Request: Can someone compile a RemoveDirt x64

[poisondeathray]

Request:

Can someone compile a 64bit RemoveDirt.dll ? The kassandro version with RestoreMotionBlocks

Background Info:
The x64 version at the bottom of this page does not work (and it says that)
avisynth.nl/index.php/RemoveDirt

wayback machine for removedirt kassandro page
http://web.archive.org/web/20140723074402/http://home.arcor.de/kassandro/RemoveDirt/RemoveDirt.htm


Thanks

[johnmeyer]

What advantage do you think you'll get with a 64-bit version? Can you not get the existing 32-bit version(s) to work at all in a Win 64-bit environment?

64-bit is not automatically "better" than 32-bit for many programs, applications, DLLs, etc.

[poisondeathray]

What advantage do you think you'll get with a 64-bit version? Can you not get the existing 32-bit version(s) to work at all in a Win 64-bit environment?

64-bit is not automatically "better" than 32-bit for many programs, applications, DLLs, etc.

Yes, it should work with mp_pipeline , but I haven't tested it yet

I expect it to be about 5-10% faster right off the bat for x64, and better for memory when using larger scripts

When you have 2K and 4K filmscans it's not fun working in x86

[poisondeathray]

The funny thing is there is a vapoursynth RemoveDirtVS x64 version based on the original kassandro plugin.... I thought surely someone has made one for avisynth already ?? My google kung fu skills are lacking and I couldn't find it

[MysteryX]

Plus it has the advantage of then integrating into an otherwise all x64 filter chain

[real.finder]

Plus it has the advantage of then integrating into an otherwise all x64 filter chain

yes and can use more than 2-4 G of ram, no headache (for many peoples) or/and overhead with mpp

[pinterf]

More than a year ago I gave it a try but it was too much work (goal: 2.6 header, remove mmx and inline asm, add new colorspaces). I think the backport from Vs is much much less pain. If there is no volunteer, I could look at it again.

[real.finder]

More than a year ago I gave it a try but it was too much work (goal: 2.6 header, remove mmx and inline asm, add new colorspaces). I think the backport from Vs is much much less pain. If there is no volunteer, I could look at it again.

this is vs source https://github.com/handaimaoh/removedirtvs

[pinterf]

Thanks, I know it, SCSelect, DupBlocks and RestoreMotionBlocks are ported. Those would be enough for me.
I was hoping to port all the other stuffs (without knowing if they are used by anyone) and there was a 0.9 and a half-ready 1.0 I think, then I stopped working on it.

[StainlessS]

Kassandro forum, no longer online:- http://videoprocessing.11.forumer.com

PAGE NOT FOUND

I did not look on waybackmachine for it.

[silverwing]

Thanks, I know it, SCSelect, DupBlocks and RestoreMotionBlocks are ported.

Hi!

Can anybody tell me where SCSelect is located in modern Avisynth (i.e. version 2.6+)?
Thank you for attention. :thanks:

[StainlessS]

SCSelect was from a Kassandro plugin.

Here an equivalent function in script:- https://forum.doom9.org/showthread.php?p=1633738#post1633738


Avisource("D:\avs\test.avi")

# RemoveDirt's SCSelect(clip input, clip scene_begin, clip scene_end, clip global_motion, float dfactor, bool debug, bool planar)

Function SCSelect_Like(clip dclip,clip Start,clip End,clip Motion, float "dfactor",bool "debug") {
# Start, End, Motion MUST all be same, dclip can be other colorspace/size (unlike SCSelect).
dfactor=Float(Default(dfactor,4.0))
debug=Default(debug,false)
Global SCM_A=0.0 Global SCM_B=0.0 Global SCM_SC=0 Global SCM_Prev=-1
Motion.ScriptClip("""
NotNext = (current_frame!=SCM_Prev+1)
Global SCM_A=(NotNext)? RT_LumaDifference(dclip,dclip,n=current_frame-1,n2=current_frame) : SCM_B
Global SCM_B= RT_LumaDifference(dclip,dclip,n=current_frame,n2=current_frame+1)
# 0 = Start of scene, 1 = End of scene, 2 = Global motion
Global SCM_SC=(current_frame==FrameCount-1)?1:(SCM_A>dfactor*SCM_B || current_frame==0)?0:(SCM_B>dfactor*SCM_A)?1:2
(SCM_SC==0) ? Start : (SCM_SC==1) ? End : Last # Choose Start, End or Motion(ie Last)
(debug)?RT_Subtitle("%d ] %6.2f %6.2f SC=%d",current_frame,SCM_A,SCM_B,SCM_SC):NOP
Global SCM_Prev=current_frame
Return Last
""",args="dfactor,Start,End,Dclip,debug") # Needs Grunt for args
return Last
}

Start=Subtitle("START OF SCENE",align=3,size=30)
End=Subtitle("END OF SCENE", align=1,size=30)
Motion=Subtitle("GLOBAL MOTION",align=5,size=30)
Return SCSelect(Last,Start,End,Motion)

EDITED ABOVE client script: Should not call two instances on function which uses global Vars.

