Quantcast Doubling Avisynth script to remove comets? - digitalFAQ Forum
  #1  
01-12-2019, 11:15 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
Can you double up on a script (RemoveSpotsMC) to finish removing comet tails?? Sanlyn helped me with this script several years ago. I am using it on another Band video(recorded a year after the initial video I used this script for...same crappy video) that has same comet tails. This time it doesn't want to remove all of them. (or most) OR, can I tweek this script to where it is stronger at removing them? The script was a 2 step process as shown below.

Thank you.

Quote:
# ###############################
# SCRIPT FOR STEP 1
# ###############################

AviSource("E:\forum\sirbyron\C\Chs1981Samplecc.avi ") #<-- adjust path and file name.
Crop(16,0,0,-10).AddBorders(0,0,0,2)
ConvertToYV12(interlaced=true)
AssumeTFF().QTGMC(preset="fast",sharpness=0.8)
# ------- Save in VirtualDub as Lagarith YV12 using fast compress" ----------------

Code:

# ###############################
# SCRIPT FOR STEP 2
# ###############################

Import("D:\Avisynth 2.5\plugins\chubbyrain2.avs")
Import("D:\Avisynth 2.5\plugins\RemoveSpotsMC.avs")

AviSource("path to Step 1 output\filename.avi") #<--- adjust path and name for step 1 file.
chubbyrain2()
ChromaShift(c=-8,u=-2,L=-4)
mergechroma(awarpsharp2(depth=32, type=1, blur=3, chroma=6))
AutoAdjust(temporal_radius=30,high_quality=true,au to_gain=true,gain_mode=1,\
dark_limit=1.0,gamma_limit=3.0,auto_balance=true)
SmoothLevels(8, 0.85, 255,16,235,chroma=200,limiter=0,tvrange=true,dithe r=100)
v1=last
e=v1.SelectEven().RemoveSpotsMC().Blur(0.4,0.2).Sh arpen(0.3)
o=v1.SelectOdd().RemoveSpotsMC().Blur(0.4,0.2).Sha rpen(0.3)
Interleave(e,o)
AssumeTFF().SeparateFields().SelectEvery(4,0,3).We ave()
Santiag()
ConvertToRGB32(matrix="Rec601",interlaced=true)
AddBorders(8,4,8,4)
return last

# ---- output is RGB32. You can save as Lagarith RGB32 in a
# ---- new file via "fast compress" and then run VirtualDub
# ---- as a separate step -- or load VirtualDub filters on
# ---- Avisynth output and save as Lagarith YV12 for an encoder
# ---- via "Full processing mode".

# ------ VirtualDub plugins used later (See .vcf file)
# - fxVHS.vdf (FlaXen VHS) Ise chroma shift only!! Disable all other filters in the window.
# - hue.vdf (hue/saturation/intensity - Donald Graft)
# - ColorMill.vdf
# - gradation.vdf

Reply With Quote
Someday, 12:01 PM
admin's Avatar
Ads / Sponsors
 
Join Date: ∞
Posts: 42
Thanks: ∞
Thanked 42 Times in 42 Posts
  #2  
01-12-2019, 11:57 PM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
Quote:
Originally Posted by sirbyron View Post
Can you double up on a script (RemoveSpotsMC) to finish removing comet tails?? Sanlyn helped me with this script several years ago. I am using it on another Band video(recorded a year after the initial video I used this script for...same crappy video) that has same comet tails. This time it doesn't want to remove all of them. (or most) OR, can I tweek this script to where it is stronger at removing them? The script was a 2 step process as shown below.

Thank you.
Anyone working on this problem will need a video sample to find out why the old script isn't working well and to develop and test new procedures if necessary.

