AviSynth Video problem? please...

[NLP]

Hi folks,

I have a script that runs perfectly on Windows 7 64 bits.

But when I run it on Windows 10 it creates a kind of "ghost effect/problem". It normally happens at the beginning of the video.


(There is an example video below with before/after results and all Windows 7 and 10 features I quoted below)


When I run it on Windows 7 64bits it does not happen.

I'm using AviSynth 2.6

And I'm using these plugins:

SetMemoryMax(512)
SetMTMode(3)
LoadPlugin("m:\temp2\RemoveGrain.dll")
LoadPlugin("m:\temp2\MaskTools2-25.dll")
LoadPlugin("m:\temp2\Depan.dll")
LoadPlugin("m:\temp2\DepanEstimate.dll")
LoadPlugin("m:\temp2\MVTools2.dll")

Example:

- Raw Video: https://drive.google.com/file/d/1PTiXiqNmK-GzzdQipJt3GDCCVY89viES/view

- Video with AviSynth Scripts Windows 10 with this ghost image problem: https://drive.google.com/file/d/1s38WbZj89OiKJd0_2HfDq51ULseThTjr/view



Does someone knows why it is happening?

Maybe there is/are some Plugin that has been updated to Windows 10?



Windows 7 Features

Microsoft Windows 7 Ultimate Operating System Name
Version 6.1.7601 Service Pack 1 Build 7601
Other Operating System Information Not available
Operating System Manufacturer Microsoft Corporation
System name CASA-PC
System manufacturer To be filled by O.E.M.
System model To be filled by O.E.M.
System type x64-based PC
AMD FX (tm) -8350 Processor Eight-Core Processor, 4000 Mhz, 4 Core (s), 8 Logical Processor (s)
BIOS version / date American Megatrends Inc. 2603, 06/26/2015
SMBIOS version 2.7
Windows folder C: \ Windows
System folder C: \ Windows \ system32
Boot device \ Device \ HarddiskVolume4
Brazil
Hardware Abstraction Layer Version = "6.1.7601.24520"
Username CASA-PC \ Tony
Time zone Brazil official time
Installed Physical Memory (RAM) 16 GB
Total physical memory 11.9 GB
8.52 GB available physical memory
Total virtual memory 23.8 GB
Available virtual memory 19.3 GB
Paging file space 11.9 GB
Paging file C: \ pagefile.sys

Windows 10 Features

Microsoft Windows 10 Pro Operating System Name
Version 10.0.19042 Build 19042
Other Operating System Description Not available
Operating System Manufacturer Microsoft Corporation
System name TONY-PC
System manufacturer System manufacturer
System Product Name System model
X64-based PC system type
SKU system SKU
AMD Ryzen 7 Processor 2700 Eight-Core Processor, 3200 Mhz, 8 Core (s), 16 Logical Processor (s)
BIOS version / date American Megatrends Inc. 2407, 07/01/2020
SMBIOS version 3.2
Embedded Controller Version 255.255
Legacy BIOS mode
BaseBoard manufacturer ASUSTeK COMPUTER INC.
BaseBoard Product TUF GAMING X570-PLUS_BR
BaseBoard Rev X.0x Version
Desktop Platform Role
Secure Boot State Not supported
PCR7 Configuration Connection Impossible
Windows folder C: \ WINDOWS
System folder C: \ WINDOWS \ system32
Boot device \ Device \ HarddiskVolume1
Brazil
Hardware Abstraction Layer Version = "10.0.19041.844"
Username Tony-PC \ Tony
Time zone Brazil official time
Installed Physical Memory (RAM) 16 GB
Total physical memory 15.9 GB
Available physical memory 11.1 GB
Total virtual memory 31.9 GB
26.5 GB available virtual memory
Paging file space 16.0 GB
Paging file C: \ pagefile.sys
Kernel DMA Protection Disabled
Virtualization-based security Not enabled
Device Encryption Support Reasons for Automatic Device Encryption Failure: Unusable TPM, PCR7 Association Not Compatible, Hardware Security Test Interface Failed, and the device is not in Modern Standby, Bus / DMA compliant devices not allowed detected, unusable TPM
Hyper-V - VM Monitor Mode Extensions Yes
Hyper-V - Second Level Address Translation Extensions Yes
Hyper-V - Virtualization Enabled on Firmware No
Hyper-V - Data Execution Protection Yes

Thank you so much in advance!

[Selur]

Do you use the same filter .dlls on Win7 64bit and Win10 64bit?
Since it seems like you use really old .dlls (MaskTools2-25) there might be a fix in newer versions of the filters of some asm code which is only used by the newer cpu.
Others probably could test if their Windows 10 showed the same issue if you would share your script,..

[jagabo]

Provide your entire script. Upload your original video, not a deinterlaced, reencoded, AVC video. Your "pos" sample hasn't been deinterlaced, has many frames out of order, and is RGB rather than YUV. Obviously something is going wrong but without seeing your script and the original source nobody can offer a fix.

[NLP]

Do you use the same filter .dlls on Win7 64bit and Win10 64bit?
Since it seems like you use really old .dlls (MaskTools2-25) there might be a fix in newer versions of the filters of some asm code which is only used by the newer cpu.
Others probably could test if their Windows 10 showed the same issue if you would share your script,..

Hi Selur,


Yes, I use the same .dlls on both windows versions.

Here is the Script in a PDF document (seems better to read than in the forum code area): https://drive.google.com/file/d/1SLqIISkkK-3N2RLcgsIOQGo5FyGGyqzZ/view?usp=sharing


Thanks for your kind help!

[NLP]

Provide your entire script. Upload your original video, not a deinterlaced, reencoded, AVC video. Your "pos" sample hasn't been deinterlaced, has many frames out of order, and is RGB rather than YUV. Obviously something is going wrong but without seeing your script and the original source nobody can offer a fix.

Hi jagabo,

I do the capture on HDMI AVC because it takes out several horizontal lines.

Entering the Panasonic DMR-ES10 and passing through the VCR JVC HR-S5912U. So, going for Avermedia C027 PcIe capture card.

That's because in DV AVI, there is much more noise.

