How are the provided validation datasets aligned? (LFW, CFPFP, CALFW, AGEDB30, CPLFW)

Hello,

I’m using the lfw, cfp_fp, agedb_30, calfw and cplfw validation datasets, and I would like to know the exact code used for cropping and aligning the images in these datasets as provided by you. Could you please share the pipeline or specific steps you follow to prepare these datasets for model evaluation?

I have seen some of the alignment codes, but I need to replicate the exact alignment process used to create the .bin test files for validation datasets like LFW, CPLFW, AGEDB30, etc.

https://github.com/deepinsight/insightface/blob/master/python-package/insightface/utils/face_align.py
https://github.com/deepinsight/insightface/blob/master/web-demos/src_recognition/face_align.py

Thanks in advance for your help.