Also note that the script you posted contains errors that will make the script crash. Among the errors are unintentional spaces inserted in the text, as here:

Quote:
e=v1.SelectEven().RemoveSpotsMC().Blur(0.4,0.2).Sh arpen(0.3)
o=v1.SelectOdd().RemoveSpotsMC().Blur(0.4,0.2).Sha rpen(0.3)
When posting scripts or text where variable-spaced fonts or posting errors will create errors, use the "code" hashmark (#) icon to wrap script text.
Reply With Quote
  #3  
01-13-2019, 08:09 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
I will try to get a sample soon. Yes, I just copied and pasted the script from an old post , next time I will use the hashmark. Didnt know about that. Would you like the sample with the above script used or the raw YUY2?
Reply With Quote
  #4  
01-13-2019, 08:42 PM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
The video alone will suffice, thanks.
Reply With Quote
  #5  
01-14-2019, 08:11 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
Here is my sample. Comets are in frames 2,40,47,93 (double comet), 94 (double comet) and 109. Possibly more but these I counted at a glance. I recopied the script with the code #. Hope that helps. Also, the colors are way off to the left in the Histogram but when the camera pans back to full field view (not in sample) they come back to center. Just mentioning that as I guess I just have to have that sacrifice. This is the only part of video closeup.

**This is the raw capture and not used with script below.** The script below I did try but didn't remove the comets.



Code:
# ###############################
#       SCRIPT FOR STEP 1
# ###############################

AviSource("E:\forum\sirbyron\C\Chs1981Samplecc.avi") #<-- adjust path and file name.
Crop(16,0,0,-10).AddBorders(0,0,0,2)
ConvertToYV12(interlaced=true)
AssumeTFF().QTGMC(preset="fast",sharpness=0.8)
# ------- Save in VirtualDub as Lagarith YV12 using fast compress" ----------------

Code:

# ###############################
#       SCRIPT FOR STEP 2
# ############################### 

Import("D:\Avisynth 2.5\plugins\chubbyrain2.avs")
Import("D:\Avisynth 2.5\plugins\RemoveSpotsMC.avs")

AviSource("path to Step 1 output\filename.avi") #<--- adjust path and name for step 1 file.
chubbyrain2()
ChromaShift(c=-8,u=-2,L=-4)
mergechroma(awarpsharp2(depth=32, type=1, blur=3, chroma=6))
AutoAdjust(temporal_radius=30,high_quality=true,auto_gain=true,gain_mode=1,\
  dark_limit=1.0,gamma_limit=3.0,auto_balance=true)
SmoothLevels(8, 0.85, 255,16,235,chroma=200,limiter=0,tvrange=true,dither=100)
v1=last
e=v1.SelectEven().RemoveSpotsMC().Blur(0.4,0.2).Sharpen(0.3)
o=v1.SelectOdd().RemoveSpotsMC().Blur(0.4,0.2).Sharpen(0.3)
Interleave(e,o)
AssumeTFF().SeparateFields().SelectEvery(4,0,3).Weave()
Santiag()
ConvertToRGB32(matrix="Rec601",interlaced=true)
AddBorders(8,4,8,4)
return last

# ---- output is RGB32. You can save as Lagarith RGB32 in a
# ---- new file via "fast compress" and then run VirtualDub 
# ---- as a separate step -- or load VirtualDub filters on 
# ---- Avisynth output and save as Lagarith YV12 for an encoder
# ---- via "Full processing mode".

# ------ VirtualDub plugins used later (See .vcf file)
# - fxVHS.vdf (FlaXen VHS) Ise chroma shift only!! Disable all other filters in the window.
# - hue.vdf (hue/saturation/intensity - Donald Graft)
# - ColorMill.vdf
# - gradation.vdf


Read more: http://www.digitalfaq.com/forum/video-restore/6379-virtualdub-filters-improve-2.html#ixzz5cdZUwIQW


Attached Files
File Type: avi DFaqSample.avi (97.57 MB, 5 downloads)
Reply With Quote
  #6  
01-15-2019, 02:55 AM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
Thanks for the sample, sirbyron, but it's uncompressed RGB. Hopefully that's not the way it's being captured. If it were the usual YUY2 compressed with huffyuv or Lagarith, your 99mb sample would only be 25mb. With lossless compression you could have included a few more frames for temporal filters to work with. Also, conversion to RGB makes many filtering factors ineffective, since some repairs have to be in the original YUV colorspace.

When you edit a sample in VirtualDub, save it using "direct stream copy" to preserve the original colorspace and avoid colorspace conversions.
Reply With Quote
  #7  
01-15-2019, 08:50 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
I think this should be correct.

Quote:
When you edit a sample in VirtualDub, save it using "direct stream copy" to preserve the original colorspace and avoid colorspace conversions.
Sorry, yes, I processed via "Full Processing Mode" when I made the sample instead of "Direct Stream Copy".

