tucker T-Net 87.5 – Paper – 2019 no code linked report
36 CHPR 86.4 – Paper Code 2017 linked, not harvested report
37 Lifshitz et al. 85.0 – Paper – 2016 no code linked report
38 EfficientPose RT 84.8 – Paper Code 2020 linked, not harvested report
39 Matrix 82.5 – Paper – 2019 no code linked report
40 DeepCut 82.40 – Paper Code 2015 linked, not harvested report
41 QP2 82.4 – Paper Code 2015 linked, not harvested report
42 Tompson et al. 82.0 – Paper Code 2014 linked, not harvested report
43 x 81.3 – Paper Code 2018 linked, not harvested report
44 IEF 81.3 – Paper Code 2015 linked, not harvested report
45 QP1 81.1 – Paper Code 2015 linked, not harvested report
46 Improved Binary Network (HourGlass) 80.9 – Paper Code 2019 linked, not harvested report

All 46 rows shown. 46 link to a paper page on this site; 5 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 8 rows have a graph line, from 7 distinct papers; 8 rows (7 papers) have at least one sample that ran. Counting each paper once: Syntology ran 27 of 81 samples; 54 unverified. Separately, 4 of those 81 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections