This forum uses cookies
This forum makes use of cookies to store your login information if you are registered, and your last visit if you are not. Cookies are small text documents stored on your computer; the cookies set by this forum can only be used on this website and pose no security risk. Cookies on this forum also track the specific topics you have read and when you last read them. Please confirm whether you accept or reject these cookies being set.

A cookie will be stored in your browser regardless of choice to prevent you being asked this question again. You will be able to change your cookie settings at any time using the link in the footer.

[HELP] Any ideas on how to improve the quality of this video?
#1
I acquired a PAL VHS from the late 1980s shot with an amateur video camera. Also in the process the tape also broke, I know I don't have a good starting point....

Here is the video acquired before deinterlacing it with Hybrid:
[Image: pre.png]

Here it is after deinterlacing it + applying some filters:
  • Upscale to 1440x1080
  • Vapoursynth QTGMC slow preset + BOB
  • Colormatrix rec601 to rec709
  • Deblock QED default value
  • TemporalDegrain2 default value
  • Denoise QTGMC (final temporal smoothing 1, EZdenoise 6,50)
  • FineDehalo default
  • Frame resizer NNEDI3 48x6, 256 neurons
 
[Image: Post.png]

Any ideas to get sharper faces or am I satisfied that I already got a good result?

Thanks!
Reply
#2
You can denoise using the QTGMC deinterlacer, no need to do it again the denoise tab.

My filters for interlaced content, in this order (you can set the filter order under Vapoursynth -> Misc -> Filter Order/Queue):

QTGMC custom preset: preset very slow, TR2=3, SourceMatch=3, Lossless=2, EZDenoise=0.50, NoisePreset="Slower", Denoiser="KNLMeansCL"
CAS .7 - .9 depending on the source
DPIRDeblock 10.0
MCDegrainSharp - default
DehaloAlpha - default
YAHR - default
Frame (see screenshot)

Also, I wouldn't convert the color scale from 601 to 709. Would leave it as is.

[Image: Capture.png]
Reply
#3
Hard to say from a still image, would need a short sample.
I would, try:
a. CAS (sharpening seems to be overexaggerated; FineSharp and 48x6 in NNEDI3 are probably at fault)
b. apply some denoising on the chroma
c. test if it makes sense to use CodeFormer (probably not)
d. SpotLess
e. some machine learning deblockers&Co

Quote:CAS .7 - .9 depending on the source
I usually try to stick below 0.7 Smile

Cu Selur

Ps.: Moved the thread to 'Small Talk' Smile
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply
#4
(23.10.2024, 05:27)Selur Wrote: I usually try to stick below 0.7 

I'm doing a ton of snowboard videos and I find that the higher CAS gives the snow a little more detail and character  Cool
Reply
#5
If, it works for your content, it's fine by me. Smile
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)