Metadata-Version: 2.1
Name: synthface
Version: 0.0.1
Summary: Face recognition model trained on facesynth
License: MIT
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: facesynth>=0.1.0
Requires-Dist: numpy>=1.24
Requires-Dist: pillow>=10.0
Provides-Extra: dev
Requires-Dist: pytest>=7.0; extra == "dev"

# synthface

Face recognition framework trained on synthetic face data from `facesynth`.

> **Status:** Not functional yet, under development.

## Installation

```bash
pip install synthface
```

Requires `facesynth>=0.1.0`.

## Quickstart

```python
from PIL import Image
from synthface import FaceRecognitionModel

model = FaceRecognitionModel()

img_a = Image.open("face_a.png")
img_b = Image.open("face_b.png")

# 1:1 Verification
result = model.verify(img_a, img_b, threshold=0.5)
print(result)

# Extract 512D embedding
embedding = model.extract(img_a)
print(embedding.shape)
```

## License

MIT License. See [LICENSE](LICENSE).
