---
license: apache-2.0
library_name: libreyolo
tags:
- computer-vision
- pose-estimation
- keypoint-detection
- coco
- libreyolo
- rfdetr
datasets:
- detection-datasets/coco
---

# LibreRFDETRn-pose

**EXTREMELY experimental** RF-DETR-n pose checkpoint for LibreYOLO.

This is a COCO-17 human pose preview checkpoint for LibreYOLO's `task="pose"` RF-DETR path. It is useful for testing and bootstrapping, but it is not a final benchmark release.

## Checkpoint

- File: `LibreRFDETRn-pose.pt`
- Family: `LibreRFDETR`
- Size: `n`
- Task: `pose`
- Classes: person only
- Keypoints: COCO-17, `(x, y, visibility)`
- Validation image size: `512`
- Additional training epochs for this checkpoint: `0`

## Initialization Method

Native RF-DETR-n detection checkpoint plus shared tensors from the trained LibreRFDETRs-pose checkpoint. No decoder cloning was needed because nano has fewer decoder layers than small.

This method keeps the size-specific detection backbone and resolution-dependent tensors, then transfers the pose-specialized shared tensors from the small pose checkpoint. The checkpoint should still be treated as experimental until a full per-size training run is published.

## COCO Keypoint Validation

Validation was run on COCO person keypoints val2017 through LibreYOLO's pose validator.

| Metric | Value |
| --- | ---: |
| keypoints mAP50-95 | `0.506590` |
| keypoints mAP50 | `0.806117` |
| keypoints mAP75 | `0.544956` |
| keypoints AR50-95 | `0.626338` |

The validation artifacts are included as `validation_metrics.json`. Initialization details are included as `initialization_summary.json`.

## Usage

```python
from libreyolo import LibreRFDETR

model = LibreRFDETR("LibreRFDETRn-pose.pt", task="pose")
results = model.predict("image.jpg", imgsz=512)
print(results[0].keypoints)
```

Autodownload in LibreYOLO emits an experimental warning for this checkpoint.

## Caveats

- Experimental checkpoint, not a final benchmark release.
- No additional fine-tuning epochs were run for this per-size checkpoint after transfer initialization.
- Pose export/runtime backends may have separate support status from PyTorch inference.
- Metrics are from LibreYOLO PR development artifacts, not from an independent external benchmark suite.