Here is the Script in a PDF document: https://drive.google.com/file/d/1SLqIISkkK-3N2RLcgsIOQGo5FyGGyqzZ/view?usp=sharing


Script

#Import("M:\temp2\source script2 Dr Paulo Pirej.avs")
# Plugins: - RemoveGrain
# - MaskTools2
# - Depan + DepanEstimate
# - MVTools2

SetMemoryMax(512)
SetMTMode(3)
LoadPlugin("M:\Temp2\RemoveGrain.dll")
LoadPlugin("M:\Temp2\MaskTools2-25.dll")
LoadPlugin("M:\Temp2\Depan.dll")
LoadPlugin("M:\Temp2\DepanEstimate.dll")
LoadPlugin("M:\Temp2\MVTools2.dll")

#mpeg2source("M:\Marcelo 27.5.20\1\01\01-1.d2v")
#mpeg2source("M:\Marcelo 999255703 - Fazer\01 ok 19m.d2v")
DirectShowSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4", fps = 25, ConvertFPS = True )
#DirectShowSource("M:\Luis Tonini 14.1.20 - Fazer\01 V14ES10 e.mp4")
#DirectShowSource("E:\Tereza Soares 13.8.20 - Fazer\01 TS 8capt V17JVCES10.mp4", fps = 25, ConvertFPS = True ) # para AVI encapsulado AVI Xvid da Avermidia ou MP4 da Aple
#DirectShowSource("M:\Neogrid 5-10-17\090 4capt V13SVSDV.avi", fps = 25, ConvertFPS = True ) # para AVI encapsulado AVI Xvid da Avermidia
#AVISource("E:\Cleiton\Super 8mm\03\03 ok.avi") 
#AVISource("E:\Luiz Itapema 23.9.20 - Fazer\18 LI SD8.avi", audio=true)
#AVISource("E:\Luiz Itapema 23.9.20 - Fazer\18 LI SD8.avi") # DV AVI
#AVISource("M:\Pedro 31-7-18\25 8capt V15ES10SDV.avi", audio=true)


SetMTMode(2)
#ConvertToYV12()
#ConvertToyv411()
#Crop(0,20,-0,-30,true)
#bob(0,0.5)

AssumeTFF()
SeparateFields()

a=last

clense(reduceflicker=false).merge(last,0.5).clense(reduceflicker=false)
mot=removegrain(11,0).removegrain(20,0).DepanEstimate(range=2)
take2=a.depaninterleave(mot,prev=2,next=2,subpixel=2)
clean1=take2.TMedian2().selectevery(5,2)

sup1 = clean1.minblur(1).removegrain(11,0).removegrain(11,0)
\ .mt_lutxy(clean1,"x 1 + y < x 2 + x 1 - y > x 2 - y ? ?",U=2,V=2)
\ .msuper(pel=2,sharp=0)
sup2 = a.msuper(pel=2,levels=1,sharp=2)
#bv23=sup1.manalyse(isb=true, truemotion=false,global=true,delta=3,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
bv22=sup1.manalyse(isb=true, truemotion=false,global=true,delta=2,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
bv21=sup1.manalyse(isb=true, truemotion=false,global=true,delta=1,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
fv21=sup1.manalyse(isb=false,truemotion=false,global=true,delta=1,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
fv22=sup1.manalyse(isb=false,truemotion=false,global=true,delta=2,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
#fv23=sup1.manalyse(isb=false,truemotion=false,global=true,delta=3,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)

#a.mdegrain3(sup2,bv21,fv21,bv22,fv22,bv23,fv23,thSAD=499) # << this is the "2-fold" variant

interleave(a.mcompensate(sup2,fv22),a.mcompensate(sup2,fv21),a,a.mcompensate(sup2,bv21),a.mcompensate(sup2,bv22))
TMedian2().selectevery(5,2)

sup3 = last.msuper(pel=2,sharp=2)
bv33=sup3.manalyse(isb=true, truemotion=false,global=true,delta=3,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
bv32=sup3.manalyse(isb=true, truemotion=false,global=true,delta=2,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
bv31=sup3.manalyse(isb=true, truemotion=false,global=true,delta=1,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv31=sup3.manalyse(isb=false,truemotion=false,global=true,delta=1,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv32=sup3.manalyse(isb=false,truemotion=false,global=true,delta=2,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv33=sup3.manalyse(isb=false,truemotion=false,global=true,delta=3,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)

last.mdegrain3(sup3,bv31,fv31,bv32,fv32,bv33,fv33,thSAD=499) # valor Original thSAD=499
LSFMod(strength=60)
Interleave() 
Weave()

#stackvertical(a,last) # Alterado por mim.
return(last)


#--------------------------------------------------------------
# Helper functions

function MinBlur(clip clp, int r, int "uv")
{
uv = default(uv,3)
uv2 = (uv==2) ? 1 : uv
rg4 = (uv==3) ? 4 : -1
rg11 = (uv==3) ? 11 : -1
rg20 = (uv==3) ? 20 : -1
medf = (uv==3) ? 1 : -200

RG11D = (r==0) ? mt_makediff(clp,clp.sbr(),U=uv2,V=uv2)
\ : (r==1) ? mt_makediff(clp,clp.removegrain(11,rg11),U=uv2,V=uv2)
\ : (r==2) ? mt_makediff(clp,clp.removegrain(11,rg11).removegrain(20,rg20),U=uv2,V=uv2)
\ : mt_makediff(clp,clp.removegrain(11,rg11).removegrain(20,rg20).removegrain(20,rg20),U=uv2,V=uv2)
RG4D = (r<=1) ? mt_makediff(clp,clp.removegrain(4,rg4),U=uv2,V=uv2)
\ : (r==2) ? mt_makediff(clp,clp.medianblur(2,2*medf,2*medf),U=uv2,V=uv2)
\ : mt_makediff(clp,clp.medianblur(3,3*medf,3*medf),U=uv2,V=uv2)
DD = mt_lutxy(RG11D,RG4D,"x 128 - y 128 - * 0 < 128 x 128 - abs y 128 - abs < x y ? ?",U=uv2,V=uv2)
clp.mt_makediff(DD,U=uv,V=uv)
return(last)
}

# median of 5 clips from Helpers.avs by G-force
Function Median2(clip "input_1", clip "input_2", clip "input_3", clip "input_4", clip "input_5", string "chroma")
{
chroma = default(chroma,"process") #default is "process". Alternates: "copy first" or "copy second"
#MEDIAN(i1,i3,i5)
Interleave(input_1,input_3,input_5)
chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
m1 = selectevery(3,1)
#MAX(MIN(i1,i3,i5),i2)
m2 = input_1.MT_Logic(input_3,"min",chroma=chroma).MT_Logic(input_5,"min",chroma=chroma).MT_Logic(input_2,"max",chroma=chroma)
#MIN(MAX(i1,i3,i5),i4)
m3 = input_1.MT_Logic(input_3,"max",chroma=chroma).MT_Logic(input_5,"max",chroma=chroma).MT_Logic(input_4,"min",chroma=chroma)
Interleave(m1,m2,m3)
chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
selectevery(3,1)
chroma == "copy first" ? last.MergeChroma(input_1) : chroma == "copy second" ? last.MergeChroma(input_2) : last
Return(last)
}

function TMedian2(clip c) {
Median2( c.selectevery(1,-2), c.selectevery(1,-1), c, c.selectevery(1,1), c.selectevery(1,2) ) }

Pre Test Video - MediaInfo

General
Complete name               : K:\Pre Test.mp4
Format                      : MPEG-4
Format profile              : Base Media
Codec ID                    : isom (mp41/mp42)
File size                   : 8.65 MiB
Duration                    : 18 s 112 ms
Overall bit rate mode       : Variable
Overall bit rate            : 4 005 kb/s
Encoded date                : UTC 2021-03-31 01:05:07
Tagged date                 : UTC 2021-03-31 01:05:07