Is there an application that will tell you the correct colorspace, ie YUY2, RGB, etc after you have made the file? I can select "File Information" in Vdub and it does tell me "Lagarith" or "Internal DIB decoder" but doesn't mention colorspace.

Just curious, When I capture, would using "No Recompression YUY2" in my capture settings instead of "Lagarith Lossless Codec" yield any better processing results? I do know the file size is about 3 times the size.

I have been away from this a few years so pardon my rustiness.


Attached Files
File Type: avi CHS82Sample.avi (98.39 MB, 2 downloads)
Reply With Quote
  #8  
01-15-2019, 09:33 PM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
Quote:
Originally Posted by sirbyron View Post
Is there an application that will tell you the correct colorspace, ie YUY2, RGB, etc after you have made the file? I can select "File Information" in Vdub and it does tell me "Lagarith" or "Internal DIB decoder" but doesn't mention colorspace.
MediaInfo or MediaInfoXP gives colorspace info such as YUV 4.2.0 (which would be YV12), or 4.2.2 I(which would be YUY2 or similar). Or use Avisynth's Info() function:

Code:
AviSource(or whatever to open a video)
Info()
http://avisynth.nl/index.php/Info

Quote:
Originally Posted by sirbyron View Post
When I capture, would using "No Recompression YUY2" in my capture settings instead of "Lagarith Lossless Codec" yield any better processing results
No. Don't capture uncompressed. You gain nothing and you just burn up CPU time.

Thanks for the new sample. Seems to look worse every time I see it. I'll give it a look-see tonight.
Reply With Quote
  #9  
01-15-2019, 09:58 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
Thanks for colorspace info.

Quote:
Seems to look worse every time I see it. I'll give it a look-see tonight.
Yep. Lol. Old footage and 3rd Gen Copy. The best I have to work with though. Thanks and appreciate.
Reply With Quote
  #10  
01-17-2019, 08:57 PM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
Sorry for the delay.

These are similar noise and resolution problems as the earlier samples. After some experimentation to make some improvements over older scripts, things were going along fairly well when....



Take a look at the girl's face in the left-hand image. I'm sure she didn't have a mustache in previous frames! I hated throwing another filter into the mix for only a single very inconvenient glitch, but you could see that little smudge a mile away. The right-hand image is the cleaned, "after filtering" version. The final result of the two scripts below is attached as CHS82_640x480.mp4.

The scripts this time are similar to the older ones, but I made some changes. I'll address those changes in the notes that follow:

Code:
# ###############################
#       SCRIPT FOR STEP 1
# ###############################

src=AviSource("D:\forum\faq\sirbyron\CHS82Sample.avi") #<-- adjust path and file name.
exclp=src.Crop(16,0,-8,-8).ConvertToYV16(interlaced=true)

src
#Crop(14,0,-2,-8)      #<- test
ConvertToYV16(interlaced=true)
AutoAdjust(temporal_radius=15,high_quality=true,auto_gain=true,gain_mode=0,\
  dark_limit=1.0,bright_limit=4,avg_safety=0.5,gamma_limit=3,external_clip=exclp)
Levels(16,0.95,255,16,235,dither=true,coring=false)
# Histogram("Levels")  #<- test
# return last          #<- test

ConvertToYV12(interlaced=true)
SeparateFields()
e=SelectEven().RemoveSpotsMC3()
o=SelectOdd().RemoveSpotsMC3()
Interleave(e,o)
Weave()
return last
# ------- Save in VirtualDub as Lagarith YV12 using "fast compress" --------
The lines with "#<- test" to the right of the statement are included to show one of the methods often used to test procedures and other elements while a script is being built. You can keep special lines in the script but disable them by placing comment markers ("#") in front.

src=AviSource("D:\forum\faq\sirbyron\CHS82Sample.a vi")
exclp=src.Crop(16,0,-8,-8).ConvertToYV16(interlaced=true)
For script#1 I needed two copies of the video, so I had to give each copy a made-up name. "src" is a name that's short for "source", and it's assigned to the incoming .avi sample. Then I made a copy in memory of the src file. I named the copy "exclp", cropped off the black borders, and changed its colorspace to YV16 (Note: you must have Avisynth version 2.6 or later to use YV16 color -- Otherwise, change the colorspace to YV12 instead of YV16). The "exclip" name is short for "external_clip" and is used by AutoAdjust as a calibration device so that black borders don't affect the overall level adjustments. By the time the code gets to AutoAdjust, both "src' and "exclp" should be the same colorspace.

