FaceX-Zoo

FaceX-Zoo copied to clipboard

训练模型时，第一步“Align the face images to 112*112 according to face_align.py.”是把图片resize成112*112大小，还是需要用到face_align.py脚本进行处理

4

Step1: Prepare the training data Align the face images to 112*112 according to face_align.py. 需要用face_align.py这个文件处理图片还是简简单单把图片resize成112*112呢

tikdik

fumingren

Mask Face Detection

1

I want to differentiate between a face wearing a mask and a face that is not. What should I do?

cho-jae-seong

Masked Face Recognition with MobileFacenet backbone

5

Thank for your work

huytuong010101

loss fluctuation during training

1

Hi, thanks for your great work! When I'm using this repo to train a customized face dataset. I find that some unusual loss fluctuation happened. The loss will decrease at...

wynterr

KeyError: 'formatters'

4

 Could you tell me why?

ZhangChaoZhong

Is there any plan to support 224x224 input models?

2

dangne

请教使用自定义人脸数据集继续训练问题

3

head_conf.yaml中的num_class目前都是72778, 想要在预训练模型的基础上使用自定义人脸数据集继续训练，需要将head_conf.yaml文件中的num_class改成自定义人脸数据集中的数量吗？

GitHubLuGeng

@wang21jun 您好，谢谢您的工程，很有用^^ 1、口罩戴到鼻子左右的地方，口罩模拟，您这边有什么建议吗？ 是制作 mask_data 数据鼻子下的口罩这样吗？

sky186

I kept running into this error when trying to use RexNet. ``` Traceback (most recent call last): File "main.py", line 91, in blackbox_success_rates = evaluate(clipped_advs, targets, blackbox_models, model_loader, epsilons, args.device)...

dangne