Video
ID                          : 1
Format                      : AVC
Format/Info                 : Advanced Video Codec
Format profile              : High@L3
Format settings             : CABAC / 2 Ref Frames
Format settings, CABAC      : Yes
Format settings, Reference  : 2 frames
Format settings, GOP        : M=3, N=15
Codec ID                    : avc1
Codec ID/Info               : Advanced Video Coding
Duration                    : 18 s 85 ms
Bit rate mode               : Variable
Bit rate                    : 3 686 kb/s
Maximum bit rate            : 4 000 kb/s
Width                       : 720 pixels
Height                      : 480 pixels
Display aspect ratio        : 4:3
Frame rate mode             : Constant
Frame rate                  : 29.970 (30000/1001) FPS
Standard                    : NTSC
Color space                 : YUV
Chroma subsampling          : 4:2:0
Bit depth                   : 8 bits
Scan type                   : Progressive
Bits/(Pixel*Frame)          : 0.356
Stream size                 : 7.95 MiB (92%)
Language                    : English
Encoded date                : UTC 2021-03-31 01:05:07
Tagged date                 : UTC 2021-03-31 01:05:07
Color range                 : Limited
Color primaries             : BT.709
Transfer characteristics    : BT.709
Matrix coefficients         : BT.709
Codec configuration box     : avcC

Audio
ID                          : 2
Format                      : AAC LC
Format/Info                 : Advanced Audio Codec Low Complexity
Codec ID                    : mp4a-40-2
Duration                    : 18 s 112 ms
Bit rate mode               : Constant
Bit rate                    : 320 kb/s
Nominal bit rate            : 128 kb/s
Channel(s)                  : 2 channels
Channel layout              : L R
Sampling rate               : 48.0 kHz
Frame rate                  : 46.875 FPS (1024 SPF)
Compression mode            : Lossy
Stream size                 : 708 KiB (8%)
Language                    : English
Encoded date                : UTC 2021-03-31 01:05:07
Tagged date                 : UTC 2021-03-31 01:05:07

Pos Test Video - MediaInfo

General
Complete name               : K:\Pos Test.avi
Format                      : AVI
Format/Info                 : Audio Video Interleave
File size                   : 181 MiB
Duration                    : 18 s 80 ms
Overall bit rate            : 84.1 Mb/s

Video
ID                          : 0
Format                      : MAGY
Codec ID                    : MAGY
Duration                    : 18 s 51 ms
Bit rate                    : 82.7 Mb/s
Width                       : 720 pixels
Height                      : 480 pixels
Display aspect ratio        : 3:2
Frame rate                  : 29.970 (30000/1001) FPS
Standard                    : NTSC
Bits/(Pixel*Frame)          : 7.983
Stream size                 : 178 MiB (98%)

Audio
ID                          : 1
Format                      : PCM
Format settings             : Little / Signed
Codec ID                    : 1
Duration                    : 18 s 80 ms
Bit rate mode               : Constant
Bit rate                    : 1 536 kb/s
Channel(s)                  : 2 channels
Sampling rate               : 48.0 kHz
Bit depth                   : 16 bits
Stream size                 : 3.31 MiB (2%)
Alignment                   : Aligned on interleaves
Interleave, duration        : 488  ms (14.62 video frames)

Thanks for your help!

[Selur]

Here is the Script in a PDF document (seems better to read than in the forum code area):

unusable due to line breaks in the pdf,... post it in a code block.
Whole thing 'feels' like an old pre-QTGMC deinterlacing script.
One thing that might cause problems is the use of DirectShowSource - since it might behave totally different depending on the source and system, better use LSMASHSource (which also supports hardware decoding).

Cu Selur

[NLP]

Here is the Script in a PDF document (seems better to read than in the forum code area):

unusable due to line breaks in the pdf,... post it in a code block.

Script

#Import("M:\temp2\source script2 Dr Paulo Pirej.avs")
# Plugins: - RemoveGrain
# - MaskTools2
# - Depan + DepanEstimate
# - MVTools2

SetMemoryMax(512)
SetMTMode(3)
LoadPlugin("M:\Temp2\RemoveGrain.dll")
LoadPlugin("M:\Temp2\MaskTools2-25.dll")
LoadPlugin("M:\Temp2\Depan.dll")
LoadPlugin("M:\Temp2\DepanEstimate.dll")
LoadPlugin("M:\Temp2\MVTools2.dll")

#mpeg2source("M:\Marcelo 27.5.20\1\01\01-1.d2v")
#mpeg2source("M:\Marcelo 999255703 - Fazer\01 ok 19m.d2v")
DirectShowSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4", fps = 25, ConvertFPS = True )
#DirectShowSource("M:\Luis Tonini 14.1.20 - Fazer\01 V14ES10 e.mp4")
#DirectShowSource("E:\Tereza Soares 13.8.20 - Fazer\01 TS 8capt V17JVCES10.mp4", fps = 25, ConvertFPS = True ) # para AVI encapsulado AVI Xvid da Avermidia ou MP4 da Aple
#DirectShowSource("M:\Neogrid 5-10-17\090 4capt V13SVSDV.avi", fps = 25, ConvertFPS = True ) # para AVI encapsulado AVI Xvid da Avermidia
#AVISource("E:\Cleiton\Super 8mm\03\03 ok.avi") 
#AVISource("E:\Luiz Itapema 23.9.20 - Fazer\18 LI SD8.avi", audio=true)
#AVISource("E:\Luiz Itapema 23.9.20 - Fazer\18 LI SD8.avi") # DV AVI
#AVISource("M:\Pedro 31-7-18\25 8capt V15ES10SDV.avi", audio=true)


SetMTMode(2)
#ConvertToYV12()
#ConvertToyv411()
#Crop(0,20,-0,-30,true)
#bob(0,0.5)

AssumeTFF()
SeparateFields()

a=last

clense(reduceflicker=false).merge(last,0.5).clense(reduceflicker=false)
mot=removegrain(11,0).removegrain(20,0).DepanEstimate(range=2)
take2=a.depaninterleave(mot,prev=2,next=2,subpixel=2)
clean1=take2.TMedian2().selectevery(5,2)

sup1 = clean1.minblur(1).removegrain(11,0).removegrain(11,0)
\ .mt_lutxy(clean1,"x 1 + y < x 2 + x 1 - y > x 2 - y ? ?",U=2,V=2)
\ .msuper(pel=2,sharp=0)
sup2 = a.msuper(pel=2,levels=1,sharp=2)
#bv23=sup1.manalyse(isb=true, truemotion=false,global=true,delta=3,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
bv22=sup1.manalyse(isb=true, truemotion=false,global=true,delta=2,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
bv21=sup1.manalyse(isb=true, truemotion=false,global=true,delta=1,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
fv21=sup1.manalyse(isb=false,truemotion=false,global=true,delta=1,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
fv22=sup1.manalyse(isb=false,truemotion=false,global=true,delta=2,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
#fv23=sup1.manalyse(isb=false,truemotion=false,global=true,delta=3,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)

