Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
license_name: openpose-non-commercial
|
| 4 |
+
license_link: https://github.com/CMU-Perceptual-Computing-Lab/openpose/blob/master/LICENSE
|
| 5 |
+
tags:
|
| 6 |
+
- pose-estimation
|
| 7 |
+
- openpose
|
| 8 |
+
- body-25
|
| 9 |
+
- pytorch
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# OpenPose-135 weights (BODY_25 + hand + face)
|
| 13 |
+
|
| 14 |
+
Self-contained PyTorch checkpoints for the CMU OpenPose **BODY_25 body** (25 keypoints),
|
| 15 |
+
**hand** (21 keypoints × 2), and **face** (70 keypoints) detectors — the exact stack
|
| 16 |
+
needed to reproduce the OpenPose-135 keypoint layout without a Caffe runtime or `mmcv`.
|
| 17 |
+
|
| 18 |
+
## Files
|
| 19 |
+
|
| 20 |
+
| File | Size | Source |
|
| 21 |
+
|---|---:|---|
|
| 22 |
+
| `body_pose_model_25.pth` | ~200 MB | CMU `pose_iter_584000.caffemodel` ported via [caffemodel2pytorch](https://github.com/vadimkantorov/caffemodel2pytorch), redistributed by [TracelessLe/OpenPose.PyTorch](https://github.com/TracelessLe/OpenPose.PyTorch) |
|
| 23 |
+
| `hand_pose_model.pth` | 147 MB | CMU hand pose model, mirrored from [lllyasviel/Annotators](https://huggingface.co/lllyasviel/Annotators) |
|
| 24 |
+
| `facenet.pth` | 154 MB | CMU OpenPose face / FaceNet, mirrored from [lllyasviel/Annotators](https://huggingface.co/lllyasviel/Annotators) |
|
| 25 |
+
|
| 26 |
+
## Usage
|
| 27 |
+
|
| 28 |
+
These are intended to be loaded by [`tools/openpose135/`](https://github.com/hohs) — the
|
| 29 |
+
self-contained PyTorch OpenPose-135 detector vendored into the PHD codebase. The detector
|
| 30 |
+
auto-downloads from this repo on first use:
|
| 31 |
+
|
| 32 |
+
```python
|
| 33 |
+
from tools.openpose135 import OpenPose135Detector
|
| 34 |
+
detector = OpenPose135Detector(device="cuda") # auto-fetches the three .pth files
|
| 35 |
+
people = detector(image_rgb)
|
| 36 |
+
```
|
| 37 |
+
|
| 38 |
+
To use these weights without that wrapper, see the architecture definitions in
|
| 39 |
+
`tools/openpose135/model.py` — the state dicts are flat (caffemodel2pytorch / direct
|
| 40 |
+
naming) and load through a small `transfer()` helper.
|
| 41 |
+
|
| 42 |
+
## License
|
| 43 |
+
|
| 44 |
+
These weights are derived from the **CMU OpenPose** project, which is licensed for
|
| 45 |
+
**non-commercial use only**:
|
| 46 |
+
|
| 47 |
+
> The OpenPose project is freely available for free non-commercial use, and may be
|
| 48 |
+
> redistributed under these conditions. Please, see the [LICENSE](https://github.com/CMU-Perceptual-Computing-Lab/openpose/blob/master/LICENSE)
|
| 49 |
+
> for further details. Interested in a commercial license? Contact the CMU Technology
|
| 50 |
+
> Transfer and Enterprise Creation office.
|
| 51 |
+
|
| 52 |
+
Redistribution here is non-commercial; downstream use inherits the same restriction.
|
| 53 |
+
|
| 54 |
+
## Attribution
|
| 55 |
+
|
| 56 |
+
- CMU OpenPose: <https://github.com/CMU-Perceptual-Computing-Lab/openpose>
|
| 57 |
+
- BODY_25 PyTorch port: <https://github.com/TracelessLe/OpenPose.PyTorch>
|
| 58 |
+
- Hand + face PyTorch port: <https://github.com/lllyasviel/ControlNet-v1-1-nightly>
|
| 59 |
+
- Original pytorch-openpose: <https://github.com/Hzzone/pytorch-openpose>
|