Help improving some old PAL TV recordings.

[Nirvash]

Hello! I've been lurking the forums all day trying to take in as much info as possible, and I've been messing with Hybrid with great success, though the program is daunting with the sheer amount of options it has.

Sample source is an old PAL TV recording recorded using a period accurate video recorder that recorded to mpeg-2 at 720x576 @ ~9k bitrate and its only input option is RCA ().

General
Complete name                  : D:\VideoTest\Medallas\20120811_2319_Medallas_Fem.mpg
Format                         : MPEG-PS
File size                      : 471 MiB
Duration                       : 7 min 5 s
Overall bit rate mode          : Variable
Overall bit rate               : 9 271 kb/s
Frame rate                     : 25.000 FPS

Video
ID                             : 224 (0xE0)
Format                         : MPEG Video
Format version                 : Version 2
Format profile                 : Main@Main
Format settings                : BVOP
Format settings, BVOP          : Yes
Format settings, Matrix        : Default
Format settings, GOP           : M=3, N=12
Format settings, picture struc : Frame
Duration                       : 7 min 5 s
Bit rate mode                  : Variable
Bit rate                       : 8 830 kb/s
Maximum bit rate               : 9 800 kb/s
Width                          : 720 pixels
Height                         : 576 pixels
Display aspect ratio           : 4:3
Frame rate                     : 25.000 FPS
Standard                       : PAL
Color space                    : YUV
Chroma subsampling             : 4:2:0
Bit depth                      : 8 bits
Scan type                      : Interlaced
Scan order                     : Top Field First
Compression mode               : Lossy
Bits/(Pixel*Frame)             : 0.852
Time code of first frame       : 00:00:00:00
Time code source               : Group of pictures header
GOP, Open/Closed               : Open
GOP, Open/Closed of first fram : Closed
Stream size                    : 448 MiB (95%)
Color primaries                : BT.601 PAL
Transfer characteristics       : BT.470 System B/G
Matrix coefficients            : BT.470 System B/G

Audio
ID                             : 192 (0xC0)
Format                         : MPEG Audio
Format version                 : Version 1
Format profile                 : Layer 2
Duration                       : 7 min 5 s
Bit rate mode                  : Constant
Bit rate                       : 256 kb/s
Channel(s)                     : 2 channels
Sampling rate                  : 48.0 kHz
Frame rate                     : 41.667 FPS (1152 SPF)
Compression mode               : Lossy
Delay relative to video        : -17 ms
Stream size                    : 13.0 MiB (3%)

Menu
Format                         : DVD-Video

So far I've been using x264 and NVenc (to see what was faster but it's about the same, 60fps), cropping black bars, upscaling to 1080p and using QTGMC and Vapoursynth to deinterlace and apply some filters and the quality improvement is noticeable.

I didn't want to alter the source, datacap users we warned:
Original file (470 MB): https://www.mediafire.com/file/ptztxw1bo0a4tkc/20120811_2319_Medallas_Fem.mpg/file

Output (220 MB): https://www.mediafire.com/file/ptvstxjeb5h7dpv/20120811_2319_Medallas_Fem_hybrid_GPU-n...arpen.mp4/file

Settings used! If not in the screenshot, they're the default.

Do you think I can improve it in any way? There are thousands of possible combinations between the different filters, without counting all the individual options and I don't even know where to start.

BTW I'm not using Filtering - Vapoursynth - Frame - Resize - NNEDI3 as Selur recommended in some post because I saw no gain in quality taking x6 time to complete.

Thanks for your help!!

[themaster1]

Derainbow & dot crawl remover is what you need

MPEG2Source("20120811_2319_Medallas_Fem.d2v", cpu=0)

# 1: REMOVE RAINBOWS
ConvertToYV16(interlaced=true) #
orig=last
ev=orig.assumetff().separatefields().selecteven()
od=orig.assumetff().separatefields().selectodd()
ev
ue_chroma = UToY(ev).blur(0,1.5).ttempsmooth(maxr=6,lthresh=15 0, strength=2)
ve_chroma = VToY(ev).blur(0,1.5).ttempsmooth(maxr=6,lthresh=15 0, strength=2)
YToUV(ue_chroma, ve_chroma)
MergeLuma(ev)
ev_filtered=last
od
uo_chroma = UToY(od).blur(0,1.5).ttempsmooth(maxr=6,lthresh=15 0, strength=4)
vo_chroma = VToY(od).blur(0,1.5).ttempsmooth(maxr=6,lthresh=15 0, strength=4)
YToUV(uo_chroma, vo_chroma)
MergeLuma(od)
od_filtered=last
interleave(ev_filtered,od_filtered)
assumefieldbased().assumetff().weave()
converttoyv12(interlaced=false)

#2: DEINTERLACE
QTGMC( Preset="fast", SourceMatch=3, Lossless=1, Sharpness=0.4, TR2=2, EdiThreads=4)
Selecteven()

#3: REMOVE DOT CRAWL (PROGRESSIVE ONLY)
main=last
main
#checkmate (thr=12,max=12, tthr2=0) #
bicubicresize(480,360) #
spline36resize(720,576)
lsfmod
filtered=last
main
mt_edge().mt_expand()
mymask=last
mt_merge(main, filtered, mymask)
#

[Selur]

assumefieldbased().assumetff().weave()
converttoyv12(interlaced=false)

shouldn't this be 'converttoyv12(interlaced=true)' ?

@Nirvash: general suggestions, looking at the setting (haven't looked at the source) you might want to set output PAR to 1:1 when going for HD and higher resolutions and probably also change the color matrix.

