ArcFace

Implementation of ArcFace¹ in PyTorch by InsightFace.

Pre-trained models

mozuma.models.arcface.pretrained.torch_arcface_insightface

ArcFace model pre-trained by InsightFace

Parameters:

Name	Type	Description	Default
device	torch.device	Torch device to initialise the model weights	required
remove_bad_faces	bool	Whether to remove the faces with bad quality from the output. This will replace features of bad faces with float("nan"). Defaults to False.	required
bad_faces_threshold	float	The cosine similarity distance to reference faces for which we consider the face is of bad quality.	required

Returns:

Type	Description
TorchArcFaceModule	Pre-trained ArcFace

Base models

The MagFace model is an implementation of a TorchModel.

mozuma.models.arcface.modules.TorchArcFaceModule

Creates face embeddings from MTCNN output

Attributes:

Name	Type	Description
device	torch.device	Torch device to initialise the model weights
remove_bad_faces	bool	Whether to remove the faces with bad quality from the output. This will replace features of bad faces with float("nan"). Defaults to False.
bad_faces_threshold	float	The cosine similarity distance to reference faces for which we consider the face is of bad quality.

Provider store

See the stores documentation for usage.

mozuma.models.arcface.stores.ArcFaceStore

Gets the pretrained state dir from OneDrive

URL: https://github.com/TreB1eN/InsightFace_Pytorch#2-pretrained-models--performance Model: IR-SE50

---

1. Jiankang Deng, Jia Guo, Xue Niannan, and Stefanos Zafeiriou. Arcface: additive angular margin loss for deep face recognition. In CVPR. 2019. ↩