Code and checkpoints release
Closed this issue · 9 comments
Hey there,
Thanks for your work, it looks amazing.
When will the code and the checkpoints be released?
Can't wait to try it out!
Yup same here!
Amazing work! Could you
Combine different images database, like using macro photos for better understanding of textures, lighting reflection
Ditto! - I'm really excited to try this out for myself!
Once the code is released, I wouldn't be surprised if Intel's value goes up...
Please release the current Code & Checkpoints!!!
As soon as it's on GitHub here, it'll be amazing to see what contributions or forks come from it....
Plllleeeassssseeeeeeee!!!!!
Amazing work!
Waiting for code!
Hey there,
Thanks for your work, it looks amazing.
When will the code and the checkpoints be released?
I cant wait to see the code! It is one of the most impressive things that I've seen in the past few years!
Hi all,
thanks for your patience! We are in the final stages of the approval process and expect to have a release before the end of the year. Due to legal issues we will not be able to release checkpoints or the full dataset. This should not prevent you from training you own version though -- our experiments showed that strong enhancements are possible with a smaller number of G-buffers, using segmentation labels, or even without the G-buffers. For experiments with G-buffers we recommend looking into capturing your own dataset. Several other research projects have demonstrated how to do this and have released code for it.
How's it going?