I initially thought it was only spatial, but in your last example, the actor with the glasses had the same ID. This suggests that over time, temporal weights might be applied, making it more consistent with Luma changes. It’s really interesting!
IMPORTAMT - As Alex cautions, this node is incredibly fragile. It can corrupt your nuke file even when the nodes is disabled. I used the node quite successfully to create a holdout. I pre-rendered my matte and disabled the cattery node. The next day the entire Nuke script was corrupt after re-opening it. All my other nodes were gone ! Only an empty input node connected to an expression node was left. I've marked this as an issue on Rafael's git repo. Cheers b
Interesting. I haven't had a full script break, that I can remember. But I do recommend keeping these tools away from main scripts. These are the type of tools that benefit from being on a separate script to make sure you don't run into any issues. Thanks for point this out for others to see. On a separate note, this specific cattery node is published by The Foundry themselves and not by Rafael Perez like others I have covered.
@@alexvillabon " this specific cattery node is published by The Foundry themselves " noted. I think I tried Rafael's version. I'll switch to foundry and see if that does better. Thanks
@@behrampatel4872 oh I see. Yeah these nodes are something to be cautious about in terms of stability. As for segment anything I personally had far better luck (and like better) Rafael’s version.
Could you , track the crypto ID so each frame it refreshes the ID asignation its stuck with the same alpha. I presume it will brek any comp. I think its a cool tool to learn from but we are not there yet.
@@juancamilo908 yeah, i guess you could. But its not terribly efficient. I have a video coming soon that does mattes and are REALLY good for much less effort than this. Im really bummed by segment anything’s performance but what I share about mattes next is an actual game changer. I just need time to record it.
I initially thought it was only spatial, but in your last example, the actor with the glasses had the same ID. This suggests that over time, temporal weights might be applied, making it more consistent with Luma changes. It’s really interesting!
I hope this gets better. However if we get a clean matte for single frames, then we could then use the output to train copycat.
Cheers
Agreed! I have something coming in the next couple of weeks that should be able to do just that :)
IMPORTAMT - As Alex cautions, this node is incredibly fragile. It can corrupt your nuke file even when the nodes is disabled.
I used the node quite successfully to create a holdout. I pre-rendered my matte and disabled the cattery node. The next day the entire Nuke script was corrupt after re-opening it.
All my other nodes were gone !
Only an empty input node connected to an expression node was left.
I've marked this as an issue on Rafael's git repo.
Cheers
b
Interesting. I haven't had a full script break, that I can remember. But I do recommend keeping these tools away from main scripts. These are the type of tools that benefit from being on a separate script to make sure you don't run into any issues. Thanks for point this out for others to see.
On a separate note, this specific cattery node is published by The Foundry themselves and not by Rafael Perez like others I have covered.
@@alexvillabon " this specific cattery node is published by The Foundry themselves " noted.
I think I tried Rafael's version. I'll switch to foundry and see if that does better.
Thanks
@@behrampatel4872 oh I see. Yeah these nodes are something to be cautious about in terms of stability.
As for segment anything I personally had far better luck (and like better) Rafael’s version.
Interesting and cool stuff but I think i'll give the boffins a bit longer to cook on this one before diving in
My feelings exactly.
Great, thanks a lot! Do you also have a video on Depth-Anything?
@@prony5145 coming soon :)
Thx for tutorial
Could you , track the crypto ID so each frame it refreshes the ID asignation its stuck with the same alpha. I presume it will brek any comp. I think its a cool tool to learn from but we are not there yet.
@@juancamilo908 yeah, i guess you could. But its not terribly efficient. I have a video coming soon that does mattes and are REALLY good for much less effort than this. Im really bummed by segment anything’s performance but what I share about mattes next is an actual game changer. I just need time to record it.