Research map / 3D vision / Human pose estimation
Heatmap and coordinate pose encoding
Refines keypoint encoding, decoding and probabilistic output representations for 2D human pose.
50 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Human pose estimation, keyword cluster “human pose · pose estimation · 3d human”
State History Benchmarks Tried Checklist Preprints Who Export
Where it stands
Quiet None of the tracked benchmarks has a report from 2025–26.
15 papers in 2023–24 and 8 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 5.5% to 2.7%.
No longer reported by papers here: COCO test-dev2017 (AP, last in 2023).
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Convolutional Pose Machines (CVPR 2016 · 2,831 citations)
- 2018 · DensePose: Dense Human Pose Estimation in the Wild (CVPR 2018 · 1,462 citations)
- 2019 · Deep High-Resolution Representation Learning for Human Pose Estimation (CVPR 2019 · 5,835 citations)
- 2020 · HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation (CVPR 2020 · 954 citations)
- 2021 · Bottom-Up Human Pose Estimation via Disentangled Keypoint Regression (CVPR 2021 · 393 citations)
- 2022 · Heatmap Distribution Matching for Human Pose Estimation (NeurIPS 2022 · 7 citations)
- 2023 · Human Pose As Compositional Tokens (CVPR 2023 · 91 citations)
- 2024 · SHaRPose: Sparse High-Resolution Representation for Human Pose Estimation (AAAI 2024 · 38 citations)
- 2025 · ProbPose: A Probabilistic Approach to 2D Human Pose Estimation (CVPR 2025 · 13 citations)
- 2026 · UDAPose: Unsupervised Domain Adaptation for Low-Light Human Pose Estimation (CVPR 2026 · 1 citation)
Earliest papers here
- Convolutional Pose Machines (CVPR 2016 · 2,831 citations)
- Human Pose Estimation With Iterative Error Feedback (CVPR 2016 · 856 citations)
- End-To-End Learning of Deformable Mixture of Parts and Deep Convolutional Neural Networks for Human Pose Estimation (CVPR 2016 · 288 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
COCO test-dev2017 · AP (higher is better)
4 papers, 2019–2023 · not reported since 2023
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 77 | Deep High-Resolution Representation Learning for Human Pose Estimation (CVPR 2019 · 5,835 citations) |
| 2022 | 77.2 | Heatmap Distribution Matching for Human Pose Estimation (NeurIPS 2022 · 7 citations) |
| 2023 | 78.3 | Human Pose As Compositional Tokens (CVPR 2023 · 91 citations) |
What has been tried
43 papers here report 490 measured changes (a component added, removed or swapped, compared with a baseline): helped 74%, no clear effect 8%, hurt 17%, mixed 2%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Increase α from 0.0 to 0.3 | COCO val, 256×192 · AP | 68.4 -> 74.8 | Helped | SHaRPose: Sparse High-Resolution Representation for Human Pose Estimation (AAAI 2024 · 38 citations) |
| Add part-based modeling | DIP-IMU, CIP, Natural Motion · SIP error (°) | 14.97 -> 13.67; 13.00 -> 11.67; 31.62 -> 18.88 | Helped | Dynamic Inertial Poser (DynaIP): Part-Based Motion Dynamics Learning for Enhanced Human Pose Estimation with Sparse Inertial Sensors (CVPR 2024 · 31 citations) |
| add crop data augmentation | CropCOCO · mAP | 72.4 -> 81.7 | Helped | ProbPose: A Probabilistic Approach to 2D Human Pose Estimation (CVPR 2025 · 13 citations) |
| add PSE I | Human3.6M / 3D · MPJPE | 51.2 -> 48.7 | Helped | SynSP: Synergy of Smoothness and Precision in Pose Sequences Refinement (CVPR 2024 · 7 citations) |
| add MSFDecoder instead of convolutional and deconvolutional layers | LenslessHuman3.6M · MPJPE | 142.13 -> 139.34 | Helped | LPSNet: End-to-End Human Pose and Shape Estimation with Lensless Imaging (CVPR 2024 · 5 citations) |
| add uniform position sampling | MPII, 128 × 128 input, 4 × 4 cell division · PCKh@0.5 | 80.83 -> 81.42 | Helped | Continuous Heatmap Regression for Pose Estimation via Implicit Neural Representation (NeurIPS 2024 · 5 citations) |
| Remove the pose head | RefHuman val, Text prompt · Mask AP | 62.2 -> 61.8 | Hurt | Referring Human Pose and Mask Estimation In the Wild (NeurIPS 2024 · 3 citations) |
| Replace estimated confidence with a constant | COCO val · mAP | 72.4 -> 67.5 | Hurt | On the Calibration of Human Pose Estimation (ICML 2024 · 2 citations) |
What reviewers will expect
Datasets and baselines named in the 9 papers of this approach from 2024–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- MPII · 4 of 9 papers
- Human3.6M · 3 of 9 papers
- COCO · 3 of 9 papers
- 3DPW · 3 of 9 papers
- AMASS · 2 of 9 papers
Baselines
- SimpleBaseline · 2 of 9 papers
- HRNet · 2 of 9 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 23 papers; from OpenAlex and the papers' affiliations).
- Nanjing University of Science and Technology · 2 papers
- Xidian University · 2 papers
- Tsinghua University · 2 papers
- Shanghai Jiao Tong University · 2 papers
- Nanjing University · 2 papers
- National University of Singapore · 2 papers
Use it
Download BibTeX of 35 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Monocular 2D-to-3D pose lifting (90 papers)
- Transformer-based multi-view 3D pose (83 papers)
- Multi-person pose estimation (51 papers)
- Temporal video pose estimation (76 papers)
Elsewhere in the topic
- Physics and sensor-based motion capture (125 papers)
- Learned camera pose estimation (96 papers)
- Category-level object pose estimation (92 papers)
- Depth-based hand pose regression (84 papers)
- Human shape and pose regression (80 papers)
- Instance-level 6D object pose (74 papers)
- Human mesh recovery from images (58 papers)
- Category-level pose estimation (54 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).