#a.mdegrain3(sup2,bv21,fv21,bv22,fv22,bv23,fv23,thSAD=499) # << this is the "2-fold" variant

interleave(a.mcompensate(sup2,fv22),a.mcompensate(sup2,fv21),a,a.mcompensate(sup2,bv21),a.mcompensate(sup2,bv22))
TMedian2().selectevery(5,2)

sup3 = last.msuper(pel=2,sharp=2)
bv33=sup3.manalyse(isb=true, truemotion=false,global=true,delta=3,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
bv32=sup3.manalyse(isb=true, truemotion=false,global=true,delta=2,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
bv31=sup3.manalyse(isb=true, truemotion=false,global=true,delta=1,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv31=sup3.manalyse(isb=false,truemotion=false,global=true,delta=1,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv32=sup3.manalyse(isb=false,truemotion=false,global=true,delta=2,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv33=sup3.manalyse(isb=false,truemotion=false,global=true,delta=3,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)

last.mdegrain3(sup3,bv31,fv31,bv32,fv32,bv33,fv33,thSAD=499) # valor Original thSAD=499
LSFMod(strength=60)
Interleave() 
Weave()

#stackvertical(a,last) # Alterado por mim.
return(last)


#--------------------------------------------------------------
# Helper functions

function MinBlur(clip clp, int r, int "uv")
{
uv = default(uv,3)
uv2 = (uv==2) ? 1 : uv
rg4 = (uv==3) ? 4 : -1
rg11 = (uv==3) ? 11 : -1
rg20 = (uv==3) ? 20 : -1
medf = (uv==3) ? 1 : -200

RG11D = (r==0) ? mt_makediff(clp,clp.sbr(),U=uv2,V=uv2)
\ : (r==1) ? mt_makediff(clp,clp.removegrain(11,rg11),U=uv2,V=uv2)
\ : (r==2) ? mt_makediff(clp,clp.removegrain(11,rg11).removegrain(20,rg20),U=uv2,V=uv2)
\ : mt_makediff(clp,clp.removegrain(11,rg11).removegrain(20,rg20).removegrain(20,rg20),U=uv2,V=uv2)
RG4D = (r<=1) ? mt_makediff(clp,clp.removegrain(4,rg4),U=uv2,V=uv2)
\ : (r==2) ? mt_makediff(clp,clp.medianblur(2,2*medf,2*medf),U=uv2,V=uv2)
\ : mt_makediff(clp,clp.medianblur(3,3*medf,3*medf),U=uv2,V=uv2)
DD = mt_lutxy(RG11D,RG4D,"x 128 - y 128 - * 0 < 128 x 128 - abs y 128 - abs < x y ? ?",U=uv2,V=uv2)
clp.mt_makediff(DD,U=uv,V=uv)
return(last)
}

# median of 5 clips from Helpers.avs by G-force
Function Median2(clip "input_1", clip "input_2", clip "input_3", clip "input_4", clip "input_5", string "chroma")
{
chroma = default(chroma,"process") #default is "process". Alternates: "copy first" or "copy second"
#MEDIAN(i1,i3,i5)
Interleave(input_1,input_3,input_5)
chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
m1 = selectevery(3,1)
#MAX(MIN(i1,i3,i5),i2)
m2 = input_1.MT_Logic(input_3,"min",chroma=chroma).MT_Logic(input_5,"min",chroma=chroma).MT_Logic(input_2,"max",chroma=chroma)
#MIN(MAX(i1,i3,i5),i4)
m3 = input_1.MT_Logic(input_3,"max",chroma=chroma).MT_Logic(input_5,"max",chroma=chroma).MT_Logic(input_4,"min",chroma=chroma)
Interleave(m1,m2,m3)
chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
selectevery(3,1)
chroma == "copy first" ? last.MergeChroma(input_1) : chroma == "copy second" ? last.MergeChroma(input_2) : last
Return(last)
}

function TMedian2(clip c) {
Median2( c.selectevery(1,-2), c.selectevery(1,-1), c, c.selectevery(1,1), c.selectevery(1,2) ) }

[Selur]

-> did you try using LSMASHSource?

[jagabo]

I do the capture on HDMI AVC because it takes out several horizontal lines.

You are not supposed remove the "horizontal lines" (interlace comb artifacts). They are supposed to be in your capture. Your capture shouldn't be compressed with a lossy codec (AVC), and it should be encoded interlaced, not progressive. You are ruining your video even before you start filtering.

And yes, DirectShowSource() should only be used as a last resort. It's not frame accurate and probably the cause for all the out-of-order frames. Forcing the frame rate to 25p is also screwing up your video -- throwing away one of every 6 frames.

[NLP]

-> did you try using LSMASHSource?

Selur,

Thank you for pointing out the problem with DirectShowSource.

I'm a beginner with AviSynth.

Do you know some guide of instructions about how to change the DirectShowSource filter to LSMASHSource? and some example about how to use/include the LSMASHSource?

I never did it and I could not find a guide yet.



Thank you again.

[jagabo]

Change

DirectShowSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4", fps = 25, ConvertFPS = True )

to

LSmashVideoSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4")

If you need audio too use:

a = LSmashAudioSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4")
v = LSmashVideoSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4")
AudioDub(v,a)

Your AviSynth installation must include LSMASH. http://avisynth.nl/index.php/LSMASHSource

[NLP]

Change

DirectShowSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4", fps = 25, ConvertFPS = True )

to

LSmashVideoSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4")

If you need audio too use:

a = LSmashAudioSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4")
v = LSmashVideoSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4")
AudioDub(v,a)

Your AviSynth installation must include LSMASH. http://avisynth.nl/index.php/LSMASHSource

Hi jagabo,

Thank you so much! It finally worked! I did not know that it was so simple.


My VHS tapes are in bad condition (very oxidized). Because of that, my VHS video capture are very problematic.

Until today I couldn't find out how to compare 5 captures of the same VHS. I mean, frame by frame and choosing the best frames.

Do you have knowledge or/and could you give me an example of a script that does that?

How can I synchronize the captures, so that they are in the same time frame?


Thank you!

[jagabo]

If the files are already synchronized:

v1 = LSmashVideoSource("file1.mp4").Subtitle("1")
v2 = LSmashVideoSource("file2.mp4").Subtitle("2")
v3 = LSmashVideoSource("file3.mp4").Subtitle("3")
v4 = LSmashVideoSource("file4.mp4").Subtitle("4")
v5 = LSmashVideoSource("file5.mp4").Subtitle("5")

StackVertical(StackHorizontal(v1, v2, v3), StackHorizontal(v4, v5, BlankClip(v5)))

That will give you a 3x2 array of the clips. Assuming you have 720x576 caps it will be 2160x1152. If you need it smaller to fit on the screen add a resize at the end. Something like BilinearResize(width/2, height/2). Note that LSmashVideoSource() only works with MP4 and MOV files. For other types of files you can use LWlibavVideoSource() instead (also included in the LSMASH package).

To sync them you can use Trim() to remove frames from the start of a clip, or Loop() to add frames:

v1 = LSmashVideoSource("file1.mp4").Subtitle("1").Trim(15,0) # remove the first 15 frames
v2 = LSmashVideoSource("file2.mp4").Subtitle("2").Loop(7,0,0) # repeat frames 0 to 0 (ie one frame) to make a total of 7 (add 6)
etc.

If the videos aren't too corrupt you can use a median filter to pick the median pixel from the group of 5:

http://avisynth.nl/index.php/Median

That is useful for eliminating outliers.

