Champagne Jin

> Sorry that there were some problems with our website. The website has been updated, and you can download the supplementary material directly from [Here](https://github.com/daooshee/BMVC2018website/blob/master/chen_bmvc18_sup.pdf). You are too kind, thank...

As I can see, it seems that you train the DecomNet and the RelightNet separately.

Honestly, I am so curious because I want to re-implement your work in PyTorch, while the result I've got is not good enough. It looks like this:  

这些论文里都有的=、= 发自我的 iPhone > 在 2019年3月30日，下午12:16，yangdashi88 写道： > > 你好，请问可以提供estimated Boundary Heatmaps部分的网络相关的资料吗？比如训练数据、网络构造、还有在跟原图进行Fusion部分的融合方法？如果能提供，非常感谢。 > > — > You are receiving this because you are subscribed to this thread. > Reply to...

This is residual connection, and the whole network is used to estimate the noise in this setting.

However, I don't know whether you have checked the network architecture. It seems that the author did not use this residual connection in their code.

Alright, thank you, I've seen it.

Hello, thanks for your work first. I am reading DeepFace, and I am quite confused about the 3D alignment step. To be specific, I cannot tell what is the affine...

Thank you so much for your sharing. Indeed, I have 2 questions about the 3D-to-2D process. First, what are the 8 unknows in the affine camera P mean? Location, gesture...

Thanks for the work, it really inspires me! I just would like to know the data processing, could you please share it? Thanks again. @sungonce