src
ConvertToYV16(interlaced=true)
AutoAdjust(temporal_radius=15,high_quality=true,au to_gain=true,gain_mode=0,\
dark_limit=1.0,bright_limit=4,avg_safety=0.5,gamma _limit=3,external_clip=exclp)

the "src" statement by itself brings focus back to the original "src' input file. All code following that line will apply to the input source. The AutoAdjust statement is similar to the old one, but I removed the chroma balance parameter. That parameter really wasn't correcting color very well and just seemed to wash things out.

Levels(16,0.95,255,16,235,dither=true,coring=false ) keeps signal levels within y=16-235 because AutoAdjust doesn't always behave itself in that regard. Note the gamma setting in Levels at 0.95. Overall brightness perception is largely affected by that figure, which you can raise and lower at your heart's content. Unfortunately neither AutoAdjust nor Levels can fully correct the oddball brightness changes caused by the camera's disastrous automatic "features".

The big change here is that in step1 I've used a RemoveSpots plugin instead of deinterlacing. SeparateFields() and breaking the fields into even and odd streams makes RemoveSpotsMC3 faster and more effective (the spots and comets look more random to the filter, so they "look like" noise). RemoveSpotsMc3 is part of a bigger scripted plugin, which is RemoveSpotsMC4.avsi. It can be downloaded at http://www.digitalfaq.com/forum/atta...vespotsmc4avsi. The .avsi actually includes 6 different versions of RemoveSpots, each more powerful than the other, and any one of which you can use as required:
- RemoveSpots().
- RemoveSpotsMC().
- RemoveSpotsMC2().
- RemoveSpotsMC3().
- RemoveSpotsMC4().
- RemoveSpotsMC5().
In this case RemoveSpotsMC3, which iterates 3 times through groups of frames, was required to get rid of the rain of comets. The more iterations are performed, the slower the plugin runs. RemoveSpotsMC3 is slooowwww (it poked along at a pathetic 2.6 fps in processing), which is why I've used it in a separate script and saved the rest for script #2. Versions MC4 and MC5 are terminally slow and actually not better than version 3. RemoveSpotsMC3 also does a little general denoising as a bonus.

You've used a version of RemoveSpots previously, so you probably have some of the support files needed. The basic requirements are the RemoveGrain package, RgTools, and MaskTools2. The last two are supplied with the QTGMC installation package, whose digitalfaq .zip download has recently been updated. If you need any of these, here are links:

DeFlicker.dll: http://avisynth.org.ru/deflicker/deflicker04.zip

DeSpot.dll 3.6.1.0: http://avisynth.org.ru/despot/despot...ot/despot.html.

The older MaskTools.dll v1.5.8: http://avisynth.nl/index.php/MaskTools. NOTE: You can have the old makstool.dll and the newer MaskTools2.dll in your plugins at the same time. Each version contains different function names, so there are no coding conflicts between the two versions.

Updated QTGMC plugins package: http://www.digitalfaq.com/forum/atta...kagenov2017zip.

MaskTools2.dll: supplied with the QTGMC package.

RGTools.dll: supplied with the QTGMC package.

RemoveDirt v-0.9 package:
http://www.digitalfaq.com/forum/atta...ovedirt_v09zip.

All of the above-named plugins are required by other complex plugins as well.

Code:
# ###############################
#       SCRIPT FOR STEP 2
# ###############################

Import("D:\Avisynth 2.5\plugins\MDG2.avs")         #<--- adjust path for plugin.