[kelvin]

[Attachment 58249 - Click to enlarge]
Hello. Could someone help me.
How to solve this problem.

I speak Portuguese, but I wanted help from you

[jagabo]

You mean the the pink-ish horizontal stripes on the red-ish background. A video sample would be helpful.

[kelvin]

Yes

[kelvin]

You mean the the pink-ish horizontal stripes on the red-ish background. A video sample would be helpful.

Sim
Link - https://www.youtube.com/watch?v=bwR6Fqr9VpI

você pode ver melhor neste fundo vermelho

[jagabo]

Youtube's recompressed video isn't a good starting point. But that banding is going to be very hard to get rid of. It's worst on the opening titles. Most of it is in the U channel but only in areas of high V values (reds/yellows).


[Attachment 58266 - Click to enlarge]

I might be inclined to just rebuild the opening titles.

LWLibavVideoSource("Chapolin.mp4", cache=false, prefer_hw=2) 
src = last

ym = mt_binarize(120).Santiag(strh=2, strv=2).GreyScale()
yellow=ym.ColorYUV(off_y=-64, off_u=-64, off_v=36)
bg = blankClip(last, color=$A00030)
Overlay(bg, yellow, mask=ym.Blur(1.0).Blur(1.0))
Trim(34,34)
Loop(56,0,0)

last+src.Trim(56,0)

[Attachment 58265 - Click to enlarge]

Then apply some lesser filtering on the rest of the video. I'll look into that tomorrow.

[NLP]

If the files are already synchronized:

v1 = LSmashVideoSource("file1.mp4").Subtitle("1")
v2 = LSmashVideoSource("file2.mp4").Subtitle("2")
v3 = LSmashVideoSource("file3.mp4").Subtitle("3")
v4 = LSmashVideoSource("file4.mp4").Subtitle("4")
v5 = LSmashVideoSource("file5.mp4").Subtitle("5")

StackVertical(StackHorizontal(v1, v2, v3), StackHorizontal(v4, v5, BlankClip(v5)))

That will give you a 3x2 array of the clips. Assuming you have 720x576 caps it will be 2160x1152. If you need it smaller to fit on the screen add a resize at the end. Something like BilinearResize(width/2, height/2). Note that LSmashVideoSource() only works with MP4 and MOV files. For other types of files you can use LWlibavVideoSource() instead (also included in the LSMASH package).

To sync them you can use Trim() to remove frames from the start of a clip, or Loop() to add frames:

v1 = LSmashVideoSource("file1.mp4").Subtitle("1").Trim(15,0) # remove the first 15 frames
v2 = LSmashVideoSource("file2.mp4").Subtitle("2").Loop(7,0,0) # repeat frames 0 to 0 (ie one frame) to make a total of 7 (add 6)
etc.

If the videos aren't too corrupt you can use a median filter to pick the median pixel from the group of 5:

http://avisynth.nl/index.php/Median

That is useful for eliminating outliers.

Hi jagabo,

I was testing this script below. It looks like it worked at the beginning.

However, I let the script run for 1 minute and it generates a 45 minutes video.

I tested many times and it always happens.

Maybe some command is missing.

Do you know why it is happening?


Script

SetMemoryMax(512)
SetMTMode(3)
LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Median.dll")

v1 = LSmashVideoSource("K:\01 TS 4capt V17JVCES10.mp4").Subtitle("1") #.Trim(0,0) # remove the first 15 frames
v2 = LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4").Subtitle("2") #.Trim(0,0)
v3 = LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4").Subtitle("3") #.Trim(0,0)
v4 = LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4").Subtitle("4") #.Trim(0,0)
v5 = LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4").Subtitle("5") #.Trim(0,0)

SetMTMode(2)
a=last
StackVertical(StackHorizontal(v1, v2, v3), StackHorizontal(v4, v5, BlankClip(v5)))
median(v1, v2, v3, v4, v5, chroma=true)
#Median(v1, v2, v3, v4, v5, sync=10, chroma=false)
#MedianBlend(v1,v2,v3,v4,v5, low=0, high=0)
return(last)

Thank you!

[kelvin]

Youtube's recompressed video isn't a good starting point. But that banding is going to be very hard to get rid of. It's worst on the opening titles. Most of it is in the U channel but only in areas of high V values (reds/yellows).


[Attachment 58266 - Click to enlarge]

I might be inclined to just rebuild the opening titles.

LWLibavVideoSource("Chapolin.mp4", cache=false, prefer_hw=2) 
src = last

ym = mt_binarize(120).Santiag(strh=2, strv=2).GreyScale()
yellow=ym.ColorYUV(off_y=-64, off_u=-64, off_v=36)
bg = blankClip(last, color=$A00030)
Overlay(bg, yellow, mask=ym.Blur(1.0).Blur(1.0))
Trim(34,34)
Loop(56,0,0)

last+src.Trim(56,0)

[Attachment 58265 - Click to enlarge]

Then apply some lesser filtering on the rest of the video. I'll look into that tomorrow.

Follow the video link with 2 minutes. - https://www.4shared.com/video/88Sz0cueea/teste.html
Video is in the best possible quality

[jagabo]

[QUOTE=NLP;2616017]

I was testing this script below. It looks like it worked at the beginning.

However, I let the script run for 1 minute and it generates a 45 minutes video.

How long are the video files? The script should generate a video that's the length of your sources. I'm not sure what happens if one cap is much longer or shorter than the others.

SetMemoryMax(512)
SetMTMode(3)
LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Median.dll")

v1 = LSmashVideoSource("K:\01 TS 4capt V17JVCES10.mp4").Subtitle("1") #.Trim(0,0) # remove the first 15 frames
v2 = LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4").Subtitle("2") #.Trim(0,0)
v3 = LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4").Subtitle("3") #.Trim(0,0)
v4 = LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4").Subtitle("4") #.Trim(0,0)
v5 = LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4").Subtitle("5") #.Trim(0,0)

SetMTMode(2)
a=last
StackVertical(StackHorizontal(v1, v2, v3), StackHorizontal(v4, v5, BlankClip(v5)))
median(v1, v2, v3, v4, v5, chroma=true)
#Median(v1, v2, v3, v4, v5, sync=10, chroma=false)
#MedianBlend(v1,v2,v3,v4,v5, low=0, high=0)
return(last)

a=last doesn't do anything here. There is no "last" defined at that point. I'm surprised it doesn't give an error (maybe it doesn't because "a" is never used so it's ignoring that line). There's no point in stacking the videos since you are going to pass them to individually to median(). You don't want the numbers (subtitle()) in the video frames when you call median. I don't know if median() works with mtmode 2. You can reduce your script to:

SetMemoryMax(512) # disabled until you know the script is working
#SetMTMode(3) # disabled until you know the script is working

LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Median.dll")

v1 = LSmashVideoSource("K:\01 TS 4capt V17JVCES10.mp4") #.Trim(0,0) # remove the first 15 frames
v2 = LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4") #.Trim(0,0)
v3 = LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4") #.Trim(0,0)
v4 = LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4") #.Trim(0,0)
v5 = LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4") #.Trim(0,0)

#SetMTMode(2) # disabled until you know the script is working
median(v1, v2, v3, v4, v5, chroma=true)
#Median(v1, v2, v3, v4, v5, sync=10, chroma=false)
#MedianBlend(v1,v2,v3,v4,v5, low=0, high=0)

Open your script in VirtualDub and scrub around to verify it's working.

