Face recognition model in face_sdk

[buiduchanh]

Hi. @wang21jun . Thank for your contribute.
I have a question about the model using for recognition, I saw many pretrain model in https://github.com/JDAI-CV/FaceX-Zoo/tree/main/training_mode but I dont know what model you used ? . In config file I saw this model name is face_recognition_1.0 but I dont know the architecture of this model ( backbone, head v..v). Can you share me the details of this model ?
Thanks.

[wang21jun]

backbone: MobileFaceNet
head: MV-Softmax
accuracy on megaface: 90.39
you can get the model by this tool: https://github.com/JDAI-CV/FaceX-Zoo/blob/main/face_sdk/utils/model_convert.py

[buiduchanh]

backbone: MobileFaceNet
head: MV-Softmax
accuracy on megaface: 90.39
you can get the model by this tool: https://github.com/JDAI-CV/FaceX-Zoo/blob/main/face_sdk/utils/model_convert.py

Thanks for your quick reply @wang21jun
By the way, I saw some sota models (backbone or head) and your explanation in this link #11 (comment). I wanna train a full model with strongest backbone and head ( Attention + MV-Softmax). I saw you used MS-Celeb and remove some identities so training set contain 72,778 identities and about 3.28M images. Can you share your preprocessed dataset ?
Thanks you

[wang21jun]

https://github.com/JDAI-CV/FaceX-Zoo/blob/main/data/files/MS-Celeb-1M-v1c-r_id_list.txt
Just use all of the images of these identities.

[ooodragon94]

hi
thanks for your great repo!

I have trouble finding "https://github.com/JDAI-CV/FaceX-Zoo/blob/main/face_sdk/utils/model_convert.py"
can you help me out?

huge thanks!