AviSource("D:\forum\faq\sirbyron\CHS82_STEP1.avi") #<--- adjust path & name for step 1 file.
AssumeTFF()
QTGMC(preset="faster",EZDenoise=4,denoiser="dfftest",ChromaNoise=true,\
    DenoiseMC=true,sharpness=0.8,FPSDivisor=2)
MDG2()
RemoveSpotsMC()
BiFrost(interlaced=false)
Cnr2(mode="ooo", scdthr=255.0, ln=255, lm=222, un=255, um=255, vn=255, vm=255)
ChromaShift(c=-8,u=-2,L=-4)
mergechroma(awarpsharp2(depth=32, type=1, blur=3, chroma=6).aWarpSharp2(depth=10))
TemporalSoften(4,4,8,15,2)
LimitedSharpenfaster(strength=250)
Crop(14,0,-2,-8).AddBorders(8,4,8,4)
AddGrainC(1.5,1.5)
ConvertToRGB32(matrix="Rec601",interlaced=false)
return last
QTGMC's parameters are changed since the last version of these scripts, to pump up more denoising, to smooth horizontal shimmer, and to do a little chroma cleaning. One parameter to note is FPSDivisor=2, which discards alternate fields and outputs 29.97fps progressive video. This can be encoded as interlaced to match other interlaced clips that might be joined to it (the encoder will use fake interlace flags during the encode. Many DVD/BD players will play it as interlaced anyway). The video sample has those really ugly sloppy interlace combing artifacts which, no matter what you do, won't go away. It's a common fault of consumer camera shutters and processing, and this technique is a common way of handling it. There's so much zoom distortion and jumpy motion anyway, no one will ever know the difference.

The new kid on the block is the MDG2.avs plugin. This is a customized version of the MaskTools2 MDegrain2 function which is used here to smooth some buzzy edges. All the support files are either in the QTGMC package or have been referenced above. This isn't an especially slow filter, but it did slow down processing to about 5.5fps. MDG2.avs can be downloaded from http://www.digitalfaq.com/forum/atta...1&d=1524917341.

A couple of unexpected black spots showed up in very sensitive and visible spots, so I had throw in another mild dose of RemoveSpots, this time the faster and more simple RemoveSpotsMC(). It's a function in the RemoveSpotsMC4.avsi plugin discussed earlier.

BiFrost is an anti-rainbow filter used here to curb some of the chroma discoloration, although there's an awful lot of it. Bifrost v2: http://www.digitalfaq.com/forum/atta...-bifrost_v2zip.

TemporalSoften(4,4,8,15,2) is an Avisynth built-in function. A very mild but effective filter. Helps smooth buzzy edges and floating grunge.

LimitedSharpenFaster.avsi is a sharpener, but not your typical under-engineered NLE type. It's "limited" in that it tries to sharpen without creating edge artifacts such as halo's or ringing. It's used here at a strength of 250, which is not far from its default. http://avisynth.nl/index.php/LimitedSharpen. It requires Masktools2 and Rgtools, which are in the QTGMC package. It also requires aWarpSharp2, which you already have.

AddGrainC(1.5,1.5) adds some very fine film-like dithered grain to help avoid an over-filtered or denuded look. The .dll is furnished with the QTGMC plugin.

At the end of the script the video is converted to RGB32 for the VirtualDub filters that I applied to Avisynth's output before saving the file. The three filters I used were ColorCamcorderDenoise v1.7, ColorMill, and gradation curves. The settings I used were saved in a .vcf file, which is attached as CHS82_VDub_settings.vcf. To load the filters and settings, open VirtualDub and click "file..." -> "load processing settings...", then locate and select the .vcf file. The three filters must exist in your VDub plugins folder or the .vcf won't work. In case you don't have them, the three VDub filters are:

ColorMill.vdf 2.1: http://www.digitalfaq.com/forum/atta...colormill21zip.

gradation.vdf 4.5: http://www.digitalfaq.com/forum/atta...1&d=1489408797.

ColorCamcorderDenoise 1.7: http://www.digitalfaq.com/forum/atta...1&d=1544578132. There are multiple versions of CCD, but you can have multiple versions in your plugins folder because they have slightly different .vdf names and internal code.


Attached Images
File Type: jpg facial spot before and after.jpg (64.6 KB, 33 downloads)
Attached Files
File Type: mp4 CHS82_640x480.mp4 (8.94 MB, 2 downloads)
File Type: vcf CHS82_VDub_settings.vcf (3.7 KB, 1 downloads)
Reply With Quote
The following users thank sanlyn for this useful post: sirbyron (01-17-2019)
  #11  
01-17-2019, 09:26 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
Wow. Looks so much better. Thank you! I look fwd to working on this over the weekend.