[jagabo]

Follow the video link with 2 minutes. - https://www.4shared.com/video/88Sz0cueea/teste.html

I can't download from that site. I get stuck with a 60 second countdown that doesn't change. You can upload files up to 500 MB directly to this site using the "Upload files/Manage attachments" button below the edit boxes.

[kelvin]

Follow the video link with 2 minutes. - https://www.4shared.com/video/88Sz0cueea/teste.html

I can't download from that site. I get stuck with a 60 second countdown that doesn't change. You can upload files up to 500 MB directly to this site using the "Upload files/Manage attachments" button below the edit boxes.

OK, I sent it as an attachment.

[jagabo]

Are you capturing VHS (or some other analog) tapes? Why are the videos 960x720 rather than 720x576? You usually want to be as close to the actual source for this type of filtering. You want to capture at 720x576, interlaced, with a lossless codec.

[kelvin]

This is the best source available.
Image was captured by television channel recording
Source was obtained with meeting of Brazilian fans
This format is to blame for the horrible Mexican television channel "Televisa".
They did a junk restoration, so we fans are trying to make the most of it.

[jagabo]

OK. I'll see what I can do.

[NLP]

[QUOTE=jagabo;2616031]

I was testing this script below. It looks like it worked at the beginning.

However, I let the script run for 1 minute and it generates a 45 minutes video.

How long are the video files? The script should generate a video that's the length of your sources. I'm not sure what happens if one cap is much longer or shorter than the others.

SetMemoryMax(512)
SetMTMode(3)
LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Median.dll")

v1 = LSmashVideoSource("K:\01 TS 4capt V17JVCES10.mp4").Subtitle("1") #.Trim(0,0) # remove the first 15 frames
v2 = LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4").Subtitle("2") #.Trim(0,0)
v3 = LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4").Subtitle("3") #.Trim(0,0)
v4 = LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4").Subtitle("4") #.Trim(0,0)
v5 = LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4").Subtitle("5") #.Trim(0,0)

SetMTMode(2)
a=last
StackVertical(StackHorizontal(v1, v2, v3), StackHorizontal(v4, v5, BlankClip(v5)))
median(v1, v2, v3, v4, v5, chroma=true)
#Median(v1, v2, v3, v4, v5, sync=10, chroma=false)
#MedianBlend(v1,v2,v3,v4,v5, low=0, high=0)
return(last)

a=last doesn't do anything here. There is no "last" defined at that point. I'm surprised it doesn't give an error (maybe it doesn't because "a" is never used so it's ignoring that line). There's no point in stacking the videos since you are going to pass them to individually to median(). You don't want the numbers (subtitle()) in the video frames when you call median. I don't know if median() works with mtmode 2. You can reduce your script to:

SetMemoryMax(512) # disabled until you know the script is working
#SetMTMode(3) # disabled until you know the script is working

LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Median.dll")

v1 = LSmashVideoSource("K:\01 TS 4capt V17JVCES10.mp4") #.Trim(0,0) # remove the first 15 frames
v2 = LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4") #.Trim(0,0)
v3 = LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4") #.Trim(0,0)
v4 = LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4") #.Trim(0,0)
v5 = LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4") #.Trim(0,0)

#SetMTMode(2) # disabled until you know the script is working
median(v1, v2, v3, v4, v5, chroma=true)
#Median(v1, v2, v3, v4, v5, sync=10, chroma=false)
#MedianBlend(v1,v2,v3,v4,v5, low=0, high=0)

Open your script in VirtualDub and scrub around to verify it's working.

Hi jagabo,

The video lasts for 1 hour and 40 minutes

The script ran very slowly, it seems to me that it is inserting all the frames of the 5 captures.

And the median is not working.

So I decided to get the Didée's script that I put at the end of this post *, and I added the LSMASHSource.

Then, this script transformed the 1 hour and 40 minute length video into a 2 hours of video. I really didn't understand why it is happening?...

Do I need to install FFmpeg to have Libavcodec?

I have K-lite Mega installed, does it no longer contain Libavcodec?



I used the command below to see the entire screen.

BilinearResize (width / 3, height / 3)

The 5 screens on the left perform normally, but the one on the right is static.


Image


[Attachment 58279 - Click to enlarge]






* Didée's script

#Import("M:\temp2\source script2 Dr Paulo Pirej.avs")
# Plugins: - RemoveGrain
# - MaskTools2
# - Depan + DepanEstimate
# - MVTools2

SetMemoryMax(512)
SetMTMode(3)
LoadPlugin("M:\Temp2\RemoveGrain.dll")
LoadPlugin("M:\Temp2\MaskTools2-25.dll")
LoadPlugin("M:\Temp2\Depan.dll")
LoadPlugin("M:\Temp2\DepanEstimate.dll")
LoadPlugin("M:\Temp2\MVTools2.dll")

#mpeg2source("M:\Marcelo 27.5.20\1\01\01-1.d2v")
#mpeg2source("M:\Marcelo 999255703 - Fazer\01 ok 19m.d2v")
DirectShowSource("M:\Welington 3.11.20 - Fazer\07 WE 3capt V17JVCES10.mp4", fps = 25, ConvertFPS = True )
#DirectShowSource("M:\Luis Tonini 14.1.20 - Fazer\01 V14ES10 e.mp4")
#DirectShowSource("E:\Tereza Soares 13.8.20 - Fazer\01 TS 8capt V17JVCES10.mp4", fps = 25, ConvertFPS = True ) # para AVI encapsulado AVI Xvid da Avermidia ou MP4 da Aple
#DirectShowSource("M:\Neogrid 5-10-17\090 4capt V13SVSDV.avi", fps = 25, ConvertFPS = True ) # para AVI encapsulado AVI Xvid da Avermidia
#AVISource("E:\Cleiton\Super 8mm\03\03 ok.avi") 
#AVISource("E:\Luiz Itapema 23.9.20 - Fazer\18 LI SD8.avi", audio=true)
#AVISource("E:\Luiz Itapema 23.9.20 - Fazer\18 LI SD8.avi") # DV AVI
#AVISource("M:\Pedro 31-7-18\25 8capt V15ES10SDV.avi", audio=true)


SetMTMode(2)
#ConvertToYV12()
#ConvertToyv411()
#Crop(0,20,-0,-30,true)
#bob(0,0.5)

