facenet-pytorch
facenet-pytorch copied to clipboard
timesler
Pretrained Pytorch face detection (MTCNN) and facial recognition (InceptionResnet) models
when all_boxes is as below: array([array([[389.81177, 94.35924, 480.92465, 220.15076]], dtype=float32), None], dtype=object) namely, second image has no bbox. ``` selected_boxes, selected_probs, selected_points = [], [], [] for boxes, points, probs,...
@timesler Hi, thanks for your sharing. Is there any detailed codes of your mtcnn implement? thanks a lot.
I want to retrain this model. Can you give me a complete training file
@iam-aditya which package 1.0.1 version are you taliking about ? _Originally posted by @SURABHI-GUPTA in https://github.com/timesler/facenet-pytorch/issues/129#issuecomment-707503050_
Hi, thanks for your great repo. Im using it with mtcnn backend and its accuracy is really well but Its too slow than my expectations too. I see that there...
``` Running on device: cpu /root/.cache/torch Traceback (most recent call last): File "mytest.py", line 33, in x_aligned, prob = mtcnn(x, return_prob=True) File "/usr/local/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in __call__ result = self.forward(*input,...
Thanks for sharing this code, This is not an issue anymore. I want to convert this model to ncnn to be able to run on mobile devices. Does anyone give...
Hi first of all thanks for your work! I am developing an app where I take faces and try to find the nearest face from a custom dataset. This custom...
As highlighted in #10, the current MTCNN design is appropriate for the project while a single detection method is implemented, but with multiple, it makes more sense to separate the...