EDIT:

Function EndOfSceneClip(clip c,Int "thSCD1",Int "thSCD2") { # All Luma Samples set 255 at EOS, else 0
thSCD1=Default(thSCD1,400) thSCD2=Default(thSCD2,130)
sup=c.MSuper(pel=1,sharp=0,rfilter=2,hpad=16, vpad=16)
bv=sup.MAnalyse(isb=true, delta=1,blksize=16)
Return c.MSCDetection(bv,thSCD1=thSCD1,thSCD2=thSCD2)
}

Function StartOfSceneClip(clip c,Int "thSCD1",Int "thSCD2") { # All Luma Samples set 255 at SOS, else 0
thSCD1=Default(thSCD1,400) thSCD2=Default(thSCD2,130)
sup=c.MSuper(pel=1,sharp=0,rfilter=2,hpad=16, vpad=16)
fv=sup.MAnalyse(isb=false,delta=1,blksize=16)
Return c.MSCDetection(fv,thSCD1=thSCD1,thSCD2=thSCD2)
}

Function SceneCutClip(clip c,Int "thSCD1",Int "thSCD2") { # All Luma pixel = 0 =Norm, 1=EOS, 2=SOS, 3=EOS & SOS
thSCD1=Default(thSCD1,400) thSCD2=Default(thSCD2,130)
Sup = c.MSuper(pel=1,sharp=0,rfilter=2,hpad=16, vpad=16)
BvEos= Sup.MAnalyse(isb=True, delta=1,blksize=16)
FvSos= Sup.MAnalyse(isb=False, delta=1,blksize=16)
Eos = c.MSCDetection(BvEos,thSCD1=thSCD1,thSCD2=thSCD2)
Sos = c.MSCDetection(FvSos,thSCD1=thSCD1,thSCD2=thSCD2)
Return MT_Lutxy(Eos,Sos,yexpr="y 0 == x 0 == 0 1 ? x 0 == 2 3 ? ?",u=-128,v=-128)
}

[silverwing]

SCSelect was from a Kassandro plugin.

Yes I know. I'm already tired of trying to compile the SCSelect function (from RemoveGrainT source) as a separate plug-in, so that it works without a segmentation fault (in AVS+ MT). :(


Here an equivalent function in script:- https://forum.doom9.org/showthread.php?p=1633738#post1633738

Good idea! Thank you! I will try. :thanks:

[StainlessS]

In the SCSelect_Like thing,


# Add 1st line inside ScriptClip
n=current_frame

And change all other current_frame instances to n. Will maybe be a bit faster (n would be Local var, current_frame is Global, and a LOT SLOWER to access than Locals).
[at least in AVS standard, not sure in AVS+, think AVS+ added a hash table for local/global var access, so may make little difference in AVS+]

EDIT: Like so


# RemoveDirt's SCSelect(clip input, clip scene_begin, clip scene_end, clip global_motion, float dfactor, bool debug, bool planar)

Function SCSelect_Like(clip dclip,clip Start,clip End,clip Motion, float "dfactor",bool "debug") {
# Start, End, Motion MUST all be same, dclip can be other colorspace/size (unlike SCSelect).
dfactor=Float(Default(dfactor,4.0))
debug=Default(debug,false)
Global SCM_A=0.0 Global SCM_B=0.0 Global SCM_SC=0 Global SCM_Prev=-1
Motion.ScriptClip("""
n = current_frame
NotNext = (n!=SCM_Prev+1)
Global SCM_A=(NotNext)? RT_LumaDifference(dclip,dclip,n=n-1,n2=n) : SCM_B
Global SCM_B= RT_LumaDifference(dclip,dclip,n=n,n2=n+1)
# 0 = Start of scene, 1 = End of scene, 2 = Global motion
Global SCM_SC=(n==FrameCount-1)?1:(SCM_A>dfactor*SCM_B || n==0)?0:(SCM_B>dfactor*SCM_A)?1:2
(SCM_SC==0) ? Start : (SCM_SC==1) ? End : Last # Choose Start, End or Motion(ie Last)
(debug)?RT_Subtitle("%d ] %6.2f %6.2f SC=%d",n,SCM_A,SCM_B,SCM_SC):NOP
Global SCM_Prev=n
Return Last
""",args="dfactor,Start,End,Dclip,debug") # Needs Grunt for args
return Last
}

[pinterf]

There were always more important filters during the last year, maybe I can resurrect the work again this spring.

[silverwing]

maybe I can resurrect the work again this spring.
YES! Dear pinterf, You can make us happy. :goodpost:
RestoreMotionBlocks () and SCSelect () are very necessary functions.
Thank you!

[silverwing]

Thank you pinterf! Now I run the function RemoveDirt (clip input, bool "_grey", int "repmode") in 64-bit avisynth. No segmentation fault. It's work normally.
This is a great job (SCSelect, RestoreMotionBlocks) (https://github.com/pinterf/RemoveDirt), many thanks! :thanks:

[pinterf]

My pleasure. And as you'd notice from the git readme, many thing were done since this release, next version is coming soon.