I do have Avisynth 2.6.0.4 but will have to download a few of the newer and updated filters. You are correct, no mustache looks better. It always amazes me the support and knowledge on this forum. If I only knew half of what most of you do I would be so happy. Again, really appreciate.
Reply With Quote
  #12  
01-18-2019, 07:07 AM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
Yes, some of the new plugins helped get better results. I'm still dissatisfied with one problem in the overall color balance. There is a visible saturation problem with cyan (green in shadow areas) but it's not a color issue, it's a saturation issue. Tape aging and storage conditions have produced a pretty convoluted histogram, with greenish-cyan saturation mainly in the midtones. You can reduce cyan saturation overall, but it affects other areas (for instance, it will turn the blue skirt and blue uniforms toward black, which you don't want). Advanced color controls like those in AfterEffects, ColorFinesse, etc., allow you to reduce saturation of specific colors only in specific ranges rather than across the full spectrum. Reducing saturation isn't the same thing as changing hue. While those pricey apps can't perform repairs the way Avisynth can, they do have advantges in other areas. Unfortunately those apps cost plenty, beyond the means of most of us.
Reply With Quote
The following users thank sanlyn for this useful post: sirbyron (01-18-2019)
  #13  
01-18-2019, 07:27 AM
lordsmurf's Avatar
lordsmurf lordsmurf is online now
Site Staff | Video
 
Join Date: Dec 2002
Posts: 8,475
Thanked 1,392 Times in 1,224 Posts
Yes, sometimes you can double filters/scripts for better effects than simply increasing the power of the script. A good example of this is my stabmod(), a simple modification of stab(), but tweaked for VHS vertical jitter and image movements.

Samples always needed, attach to posts, don't make us ask.

Hash marks are lines on a football field. (I'm amused because you learn this on day 1 in marching band. )
# is a pound symbol ... or hashtag if you're a Twitter junkie.

Comets are magnetic dropouts. A sample would confirm that you're using the right terms.

I'm not as worried about RGB vs. YUV at this juncture. One error at a time. But we do need/want lossless, never uncompressed.

If the image from yesterday is not cropped, full VHS image, then I'd wonder if this is a case where the VCR, or something else, is muddying the image. The details are smushed, lost. Or maybe it's nth gen? Or both.

When my video system is reed back up, I'll give this another look.

- Did my advice help you? Then become a Premium Member and support this site.
- Find television shows, cartoons, DVDs and Blu-ray releases at the TVPast forums.
Reply With Quote
  #14  
01-18-2019, 08:04 AM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
As far as I can gather from previous posts and threads of the same or similar event, the samples are caps from nth-generation (unspecified) tapes.
Reply With Quote
  #15  
01-18-2019, 07:11 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
Quote:
You can reduce cyan saturation overall, but it affects other areas (for instance, it will turn the blue skirt and blue uniforms toward black, which you don't want).
Yes, I did notice that while playing around in Colormill and a few other filters.

Quote:
Advanced color controls like those in AfterEffects, ColorFinesse, etc., allow you to reduce saturation of specific colors only in specific ranges rather than across the full spectrum. Reducing saturation isn't the same thing as changing hue. While those pricey apps can't perform repairs the way Avisynth can, they do have advantges in other areas. Unfortunately those apps cost plenty, beyond the means of most of us.
Yes, noticed very pricey. I do see Adobe offers a monthly plan? Looks like that ColorFinesse is to buy outright. Heck, even if I did have those I'd really need to learn all about how to use the "ranges" stuff. I see a lot of cool software with photography and video using these color wheels/diagrams for color correction but I'd be so lost. Would really like to learn though.


Quote:
Samples always needed, attach to posts, don't make us ask.

Hash marks are lines on a football field. (I'm amused because you learn this on day 1 in marching band. )
# is a pound symbol ... or hashtag if you're a Twitter junkie.
Post #7 does have my correct sample. The first sample from post #5 was incorrect. I've been away from this for awhile as life interrupts .. but things are slowly coming back into focus. YES! Day 1 you are correct .. Hash marks ARE lines on the football field! Those other terms are secondary!

