20.03.2024, 18:50
I don't think that this is the problem
I attached a sample. So that you can replicate the results.
I simplified the script, so that the average is simply calculated as
you can move the weight with value = 90, to see how each frame impact on the average.
No all frames have the same problem. The most interesting is the frame n.439 try to use it as reference.
good luck!
Dan
In the script the default is dd_method=2, so in the first case you are using both the filters, in the second you are using only deoldify (not resized, because also chroma_resize=True by default).
I attached a sample. So that you can replicate the results.
I simplified the script, so that the average is simply calculated as
clip = core.misc.AverageFrames(clip, [0, 0, 0, 0, 0, 0, 0, 0, 0, 90, 10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], scale=100, scenechange = False, planes=[1,2])
you can move the weight with value = 90, to see how each frame impact on the average.
No all frames have the same problem. The most interesting is the frame n.439 try to use it as reference.
good luck!
Dan
(20.03.2024, 18:50)Selur Wrote: Got a question:
should only call deoldify, right?clip = ddeoldify(clip=clip, model=0, render_factor=31, sat=[1.00,1], hue=[0.00,0], chroma_resize=False, color_stabilizer=[False,False,False,11,"center",True])
Strange thing is, when I change:
toif dd_method == 0 or (dd_method != 1):
clipa = clip.std.ModifyFrame(clip, ddeoldify_colorize)
speed increases form 7.77fps to 9.24fps.if dd_method == 0 or (dd_method != 1):
return clip.std.ModifyFrame(clip, ddeoldify_colorize)
It seems like I overlooked something and not only simple deoldify is applied.
Any idea what is happening?
Cu Selur
In the script the default is dd_method=2, so in the first case you are using both the filters, in the second you are using only deoldify (not resized, because also chroma_resize=True by default).