AssumeTFF()
SeparateFields()

a=last

clense(reduceflicker=false).merge(last,0.5).clense(reduceflicker=false)
mot=removegrain(11,0).removegrain(20,0).DepanEstimate(range=2)
take2=a.depaninterleave(mot,prev=2,next=2,subpixel=2)
clean1=take2.TMedian2().selectevery(5,2)

sup1 = clean1.minblur(1).removegrain(11,0).removegrain(11,0)
\ .mt_lutxy(clean1,"x 1 + y < x 2 + x 1 - y > x 2 - y ? ?",U=2,V=2)
\ .msuper(pel=2,sharp=0)
sup2 = a.msuper(pel=2,levels=1,sharp=2)
#bv23=sup1.manalyse(isb=true, truemotion=false,global=true,delta=3,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
bv22=sup1.manalyse(isb=true, truemotion=false,global=true,delta=2,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
bv21=sup1.manalyse(isb=true, truemotion=false,global=true,delta=1,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
fv21=sup1.manalyse(isb=false,truemotion=false,global=true,delta=1,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
fv22=sup1.manalyse(isb=false,truemotion=false,global=true,delta=2,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)
#fv23=sup1.manalyse(isb=false,truemotion=false,global=true,delta=3,blksize=16,overlap=8,search=5,searchparam=4,DCT=5)

#a.mdegrain3(sup2,bv21,fv21,bv22,fv22,bv23,fv23,thSAD=499) # << this is the "2-fold" variant

interleave(a.mcompensate(sup2,fv22),a.mcompensate(sup2,fv21),a,a.mcompensate(sup2,bv21),a.mcompensate(sup2,bv22))
TMedian2().selectevery(5,2)

sup3 = last.msuper(pel=2,sharp=2)
bv33=sup3.manalyse(isb=true, truemotion=false,global=true,delta=3,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
bv32=sup3.manalyse(isb=true, truemotion=false,global=true,delta=2,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
bv31=sup3.manalyse(isb=true, truemotion=false,global=true,delta=1,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv31=sup3.manalyse(isb=false,truemotion=false,global=true,delta=1,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv32=sup3.manalyse(isb=false,truemotion=false,global=true,delta=2,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)
fv33=sup3.manalyse(isb=false,truemotion=false,global=true,delta=3,blksize=8,overlap=4,search=5,searchparam=4,DCT=5)

last.mdegrain3(sup3,bv31,fv31,bv32,fv32,bv33,fv33,thSAD=499) # valor Original thSAD=499
LSFMod(strength=60)
Interleave() 
Weave()

#stackvertical(a,last) # Alterado por mim.
return(last)


#--------------------------------------------------------------
# Helper functions

function MinBlur(clip clp, int r, int "uv")
{
uv = default(uv,3)
uv2 = (uv==2) ? 1 : uv
rg4 = (uv==3) ? 4 : -1
rg11 = (uv==3) ? 11 : -1
rg20 = (uv==3) ? 20 : -1
medf = (uv==3) ? 1 : -200

RG11D = (r==0) ? mt_makediff(clp,clp.sbr(),U=uv2,V=uv2)
\ : (r==1) ? mt_makediff(clp,clp.removegrain(11,rg11),U=uv2,V=uv2)
\ : (r==2) ? mt_makediff(clp,clp.removegrain(11,rg11).removegrain(20,rg20),U=uv2,V=uv2)
\ : mt_makediff(clp,clp.removegrain(11,rg11).removegrain(20,rg20).removegrain(20,rg20),U=uv2,V=uv2)
RG4D = (r<=1) ? mt_makediff(clp,clp.removegrain(4,rg4),U=uv2,V=uv2)
\ : (r==2) ? mt_makediff(clp,clp.medianblur(2,2*medf,2*medf),U=uv2,V=uv2)
\ : mt_makediff(clp,clp.medianblur(3,3*medf,3*medf),U=uv2,V=uv2)
DD = mt_lutxy(RG11D,RG4D,"x 128 - y 128 - * 0 < 128 x 128 - abs y 128 - abs < x y ? ?",U=uv2,V=uv2)
clp.mt_makediff(DD,U=uv,V=uv)
return(last)
}

# median of 5 clips from Helpers.avs by G-force
Function Median2(clip "input_1", clip "input_2", clip "input_3", clip "input_4", clip "input_5", string "chroma")
{
chroma = default(chroma,"process") #default is "process". Alternates: "copy first" or "copy second"
#MEDIAN(i1,i3,i5)
Interleave(input_1,input_3,input_5)
chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
m1 = selectevery(3,1)
#MAX(MIN(i1,i3,i5),i2)
m2 = input_1.MT_Logic(input_3,"min",chroma=chroma).MT_Logic(input_5,"min",chroma=chroma).MT_Logic(input_2,"max",chroma=chroma)
#MIN(MAX(i1,i3,i5),i4)
m3 = input_1.MT_Logic(input_3,"max",chroma=chroma).MT_Logic(input_5,"max",chroma=chroma).MT_Logic(input_4,"min",chroma=chroma)
Interleave(m1,m2,m3)
chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
selectevery(3,1)
chroma == "copy first" ? last.MergeChroma(input_1) : chroma == "copy second" ? last.MergeChroma(input_2) : last
Return(last)
}

function TMedian2(clip c) {
Median2( c.selectevery(1,-2), c.selectevery(1,-1), c, c.selectevery(1,1), c.selectevery(1,2) ) }

[jagabo]

The script ran very slowly, it seems to me that it is inserting all the frames of the 5 captures.

Which script?

And the median is not working.

What do you mean?

So I decided to get the Didée's script that I put at the end of this post *, and I added the LSMASHSource.

That script has nothing to do with the ajk's median() function. Andi it's completely in appropriate for your situation.

Then, this script transformed the 1 hour and 40 minute length video into a 2 hours of video. I really didn't understand why it is happening?...

Just to test, I used your short sample and the script did not do that. Did you make some other changes?

Do I need to install FFmpeg to have Libavcodec?

LWlibavVideoSource() is obviously working -- given you image with the stacked videos. If it wasn't working you would get an error message about an unknow function.

I have K-lite Mega installed, does it no longer contain Libavcodec?

That has nothing to do wiht LWlibavVideoSource().

I used the command below to see the entire screen.

BilinearResize (width / 3, height / 3)

The 5 screens on the left perform normally, but the one on the right is static.

I think don't understand what's going on. I suspect you pressed the Play Input Video button (second from the left) or selected Go -> Preview Input. That would only display the video in the left pane, the input pane. The right pane is the output pane -- the video after VirtualDub's filtering. If you want to see both panes press the Play Output Video (third from the left) or select Go -> Preview Filtered. But if you're not performing any filtering in VirtualDub there's no need to see the output video.

[NLP]

The script ran very slowly, it seems to me that it is inserting all the frames of the 5 captures.

Which script?

And the median is not working.

What do you mean?

So I decided to get the Didée's script that I put at the end of this post *, and I added the LSMASHSource.

That script has nothing to do with the ajk's median() function. Andi it's completely in appropriate for your situation.

Then, this script transformed the 1 hour and 40 minute length video into a 2 hours of video. I really didn't understand why it is happening?...

Just to test, I used your short sample and the script did not do that. Did you make some other changes?

Do I need to install FFmpeg to have Libavcodec?

LWlibavVideoSource() is obviously working -- given you image with the stacked videos. If it wasn't working you would get an error message about an unknow function.

I have K-lite Mega installed, does it no longer contain Libavcodec?

That has nothing to do wiht LWlibavVideoSource().

I used the command below to see the entire screen.

BilinearResize (width / 3, height / 3)

The 5 screens on the left perform normally, but the one on the right is static.