Quote:
As far as I can gather from previous posts and threads of the same or similar event, the samples are caps from nth-generation (unspecified) tapes.
Correct, these are 3rd generation copies. Ironically they have been inside in my clothes closet in a box all these years with other videos so I wouldnt think the weather aspect should have affected them too much, I think just the process they were copied years ago (plus yes, age to a degree .. or more?)
Reply With Quote
  #16  
01-19-2019, 01:46 AM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
Quote:
Juan:"And what do you call this symbol ' # '?"
Peter:"It has many names, it's often called the pound sign in America, but everywhere else, ' # ' is called hash."

- http://www.blairenglish.com/exercise...s_names_1.html



"List of character, symbol & sign names"
https://www.prepressure.com/fonts/ba...haracter-names





Where do you guys think the expression "hash-tag" came from?
Hint: it didn't come from football.

--
--


Attached Images
File Type: png hash.png (6.8 KB, 24 downloads)
File Type: png hash 2.png (11.8 KB, 24 downloads)
Reply With Quote
  #17  
01-19-2019, 04:55 AM
lordsmurf's Avatar
lordsmurf lordsmurf is online now
Site Staff | Video
 
Join Date: Dec 2002
Posts: 8,475
Thanked 1,392 Times in 1,224 Posts
Quote:
Originally Posted by sanlyn View Post
Where do you guys think the expression "hash-tag" came from?
Hint: it didn't come from football.
Touche! Point goes to sanlyn.

- Did my advice help you? Then become a Premium Member and support this site.
- Find television shows, cartoons, DVDs and Blu-ray releases at the TVPast forums.
Reply With Quote
  #18  
01-19-2019, 10:46 AM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
Quote:
Touche! Point goes to sanlyn.
Aha.....
Reply With Quote
  #19  
01-20-2019, 07:36 PM
sirbyron sirbyron is offline
Premium Member
 
Join Date: Jun 2012
Location: NC, USA
Posts: 46
Thanked 0 Times in 0 Posts
Ok. I might get flamed here but I must ask.

I noticed I have been capturing at 640x480 on my ATI 7500 card. I have been reading back posts saying that 720x480 gives more data to work with. So, I guess I have several questions.
1) Should I go back and recapture this video in 720x480?
2) Will there be any chance of improvement (considering how bad the source is)?
3) Will these same two scripts still work? I am assuming "no". If not, how much trouble to tweek for the 720x480 capture?

btw..Step one Script took about 18 hrs to finish on my P4 machine. lol. Dont mind though, looks great. Thanks!
Reply With Quote
  #20  
01-20-2019, 08:55 PM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,446
Thanked 1,074 Times in 895 Posts
The capture size depends on what your final objective is. If you want DVD or standard def Bluray, you're better off with 720x480. It can always be downsized slightly to a 4:3 square pixel format (it's horizontal-only downscaling) for web mounting or such. If internet or non-anamorhic final formats are desired, 640x480 is preferable, although remember that if you want readily shareable formats such as DVD or BD you will have to upscale to get there, and upscaling is a bit more demanding than downscaling because upscaling makes fewer pixels do more work.

Yes, some filters have to do a lot of work on badly damaged video and will be very slow. Another filter that does some heavy-duty cleaning and can cause heavy-duty artifacts along the way (with some sources), is a median averaging filter such as FixRipsp2. In tried it with this video. If RemoveSpotsMC3 was running at a mere 2.6 fps on my system, FixRipsP2 ran at just under 1 fps. I was glad MC3 worked.
Reply With Quote
Reply




Similar Threads
Thread Thread Starter Forum Replies Last Post
Problem with custom script in Avisynth to stabilize video? benzio Restore, Filter, Improve Quality 0 03-07-2018 11:27 AM
VCR simulator script, Avisynth fake VHS look! jmac698 Restore, Filter, Improve Quality 1 02-12-2018 06:17 AM
Avisynth script to fix offset interlacing? lordsmurf Restore, Filter, Improve Quality 4 02-10-2018 05:34 PM
Avisynth script for QTGMC, encode x264 - no video after deinterlace? mo418 Capture, Record, Transfer 18 09-16-2015 01:06 PM

Thread Tools



 
All times are GMT -5. The time now is 07:51 PM