# 3D Human Pose Estimation on Real-world captures 40k frames

Source: [3D Human Pose Estimation on Real-world captures 40k frames](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-real-world-captures-40k-frames>)

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

Data updated: 2026-02-26T05:42:06.635Z

Evaluation of 3D Human Pose Estimation performance on the Real-world captures 40k frames dataset, reporting component-wise metrics including Epa-mpjpe, Eacc, and Evel.

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 Human Pose Estimation |
| Dataset (including any stated split) | Real-world captures 40k frames |
| Metric | Epa-mpjpe |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 3 of 3 |
| 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

- [Epa-mpjpe](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-real-world-captures-40k-frames.md?metric=Epa-mpjpe>): Lower is better; unit: SCALAR.

- [Success Rate](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-real-world-captures-40k-frames.md?metric=Success%20Rate>): Higher is better; unit: PERCENT.

- [Eacc](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-real-world-captures-40k-frames.md?metric=Eacc>): Higher is better; unit: PERCENT.

- [Evel](<https://www.sota2.com/research/sota/3d-human-pose-estimation-on-real-world-captures-40k-frames.md?metric=Evel>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Epa-mpjpe | Result date | Report paper |
| --- | --- | --- | --- |
| SimXR | 73 | Not recorded | 68fa72f81e209131cea59a67 |
| UnrealEgo | 81 | Not recorded | 68fa72f81e209131cea59a67 |
| KinPoly-v | 83 | Not recorded | 68fa72f81e209131cea59a67 |

### SimXR

Epa-mpjpe: 73. Result ID: 69928be03a6c3148490fafea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72f81e209131cea59a67.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72f81e209131cea59a67.md>)

[Implementation](<https://zhengyiluo.github.io/SimXR/>)

Recorded evaluation configuration: {"Size":"59.7MB","Physics":true}. This record does not assert independent reproduction.

### UnrealEgo

Epa-mpjpe: 81. Result ID: 69928be03a6c3148490fafe8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72f81e209131cea59a67.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d2d1e209131cea381a1.md>)

[Implementation](<https://4dqv.mpi-inf.mpg.de/UnrealEgo/>)

Recorded evaluation configuration: {"Size":"554.5MB","Physics":false}. This record does not assert independent reproduction.

### KinPoly-v

Epa-mpjpe: 83. Result ID: 69928be03a6c3148490fafe9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72f81e209131cea59a67.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dc81e209131cea3c27d.md>)

[Implementation](<https://zhengyiluo.github.io/projects/kin_poly/>)

Recorded evaluation configuration: {"Size":"98.9MB","Physics":true}. This record does not assert independent reproduction.

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