# Animal Pose Estimation on Marmosets (test)

Source: [Animal Pose Estimation on Marmosets (test)](<https://www.sota2.com/research/sota/animal-pose-estimation-on-marmosets-test>)

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

Data updated: 2026-02-26T05:22:36.819Z

Evaluation of Animal Pose Estimation performance on the Marmosets test split, reporting Average Precision (AP) 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 | Animal Pose Estimation |
| Dataset (including any stated split) | Marmosets (test) |
| Metric | AP |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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/animal-pose-estimation-on-marmosets-test.md?metric=AP>): Higher is better; unit: PERCENT.

- [AP (Large)](<https://www.sota2.com/research/sota/animal-pose-estimation-on-marmosets-test.md?metric=AP%20(Large)>): Higher is better; unit: PERCENT.

- [AP (Medium)](<https://www.sota2.com/research/sota/animal-pose-estimation-on-marmosets-test.md?metric=AP%20(Medium)>): Higher is better; unit: PERCENT.

- [AP (Heavy)](<https://www.sota2.com/research/sota/animal-pose-estimation-on-marmosets-test.md?metric=AP%20(Heavy)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AP | Result date | Report paper |
| --- | --- | --- | --- |
| BUCTD-CoAM-W48 | 93.7 | Not recorded | 68fa5f511e209131ce9e8f68 |
| BUCTD-preNet-W48 | 93.3 | Not recorded | 68fa5f511e209131ce9e8f68 |
| PETR | 93.2 | Not recorded | 68fa5f511e209131ce9e8f68 |
| CID-W32 | 92.5 | Not recorded | 68fa5f511e209131ce9e8f68 |
| HRNet-W48 | 91.6 | Not recorded | 68fa5f511e209131ce9e8f68 |
| BUCTD-CoAM-W48 | 91.6 | Not recorded | 68fa5f511e209131ce9e8f68 |
| HRNet-W48 | 91 | Not recorded | 68fa5f511e209131ce9e8f68 |
| BUCTD-preNet-W48 | 90.4 | Not recorded | 68fa5f511e209131ce9e8f68 |
| DLCRNet | 80.1 | Not recorded | 68fa5f511e209131ce9e8f68 |
| HRNet-AE | 65.1 | Not recorded | 68fa5f511e209131ce9e8f68 |
| DEKR | 61.4 | Not recorded | 68fa5f511e209131ce9e8f68 |
| ResNet-AE | 45 | Not recorded | 68fa5f511e209131ce9e8f68 |

### BUCTD-CoAM-W48

AP: 93.7. Result ID: 698bc2807a967eddb7712614.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Implementation](<https://github.com/amathislab/BUCTD>)

Recorded evaluation configuration: {"Method Category":"Hybrid","Condition Source":"PETR","Sampling":"Empirical","Backbone":"W48"}. This record does not assert independent reproduction.

### BUCTD-preNet-W48

AP: 93.3. Result ID: 698bc2807a967eddb7712613.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Implementation](<https://github.com/amathislab/BUCTD>)

Recorded evaluation configuration: {"Method Category":"Hybrid","Condition Source":"CID-W32","Sampling":"Empirical","Backbone":"W48"}. This record does not assert independent reproduction.

### PETR

AP: 93.2. Result ID: 698bc2807a967eddb771260f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

Recorded evaluation configuration: {"Method Category":"Bottom-up"}. This record does not assert independent reproduction.

### CID-W32

AP: 92.5. Result ID: 698bc2807a967eddb771260d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

Recorded evaluation configuration: {"Method Category":"Bottom-up","Backbone":"W32"}. This record does not assert independent reproduction.

### HRNet-W48

AP: 91.6. Result ID: 698bc2807a967eddb7712611.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea15032.md>)

[Implementation](<https://github.com/HRNet>)

Recorded evaluation configuration: {"Method Category":"Top-down","Detector":"Faster R-CNN","Backbone":"W48"}. This record does not assert independent reproduction.

### BUCTD-CoAM-W48

AP: 91.6. Result ID: 698bc2807a967eddb7712615.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Implementation](<https://github.com/amathislab/BUCTD>)

Recorded evaluation configuration: {"Method Category":"Hybrid","Condition Source":"DLCRNet","Sampling":"Generative (sigma)","Backbone":"W48"}. This record does not assert independent reproduction.

### HRNet-W48

AP: 91. Result ID: 698bc2807a967eddb7712610.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea15032.md>)

[Implementation](<https://github.com/HRNet>)

Recorded evaluation configuration: {"Method Category":"Top-down","Detector":"YOLOv3","Backbone":"W48"}. This record does not assert independent reproduction.

### BUCTD-preNet-W48

AP: 90.4. Result ID: 698bc2807a967eddb7712612.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Implementation](<https://github.com/amathislab/BUCTD>)

Recorded evaluation configuration: {"Method Category":"Hybrid","Condition Source":"DLCRNet","Sampling":"Empirical","Backbone":"W48"}. This record does not assert independent reproduction.

### DLCRNet

AP: 80.1. Result ID: 698bc2807a967eddb771260c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

Recorded evaluation configuration: {"Method Category":"Bottom-up"}. This record does not assert independent reproduction.

### HRNet-AE

AP: 65.1. Result ID: 698bc2807a967eddb771260b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

Recorded evaluation configuration: {"Method Category":"Bottom-up","Backbone":"HRNet"}. This record does not assert independent reproduction.

### DEKR

AP: 61.4. Result ID: 698bc2807a967eddb771260e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69ad1e209131cea234c4.md>)

[Implementation](<https://github.com/HRNet/DEKR>)

Recorded evaluation configuration: {"Method Category":"Bottom-up"}. This record does not assert independent reproduction.

### ResNet-AE

AP: 45. Result ID: 698bc2807a967eddb771260a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f511e209131ce9e8f68.md>)

Recorded evaluation configuration: {"Method Category":"Bottom-up","Backbone":"ResNet"}. This record does not assert independent reproduction.

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