---
license: mit
library_name: libreyolo
tags:
  - image-to-image
  - surface-normals
  - moge
---

# LibreMoGe2s-normal

MoGe-2 ViT-S/14 surface-normal estimator, repackaged in LibreYOLO checkpoint format.

## Source

Derived from [Ruicheng/moge-2-vits-normal](https://huggingface.co/Ruicheng/moge-2-vits-normal)
at revision `679230677b4d282c6f304189a93e98e14f085902`, the exact commit LibreYOLO pins.
Copyright (c) Microsoft Corporation. Licensed under the MIT License.

The DINOv2 encoder MoGe-2 builds on is separately licensed Apache-2.0 by
Meta AI.

## Modifications

State-dict key remapping only. Learned parameters are unchanged.
See `weights/convert_moge2_weights.py` in the
[LibreYOLO source repository](https://github.com/LibreYOLO/libreyolo).

## Usage

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreMoGe2s-normal.pt")
result = model.predict("image.jpg")[0]
normals = result.normals
```

## License

MIT License. See the [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE) files in
this repository.