# 3D Pose Estimation on MM-Fi Setting 2 - Cross-Subject

Source: [3D Pose Estimation on MM-Fi Setting 2 - Cross-Subject](<https://www.sota2.com/research/sota/3d-pose-estimation-on-mm-fi-setting-2-cross-subject>)

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

Data updated: 2026-04-08T07:06:56.056Z

Evaluation of 3D Pose Estimation performance on the MM-Fi Setting 2 (Cross-Subject) configuration, reporting metrics such as PCK thresholds and Mean Per Joint Position Error (MPJPE) variants.

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 | 3D Pose Estimation |
| Dataset (including any stated split) | MM-Fi Setting 2 - Cross-Subject |
| Metric | PCK@20 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [PCK@20](<https://www.sota2.com/research/sota/3d-pose-estimation-on-mm-fi-setting-2-cross-subject.md?metric=PCK%4020>): Higher is better; unit: PERCENT.

- [PCK@50](<https://www.sota2.com/research/sota/3d-pose-estimation-on-mm-fi-setting-2-cross-subject.md?metric=PCK%4050>): Higher is better; unit: PERCENT.

- [MPJPE](<https://www.sota2.com/research/sota/3d-pose-estimation-on-mm-fi-setting-2-cross-subject.md?metric=MPJPE>): Lower is better; unit: SCALAR.

- [PA-MPJPE](<https://www.sota2.com/research/sota/3d-pose-estimation-on-mm-fi-setting-2-cross-subject.md?metric=PA-MPJPE>): Lower is better; unit: SCALAR.

## Recorded results

| Method | PCK@20 | Result date | Report paper |
| --- | --- | --- | --- |
| DT-Pose | 41.9 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| HPE-Li | 38.2 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| DT-Pose | 37.7 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| HPE-Li | 36.5 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| MetaFi++ | 36.4 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| MetaFi++ | 32.3 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| DT-Pose | 28.5 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| HPE-Li | 26.9 | Not recorded | 68fa76ae1e209131cea6e6b0 |
| MetaFi++ | 24 | Not recorded | 68fa76ae1e209131cea6e6b0 |

### DT-Pose

PCK@20: 41.9. Result ID: 69d5fd1c47f34ce924be09ac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 1"}. This record does not assert independent reproduction.

### HPE-Li

PCK@20: 38.2. Result ID: 69d5fd1c47f34ce924be09a9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 1"}. This record does not assert independent reproduction.

### DT-Pose

PCK@20: 37.7. Result ID: 69d5fd1c47f34ce924be09ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 3"}. This record does not assert independent reproduction.

### HPE-Li

PCK@20: 36.5. Result ID: 69d5fd1c47f34ce924be09ab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 3"}. This record does not assert independent reproduction.

### MetaFi++

PCK@20: 36.4. Result ID: 69d5fd1c47f34ce924be09a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 1"}. This record does not assert independent reproduction.

### MetaFi++

PCK@20: 32.3. Result ID: 69d5fd1c47f34ce924be09a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 3"}. This record does not assert independent reproduction.

### DT-Pose

PCK@20: 28.5. Result ID: 69d5fd1c47f34ce924be09ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 2"}. This record does not assert independent reproduction.

### HPE-Li

PCK@20: 26.9. Result ID: 69d5fd1c47f34ce924be09aa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 2"}. This record does not assert independent reproduction.

### MetaFi++

PCK@20: 24. Result ID: 69d5fd1c47f34ce924be09a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76ae1e209131cea6e6b0.md>)

Recorded evaluation configuration: {"Protocol":"Protocol 2"}. This record does not assert independent reproduction.

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