DCT-NET.Pytorch icon indicating copy to clipboard operation
DCT-NET.Pytorch copied to clipboard

unofficial implementation of DCT-Net: Domain-Calibrated Translation for Portrait

Results 10 DCT-NET.Pytorch issues
Sort by recently updated
recently updated
newest added

十分感谢你的贡献,但我有几个问题想要咨询一下。 我发现你在训练ccn的时候没有做论文中提到的blend操作,Geometry expansion module我也没有找到,这是有什么原因吗? 另外论文作者提到他用相同的模型对头部和背景进行了两次推理,只是对头部模型进行了额外的优化,但在你复现中似乎只有一个模型并且只做了一次推理,是有什么改动吗? 另外show中展示的效果似乎不是很好,有更好的效果图可以分享一下吗?谢谢

Hi @LeslieZhoa ! How can you compare the dictionary and the float? https://github.com/LeslieZhoa/DCT-NET.Pytorch/blob/b303f416c2caf05114fd38f10b0570df903b4462/trainer/ModelTrainer.py#L124 This method is called here: https://github.com/LeslieZhoa/DCT-NET.Pytorch/blob/b303f416c2caf05114fd38f10b0570df903b4462/trainer/ModelTrainer.py#L96 But the val_loss obtained in CCNTrainer.evalution is an empty dictionary: https://github.com/LeslieZhoa/DCT-NET.Pytorch/blob/b303f416c2caf05114fd38f10b0570df903b4462/trainer/CCNTrainer.py#L196-L206...

thanks. a nice work!

What is the image size used for training CCN, Why can't the matrix be multiplied when I input 1024 in training, and The problem of ' 'Nan or Inf found...

生成人脸都比较好,就是处理整图背景表现出有一块一块的黑色的

![2023-03-27 09-50-56 的屏幕截图](https://user-images.githubusercontent.com/29244715/227822148-e056f78e-966d-4eb2-a024-2e1fdac3e216.png) Are the following training results correct? Can the CCN stage be directly adopted https://github.com/rosinality/stylegan2-pytorch? to train, or what changes did you make in the CCN stage In...

Trained for 9 epochs, 4000 times, ![2023-03-27 19-39-01 的屏幕截图](https://user-images.githubusercontent.com/29244715/227931058-84841a92-0e79-4855-8a69-3b100c902798.png) ![2023-03-27 19-41-40 的屏幕截图](https://user-images.githubusercontent.com/29244715/227931611-6ff4c60d-77ab-426d-9f98-88dc71f2e164.png)

![2](https://user-images.githubusercontent.com/29244715/224950519-c12325a4-2c18-40f6-9079-33b10b27a8f7.jpg) ![2sN6NWwfNF_small](https://user-images.githubusercontent.com/29244715/224950576-752e2fd2-a22d-44e3-a476-ebfa6a1df60c.jpg) ![beyonce](https://user-images.githubusercontent.com/29244715/224950701-6dcb02bd-d077-403d-baf8-8043b0037c1c.jpeg) What's the matter with these black heads?

非常感谢作者伟大的工作,我尝试用ttn代码训练(ccn,face expression estimator等都没发现啥问题),大概训练了50epoch,但是效果不太理想。是训练时间过短还是有其他原因。