tryink

> I was wondering the same... @iluvrachel did the normalization step work? I still didn't proceed to "step 5" because I wonder I will train on un-normalized labels :( I...

Hi, I also want to add the time smoothing module into this implementation, but I just have no idea what to do with the training period. Could you please give...

> Hi, I'd like to ask what the latest and best technique of pose synthesis is now? Is everybody dance now paper still the best or are there better ones?...

Have you solved the question by far? I wonder the same.

> Either you take the output mesh in FLAME topology and extract the 3D landmarks from this mesh using the provided barycentric landmark embedding (you could do that by calling...

> I have better results than this, princessmittens, can you leave your e-mail and I will reach out to you? > > > No-I have talked to @hanikh. I don't...

> I am getting a similar error is Node Matmul. Any idea with that? hi,have you solved the problem? I encountered the same and have no ideas

> After I run `lfw_eval.py`, there occurred this error: > > ``` > Traceback (most recent call last): > File "lfw_eval.py", line 122, in > _, result = eval(net.sphere().to('cuda'), model_path='checkpoint/CosFace_24_checkpoint.pth')...

> 您好！ 感谢这篇高质量的工作以及靠谱的开源！ > > 我在提供的CelebA-HQ上获取了很好的复现结果。但是，在其他图片上进行测试时，lmk和segment map我可以正常提取并可视化得到不错的结果，对于code（也就是W+ latent space embedding），因为**pSp提供了多个checkpoint**，似乎没有指明该使用哪个？请问能否提供相关信息。我目前获取encoder的输出，即code，测试了 with&without average latent code的结果，看来都是错误的。 > > 提供的数据集的结果：  > > 我自己提取code/landmark/segment map的结果：  > > 感谢！ 您好，请问您用的是哪个版本的pSp_checkpoint用于测试呢？

> Thanks for your interest, we use the StyleGAN inversion checkpoint (https://drive.google.com/file/d/1bMTNWkh5LArlaWSc_wa8VKyq2V42T2z0/view). Note that you need to align the faces before inversion. Hi, tanks for your reply! By the way,...