# Pose Estimation on AP-10K 2021 (test)

Source: [Pose Estimation on AP-10K 2021 (test)](<https://www.sota2.com/research/sota/pose-estimation-on-ap-10k-2021-test>)

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

Data updated: 2026-02-26T05:22:45.153Z

Evaluation of Pose Estimation performance on the test split of the AP-10K 2021 dataset, reporting standard AP and AR metrics.

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 | Pose estimation |
| Dataset (including any stated split) | AP-10K 2021 (test) |
| Metric | AP |
| Unit | PERCENT |
| Direction | Higher 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

- [AP](<https://www.sota2.com/research/sota/pose-estimation-on-ap-10k-2021-test.md?metric=AP>): Higher is better; unit: PERCENT.

- [AP50](<https://www.sota2.com/research/sota/pose-estimation-on-ap-10k-2021-test.md?metric=AP50>): Higher is better; unit: PERCENT.

- [AP75](<https://www.sota2.com/research/sota/pose-estimation-on-ap-10k-2021-test.md?metric=AP75>): Higher is better; unit: PERCENT.

- [AR](<https://www.sota2.com/research/sota/pose-estimation-on-ap-10k-2021-test.md?metric=AR>): Higher is better; unit: PERCENT.

- [AR50](<https://www.sota2.com/research/sota/pose-estimation-on-ap-10k-2021-test.md?metric=AR50>): Higher is better; unit: PERCENT.

- [AR75](<https://www.sota2.com/research/sota/pose-estimation-on-ap-10k-2021-test.md?metric=AR75>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP | Result date | Report paper |
| --- | --- | --- | --- |
| ViTAEv2-S | 71.8 | Not recorded | 68fa6e0c1e209131cea3dacf |
| Swin-T | 68.9 | Not recorded | 68fa6e0c1e209131cea3dacf |
| ResNet | 68.1 | Not recorded | 68fa6e0c1e209131cea3dacf |

### ViTAEv2-S

AP: 71.8. Result ID: 698ff6d33a6c314849fd1f37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e0c1e209131cea3dacf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e0c1e209131cea3dacf.md>)

Recorded evaluation configuration: {"Backbone":"ViTAEv2-S","Framework":"SimpleBaseline","Number of Parameters":"23.1 M","Training Epochs":210,"Image size":"256 x 256","Optimizer":"Adam"}. This record does not assert independent reproduction.

### Swin-T

AP: 68.9. Result ID: 698ff6d33a6c314849fd1f36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e0c1e209131cea3dacf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a8c1e209131cea28b62.md>)

[Implementation](<https://github.com/microsoft/Swin-Transformer>)

Recorded evaluation configuration: {"Backbone":"Swin-T","Framework":"SimpleBaseline","Number of Parameters":"32.8 M","Training Epochs":210,"Image size":"256 x 256","Optimizer":"Adam"}. This record does not assert independent reproduction.

### ResNet

AP: 68.1. Result ID: 698ff6d33a6c314849fd1f35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e0c1e209131cea3dacf.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Framework":"SimpleBaseline","Number of Parameters":"34.1 M","Training Epochs":210,"Image size":"256 x 256","Optimizer":"Adam"}. This record does not assert independent reproduction.

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