Hqdn3dY
From Avisynth wiki
Abstract | |
---|---|
Author | Loren Merritt, Andrew Revvo |
Version | v1.0.4 |
Download | Hqdn3dY-1.0.4.7z |
Category | Spatial-Temporal Denoisers |
License | GPLv2 |
Discussion |
Contents |
Description
hqdn3dY is a modfication of the hqdn3d filter.
Changes include:
- Chroma processing disabled.
- Parameters ls and lt were renamed to sp and tp, respectively.
- Parameters cs, ct, and restart were removed.
-
hqdn3d16Y
16-bit processing using the interleaved 16bit hack aka Double Width (similar hack to Stack16).
Requirements
- [x86] AviSynth+ or AviSynth 2.6 or greater
- [x64] AviSynth+
- Progressive input only
- Supported color formats: Y8
Syntax and Parameters
- hqdn3dY (clip, float "sp", float "tp")
- hqdn3d16Y (clip, float "sp", float "tp")
- clip =
- Input clip.
- clip =
- float sp = 4.0
- sp (0-255) is the strength of the spatial filtering on the luma channel. Increasing this value will improve the smoothing but may overblur; anything above about 10 is probably not a good idea.
- float sp = 4.0
- float tp = 6.0
- tp (0-255) is the strength of the temporal filtering on the luma channel. Increasing this value will improve the smoothing but may cause ghosting; anything above about 13 is probably not a good idea.
- float tp = 6.0
Examples
Hqdn3dY with all default values:
AviSource("Blah.avi") Hdn3dY(sp=4.0, tp=6.0)
16 bit processing:
ConvertTo16bit() # if your source is not already 16-bit ConvertToDoubleWidth() hqdn3d16Y (sp=4.0, tp=6.0) ConvertFromDoubleWidth(bits=16)
To denoise luma and chroma (YUV):
y = ConvertToY8() u = UToY8() v = VToY8() y = y.Hqdn3dY(sp=10.0, tp=0.0) u = u.Hqdn3dY(sp=10.0, tp=0.0) v = v.Hqdn3dY(sp=10.0, tp=0.0) YToUV(u,v,y)
Or to denoise RGB:
r = ShowRed("Y8") g = ShowGreen("Y8") b = ShowBlue("Y8") r = r.Hqdn3dY(sp=10.0, tp=0.0) g = g.Hqdn3dY(sp=10.0, tp=0.0) b = b.Hqdn3dY(sp=10.0, tp=0.0) MergeRGB(r,g,b)
You can view the result of an individual channel and tweak it by entering a channel clip variable (y,u,v,r,g,b) after this code.
Changelog
Version Date Changes
v1.0.4 2018/01/09 - Compiled with VC++ 2017.5. - Minor improvements. v1.0.3 2017/09/27 - Compiled with VC++ 2017.3. - Video Artifact paths. v1.0.2 2016/11/14 - Compiled with VC++ 2015.3. - Video Artifact-ready. - Batch scripts corrected. v1.0.1 2016/06/22 - Compiled with VC++ 2015 RC 3. - Paths updated. v1.0.0 2016/02/13 - Hqdn3dY Andrew Revvo: 16-bit version is added. v1.0.0b 2016/02/04 - Hqdn3dY Andrew Revvo: Usage instruction is added. v1.0.0a 2016/01/14 - Hqdn3dY - Removed U,V code. - Refactoring. v0.11 2005/01/25 - Fix a floating-point exception. v0.10 2004/10/25 - Initial release.
Archived Downloads
Version | Download | Mirror |
---|---|---|
v1.0.4 | Hqdn3dY-1.0.4.7z | |
v1.0.1 | Hqdn3dY.7z | Hqdn3dY-1.0.1.7z |
External Links
Back to External Filters ←