Closed apophatique closed 2 years ago
As we mention in the readme in the folder, the code follows standard techniques and evaluations. Specifically, our code modifies the ArcFace original repo; you can use the evaluation code provided there: https://github.com/deepinsight/insightface/tree/master/evaluation/IJB
Make sure that you are using Π-net network for your loaded weights.
Could you add test.py, please? @grigorisg9gr
hello!
yes, that would be great if you could add test.py!
thank you!!!
We appreciate your interest, however we are busy this period. Feel free to make a pull request if you believe it's very different from the instructions we provide above. Thanks for understanding.
Hello @grigorisg9gr! Can you please elaborate more on how to use Π-net network so I can load the weights? How can I use the json file and parameter file provided to test it?
Hello @grigorisg9gr! Can you please elaborate more on how to use Π-net network so I can load the weights? How can I use the json file and parameter file provided to test it?
load the model first and then get the feature embedding for the face image https://github.com/grigorisg9gr/polynomial_nets/blob/master/face_recognition/eval/verification.py#L555
https://github.com/deepinsight/insightface/blob/master/recognition/_evaluation_/ijb/ijb_evals.py#L25 https://github.com/deepinsight/insightface/blob/master/recognition/_evaluation_/ijb/ijb_evals.py#L38
Hi, i'm new to machine learning
I have a faces database - it is just a directory with faces image which are named as man's name on photo. I need to recognize all people from the database in input image Can you please tell me how to run face recognition module to complete this task? I only see train.py, but no test.py
Thanks