# Multi-person 3D Pose Estimation on Association Dataset (test)

Source: [Multi-person 3D Pose Estimation on Association Dataset (test)](<https://www.sota2.com/research/sota/multi-person-3d-pose-estimation-on-association-dataset-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:58:19.278Z

Evaluation of Multi-person 3D Pose Estimation performance on the test split of the Association Dataset.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Multi-person 3D Pose Estimation |
| Dataset (including any stated split) | Association Dataset (test) |
| Metric | Precision |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Precision](<https://www.sota2.com/research/sota/multi-person-3d-pose-estimation-on-association-dataset-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/multi-person-3d-pose-estimation-on-association-dataset-test.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/multi-person-3d-pose-estimation-on-association-dataset-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| Zhang et al. | 97.1 | Not recorded | 68fa6aee1e209131cea2b21e |
| Shape-aware Multi-Person Pose Estimation | 90.1 | Not recorded | 68fa6aee1e209131cea2b21e |
| Shape-aware Multi-Person Pose Estimation | 83.7 | Not recorded | 68fa6aee1e209131cea2b21e |
| Dong et al. | 71 | Not recorded | 68fa6aee1e209131cea2b21e |
| VoxelPose | 68.8 | Not recorded | 68fa6aee1e209131cea2b21e |
| VoxelPose | 55.1 | Not recorded | 68fa6aee1e209131cea2b21e |

### Zhang et al.

Precision: 97.1. Result ID: 698e4235ffc3901de09c17d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67211e209131cea1481a.md>)

Recorded evaluation configuration: {"discards temporal information":true}. This record does not assert independent reproduction.

### Shape-aware Multi-Person Pose Estimation

Precision: 90.1. Result ID: 698e4235ffc3901de09c17d4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Implementation](<https://ait.ethz.ch/projects/2021/multi-human-pose/>)

Recorded evaluation configuration: {"optimization stage":"final"}. This record does not assert independent reproduction.

### Shape-aware Multi-Person Pose Estimation

Precision: 83.7. Result ID: 698e4235ffc3901de09c17d3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Implementation](<https://ait.ethz.ch/projects/2021/multi-human-pose/>)

Recorded evaluation configuration: {"optimization stage":"init"}. This record does not assert independent reproduction.

### Dong et al.

Precision: 71. Result ID: 698e4235ffc3901de09c17d2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654d1e209131cea09772.md>)

[Implementation](<https://zju-3dv.github.io/mvpose/>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### VoxelPose

Precision: 68.8. Result ID: 698e4235ffc3901de09c17d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68351e209131cea1acb8.md>)

[Implementation](<https://github.com/microsoft/voxelpose-pytorch>)

Recorded evaluation configuration: {"uses 3D bounding box ground truth":true}. This record does not assert independent reproduction.

### VoxelPose

Precision: 55.1. Result ID: 698e4235ffc3901de09c17cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6aee1e209131cea2b21e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68351e209131cea1acb8.md>)

[Implementation](<https://github.com/microsoft/voxelpose-pytorch>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.