I think don't understand what's going on. I suspect you pressed the Play Input Video button (second from the left) or selected Go -> Preview Input. That would only display the video in the left pane, the input pane. The right pane is the output pane -- the video after VirtualDub's filtering. If you want to see both panes press the Play Output Video (third from the left) or select Go -> Preview Filtered. But if you're not performing any filtering in VirtualDub there's no need to see the output video.

Hi Jagabo

I'm sorry, I made a big mess in my preview post (above).

I posted the script that you gave me, the one below.

SetMemoryMax (512) # desabilitado até que você saiba que o script está funcionando
#SetMTMode (3) # desabilitado até que você saiba que o script está funcionando

LoadPlugin ("M: \ Temp2 \ LSMASHSource.dll")
LoadPlugin ("M: \ Temp2 \ Median.dll")

v1 = LSmashVideoSource ("K: \ 01 TS 4capt V17JVCES10.mp4") # .Trim (0,0) # remove os primeiros 15 quadros
v2 = LSmashVideoSource ("K: \ 01 TS 5capt V17JVCES10 e.mp4") # .Trim (0,0)
v3 = LSmashVideoSource ("K: \ 01 TS 6capt V17JVCES10 e.mp4") # .Trim (0,0)
v4 = LSmashVideoSource ("K: \ 01 TS 8capt V17JVCES10 e.mp4") # .Trim (0,0)
v5 = LSmashVideoSource ("K: \ 01 TS 10capt V17JVCES10 e.mp4") # .Trim (0,0)

#SetMTMode (2) # desabilitado até que você saiba que o script está funcionando
mediana (v1, v2, v3, v4, v5, croma = verdadeiro)
#Median (v1, v2, v3, v4, v5, sync = 10, chroma = falso)
#MedianBlend (v1, v2, v3, v4, v5, baixo = 0, alto = 0)

As I could not get the correct Preview because I pressed the Play Input Video button (the second from the left)

This would only display the video in the left panel, as you predicted.

So I thought the median command [ median(v1, v2, v3, v4, v5, chroma=true) ] was not working.

Soon, I had the idea of ​​inserting the code into the script.

BilinearResize (width / 3, height / 3)


That ended up creating more confusion.

I shouldn't have done that, I should have posted the result of the original script.

So now I followed your last recommendation.

To see the two panels, press Play Output Video (third from left)

Then I noticed that the "median" is working, but generating many more frames.

It loses the sync of the frames as it advances in time, causing distortions in some frames in the video. It’s getting in very slow motion.


To give you an example The video below with only 1 minute long. The script created a video with more than an hour.


https://drive.google.com/file/d/1IIK_Ps_TJezi_bleiNK1f7PbBaNtR7oM/view


So I thought I was missing some codec on my system.

So following your guidance, I used the script from ajk and Andi

code:

# editar os plugins que vai usar
SetMemoryMax(512)
#SetMTMode(3)

LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Average_21Oct05.dll")
LoadPlugin("M:\Temp2\Median.dll")

clip1=LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4").Subtitle("2") #.Trim(0,0)
clip2=LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4").Subtitle("3") #.Trim(0,0)
clip3=LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4").Subtitle("4") #.Trim(0,0)
clip4=LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4").Subtitle("5") #.Trim(0,0)

#SetMTMode(2)
M1=Median1(clip1,clip2,clip3)
M2=Median1(clip2,clip3,clip4)
M3=Median1(clip1,clip3,clip4)
M4=Median1(clip1,clip2,clip4)
Average(M1, 0.25, M2, 0.25, M3, 0.25, M4, 0.25)

Function Median1(clip input_1, clip input_2, clip input_3, string "chroma")
{# median of 3 clips from Helpers.avs by G-force

chroma = default(chroma,"process") #default is "process". Alternates: "copy first" or "copy second"

Interleave(input_1,input_2,input_3)
chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
SelectEvery(3,1)

chroma == "copy first" ? last.MergeChroma(input_1) : chroma == "copy second" ? last.MergeChroma(input_2) : last

Return(last)
}

Running the ajk script, I noticed that it got better at the beginning.

It loses the sync of the frames as it advances in time, causing distortions in some frames in the video.

It is getting in slow motion, but faster than the previous one.
But it generated a 20-hour video, I don't know how long I let the script run.


But, again, another example, the video below with 1 minute long. The script created a video with more than an hour.


https://drive.google.com/file/d/1lKCz1HZEIgsMtAWKkKaaPfrjVF7Eh4Wo/view


So I believe there must be something wrong with my PC system.

I am using AviSynth 2.6, what is your recommendation? Maybe upgrade to 32-bit Plus?



Thank you Jagabo for all your help.

[jagabo]

Your videos have very long running times in part because frame rate has changed to ~1 fps. When I run that script on my computer I don't get that. But I'm running a recent built of AviSynth+ 64 bit, not 32 bit AviSynth 2.6 mt. So all the filters are probably different versions too.

The median1() function in your script works similarly to ajk's median with three input videos. But it has no ability to resync if any of the videos have missing or extra frames.

Also the way you're building all those Mx clips doesn't make any sense. Your script should look more like:

SetMemoryMax(512)
#SetMTMode(3)

LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Median.dll")

Function Median1(clip input_1, clip input_2, clip input_3, string "chroma")
{# median of 3 clips from Helpers.avs by G-force
    chroma = default(chroma,"process") #default is "process". Alternates: "copy first" or "copy second"

    Interleave(input_1,input_2,input_3)
    chroma == "process" ? Clense(reduceflicker=false) : Clense(reduceflicker=false,grey=true)
    SelectEvery(3,1)

    chroma == "copy first" ? last.MergeChroma(input_1) : chroma == "copy second" ? last.MergeChroma(input_2) : last

    Return(last)
}

clip1=LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4").Subtitle("2") #.Trim(0,0)
clip2=LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4").Subtitle("3") #.Trim(0,0)
clip3=LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4").Subtitle("4") #.Trim(0,0)
clip4=LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4").Subtitle("5") #.Trim(0,0)

# ignoring one of the videos because Median1() only accepts 3 video:
Median1(clip1, clip2, clip3)
AssumeFPS(clip1.framerate) #fixing the frame rate, assuming clip1 has the correct frame rate

Or using ajk's median():

SetMemoryMax(512)
#SetMTMode(3)

LoadPlugin("M:\Temp2\LSMASHSource.dll")
LoadPlugin("M:\Temp2\Median.dll")

clip1=LSmashVideoSource("K:\01 TS 5capt V17JVCES10 e.mp4").Subtitle("2") #.Trim(0,0)
clip2=LSmashVideoSource("K:\01 TS 6capt V17JVCES10 e.mp4").Subtitle("3") #.Trim(0,0)
clip3=LSmashVideoSource("K:\01 TS 8capt V17JVCES10 e.mp4").Subtitle("4") #.Trim(0,0)
clip4=LSmashVideoSource("K:\01 TS 10capt V17JVCES10 e.mp4").Subtitle("5") #.Trim(0,0)

# ignoring one of the videos because Median() only accepts 3 video:
Median(clip1, clip2, clip3)
# I'm guessing the frame rate will be correct in this case