[Nirvash]

@Selur

In hindsight I should have added some screenshots comparing the source and the output I have so far so you don't have to download the whole video. Here they are:


[Attachment 83085 - Click to enlarge]

[Attachment 83081 - Click to enlarge]
This is much, much cleaner, but of course the faces remain garbled because the original is garbage.

[Attachment 83083 - Click to enlarge]
Again, overall pretty clean and like the faces, the buttons on the jacket were garbage and will always look like garbage.

@10themaster1 Thanks! So should I disable the other filters (de-noise, de-halo, de-ring and sharpen)?
Also, how do I use your code? Do I just paste that into vsViewer? And how do I generate that code with my own settings?

This code Hybrid gave me doesn't look like your code.

# Imports
import vapoursynth as vs
# getting Vapoursynth core
import ctypes
import sys
import os
core = vs.core
# Import scripts folder
scriptPath = 'C:/Program Files/Hybrid/64bit/vsscripts'
sys.path.insert(0, os.path.abspath(scriptPath))
# Loading Support Files
Dllref = ctypes.windll.LoadLibrary("C:/Program Files/Hybrid/64bit/vsfilters/Support/libfftw3f-3.dll")
# loading plugins
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DenoiseFilter/CTMF/CTMF.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DenoiseFilter/VagueDenoiser/libneovaguedenoiser.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/GrainFilter/RemoveGrain/RemoveGrainVS.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/GrainFilter/AddGrain/AddGrain.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DenoiseFilter/DFTTest/DFTTest.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DenoiseFilter/NEO_FFT3DFilter/neo-fft3d.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/EEDI3m.dll")# vsQTGMC
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/ResizeFilter/nnedi3/vsznedi3.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/libmvtools.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/scenechange.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/MiscFilter/MiscFilters/MiscFilters.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DeinterlaceFilter/Bwdif/Bwdif.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/SourceFilter/d2vSource/DGDecode.dll")
# Import scripts
import mvsfunc
import G41Fun
import havsfunc
import validate
# Source: 'D:\VideoTest\Medallas\20120811_2319_Medallas_Fem.mpg'
# Current color space: YUV420P8, bit depth: 8, resolution: 684x380, frame rate: 25fps, scanorder: top field first, yuv luminance scale: limited, matrix: 470bg, transfer: bt.470 system b/g, primaries: bt.601 pal, format: mpeg-2
# Loading D:\VideoTest\Medallas\20120811_2319_Medallas_Fem.mpg using DGDecode
clip = core.dgdecode.MPEG2Source("B:/Temp/mpg_5f820ce2b0c0fd990c6f0c5430b365c6_853323747.d2v",info=3)# 25 fps, scanorder: top field first
frame = clip.get_frame(0)
# setting color matrix to 470bg.
clip = core.std.SetFrameProps(clip, _Matrix=vs.MATRIX_BT470_BG)
# setting color transfer (vs.TRANSFER_BT470_BG), if it is not set.
if validate.transferIsInvalid(clip):
  clip = core.std.SetFrameProps(clip=clip, _Transfer=vs.TRANSFER_BT470_BG)
# setting color primaries info (to vs.PRIMARIES_BT470_BG), if it is not set.
if validate.primariesIsInvalid(clip):
  clip = core.std.SetFrameProps(clip=clip, _Primaries=vs.PRIMARIES_BT470_BG)
# setting color range to TV (limited) range.
clip = core.std.SetFrameProps(clip=clip, _ColorRange=vs.RANGE_LIMITED)
# making sure frame rate is set to 25fps
clip = core.std.AssumeFPS(clip=clip, fpsnum=25, fpsden=1)
# making sure the detected scan type is set (detected: top field first)
clip = core.std.SetFrameProps(clip=clip, _FieldBased=vs.FIELD_TOP) # tff
# Deinterlacing using QTGMC
clip = havsfunc.QTGMC(Input=clip, Preset="Placebo", TFF=True) # new fps: 50
# Making sure content is preceived as frame based
clip = core.std.SetFrameProps(clip=clip, _FieldBased=vs.FIELD_PROGRESSIVE) # progressive
clip = core.std.Crop(clip=clip, left=18, right=16, top=88, bottom=82)# cropping to 650x210
# rainbow removal using LUTDeRainbow
clip = havsfunc.LUTDeRainbow(input=clip)
# denoising using VagueDenoiser
clip = core.neo_vd.VagueDenoiser(clip=clip)
# sharpening using FineSharp
clip = G41Fun.FineSharp(clip=clip)
# applying dehalo using DeHalo_alpha
clip = havsfunc.DeHalo_alpha(clip)
# deringing using HQDeringmod
clip = havsfunc.HQDeringmod(clip, nrmode=2, darkthr=3.0)
# Resizing using 6 - luma bicubic, chroma bilinear
clip = core.fmtc.resample(clip=clip, kernel="bilinear", w=3344, h=1080, interlaced=False, interlacedd=False) # resolution 3344x1080 before YUV420P8 after YUV420P16
# adjusting output color from: YUV420P16 to YUV420P8 for x264Model
clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P8, range_s="limited", dither_type="error_diffusion")
# set output frame rate to 50fps (progressive)
clip = core.std.AssumeFPS(clip=clip, fpsnum=50, fpsden=1)
# output
clip.set_output()

[Selur]

About your settings:
a. QTGMC with Placebo seems overkill
b. about the resizer: I would recommend using another one
c. you might want to add some deblocking
d. your resizing seems wrong
side note: there a also some machine learning based addons for Hybrid, if you have the GPU for it