Kuan-Yu Huang (Peter) comments

Results 9 comments of


                                            Kuan-Yu Huang (Peter)

你好，当运行test.py时出现问题，不知道如何解决。

Hi @TeenageLi , 不曉得您有沒有查看過repo內提供的Colab notebook， [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/peteryuX/arcface-tf2/blob/master/notebooks/colab-github-demo.ipynb) 如果可以跑，您可以觀察一下您的環境建設是否都相同。

hello！我想请教一下这个项目中使用的resnet50是tf官方的网络吧，原版insightface论文中使用的是修改后的resnet50，有对比过哪个效果更好吗？

Hi @FelixZhang7 ! Can you point out what the difference between them? I think maybe I ignored it.

hello！我想请教一下这个项目中使用的resnet50是tf官方的网络吧，原版insightface论文中使用的是修改后的resnet50，有对比过哪个效果更好吗？

Hi @FelixZhang7 , 感謝您提點我這些資訊，聽起來這些改動應該是會帶來性能上的提升。不過，我好像沒有在[paper](https://arxiv.org/abs/1801.07698)中看到相關的文字說明，請問我是不是遺漏了甚麼?

hello！我想请教一下这个项目中使用的resnet50是tf官方的网络吧，原版insightface论文中使用的是修改后的resnet50，有对比过哪个效果更好吗？

@FelixZhang7 ，還是謝謝您的提點，關於LResnet50E-IR的部分，我會嘗試實現並完善於這個項目中~ =)

image with lower psnr , appearing visually good

Hey, @nan-rock. Higher PSNR dose not mean better visual quality but fidelity.

image with lower psnr , appearing visually good

Try np.clip() value to 0 - 255 before astype to uint8

Can you add inference script?

Hey, @ThiagoMateo. The script can be modified for batching inference in simple way, as you use for-loop to process the NMS and decoding function. The network part can support batch...

Can you add inference script?

Did you try `tf.shape(x)[0]` to get the inference batch size?

Can you add inference script?

Hey Guys, I think I share not suitable information for you. The simpler way to figure this problem out is moving the decode nms out of keras model, which like...