Hi Alex! Thank you so much for this video... and the others. My question is if there are any possibility to export/save the Image or mask of the BRIAAI Matting node as Prompt output to obtain a mask, thanks in advance
Somehow I missed your comment! Yes you can. Just connect that as your output. I have a video coming in the next few weeks about matte extractions inside of nuke that is MUCH better than that though.
I am getting this error message: "Input channels 12 does not match model in_channels 8, 'opt_background' latent input should be used with the IC-Light 'fbc' model, and only with it". Any idea why?
I'd refer you to the comfyui reddit - there are some really helpful people there. Im sure if you search there you will find the answer. Alternatively watch one of the countless comfyui installation videos on youtube.
Hi Alex, thanks for this tutorial! I tried this approach without much success so I have a few questions: 1. How to make it not distort input person face? 2. I it possible to make it work with anything other then a head shot? I would suggest to change a title for this video since the approach shown seams to be specialized on portraits and not "any shot".
This is very interesting! I feel like this could also be used in a way to fix flicker much better than Davinci Resolve's deflicker? Say you have source footage but there is some jitter in your stylized footage between frames - could you train it in a similar fashion?
Potentially. Im looking into relighting moving shots with comfy (without normals) at the moment. But as always, temporal consistency is the killer. There are other ways of mitigating the chatter/noise along with copycat. If I get a decent result, ill share my findings.
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
Awesome stuff ❤
This is just great stuff, need more views!
going to follow closer!, what Nuke add on are you using the get those extra 'hotkeys'(?) frame 20:24
thanks
That’s w_hotbox. Been using it for years, cant live without it.
You can download it here: www.nukepedia.com/python/ui/w_hotbox
Hi Alex! Thank you so much for this video... and the others. My question is if there are any possibility to export/save the Image or mask of the BRIAAI Matting node as Prompt output to obtain a mask, thanks in advance
Somehow I missed your comment! Yes you can. Just connect that as your output. I have a video coming in the next few weeks about matte extractions inside of nuke that is MUCH better than that though.
I am getting this error message: "Input channels 12 does not match model in_channels 8, 'opt_background' latent input should be used with the IC-Light 'fbc' model, and only with it". Any idea why?
@@sombrero404 i would post on the project’s github issues page.
This is awesome!
Amazing stuff
Merci, super cool !
hey I am not able to see the option "manager" on my ComfyUI any idea why?
You can get it from here: github.com/ltdrdata/ComfyUI-Manager
@@alexvillabon Hey I got that now, but now it is giving me the error that I am missing a git extension
I'd refer you to the comfyui reddit - there are some really helpful people there. Im sure if you search there you will find the answer. Alternatively watch one of the countless comfyui installation videos on youtube.
Hi Alex, thanks for this tutorial! I tried this approach without much success so I have a few questions:
1. How to make it not distort input person face?
2. I it possible to make it work with anything other then a head shot?
I would suggest to change a title for this video since the approach shown seams to be specialized on portraits and not "any shot".
Brilliant!
Genius
where can I get the pmask node?
Any position mask node will do. There are a bunch on nukepedia such as pmatte
This is very interesting!
I feel like this could also be used in a way to fix flicker much better than Davinci Resolve's deflicker? Say you have source footage but there is some jitter in your stylized footage between frames - could you train it in a similar fashion?
Potentially. Im looking into relighting moving shots with comfy (without normals) at the moment. But as always, temporal consistency is the killer. There are other ways of mitigating the chatter/noise along with copycat. If I get a decent result, ill share my findings.
Did you manage to retrieve the details of the normal map?
I never could even after a lot more epoch. ( I have pushed it to 1 Millions)