Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
|
|
Thread Tools | Search this Thread | Display Modes |
25th February 2012, 14:00 | #81 | Link |
Registered User
Join Date: Feb 2003
Location: Russia, Moscow
Posts: 854
|
Hi All!
Please advice how calculate radius for Gaussian blur based on variance. Code:
radius=3*sqrt(Var)+0.5 Code:
radius=2*sqrt(Var) yup. Last edited by yup; 27th February 2012 at 08:01. |
27th February 2012, 09:36 | #82 | Link |
Registered User
Join Date: Feb 2003
Location: Russia, Moscow
Posts: 854
|
I find this relation between radius and sigma
http://en.wikipedia.org/wiki/Talk%3AGaussian_blur yup. |
29th March 2012, 16:56 | #83 | Link |
Registered User
Join Date: Feb 2003
Location: Russia, Moscow
Posts: 854
|
Hi all!
When I try update FFTW to 3.3.1 from 3.2.2. I get error with Avisynth 2.6 Alpha 3 build and SeT last build when call unsharp. I tested with new FFTW fft3dfilter, MVTools and dfttest (not need rename dll) without problem. only variableblur 0.5 (Tritical release) give crash at VirtualDub. yup. Last edited by yup; 30th March 2012 at 11:57. Reason: Add new info |
31st March 2012, 04:24 | #84 | Link |
Registered User
Join Date: Feb 2003
Location: Russia, Moscow
Posts: 854
|
Hi All!
Before I tested Tritical build, now last neuron2 build situation the same. Please help. I use unsharp function for local contrast enhancement. New fft3w give some speed improvement, but impossible use with variableblur. yup. |
23rd May 2012, 05:58 | #86 | Link |
Registered User
Join Date: Dec 2003
Location: MO, US
Posts: 999
|
I updated my modified version of variableblur. Link is same as before. Changes:
Code:
0.6 Fix border=3 bug for gfunc/gfuncc >= 0 (gamma processing was not done on border values) Make radius for gaussianblur 4*std instead of 3*std Add border=4 (mirroring) option to gaussianblur and make it the default Dynamically load fftw (libfftwf3-3.dll) instead of statically linking to old fftw3.lib Multithread gaussianblur using OpenMP and use FFTW multithreaded plans SSE/SSE2 optimizations. Gaussianblur now requires SSE2 capable cpu. |
24th May 2012, 22:13 | #88 | Link |
Registered User
Join Date: Dec 2003
Location: MO, US
Posts: 999
|
Another update. Same link as before. Again changes are for gaussianblur/unsharp.
Code:
0.7 Fix integrate=true giving incorrect results due to lack of precision in weight calculation. Now gives results much closer to integrate=false, as would be expected. Add support for YUY2, RGB24, and RGB32 input to gaussianblur/unsharp Last edited by tritical; 24th May 2012 at 22:16. |
16th November 2012, 07:21 | #91 | Link | |
Registered User
Join Date: Aug 2008
Location: Isle of Man
Posts: 588
|
Quote:
|
|
24th August 2016, 10:28 | #93 | Link |
Registered User
Join Date: Feb 2003
Location: Russia, Moscow
Posts: 854
|
Hi all!
Now I continue use this plugin mainly for sharpening.Now I am switch to aisynth+ 64 bit and can not find 64 bit plugin for sharpening. IMO unsharp mask make this work better than other for my source (VHS capture). May be possible compile this plugin for 64 bit? Or advice sharpening for 64 bit (I see StaxRip bundle and do not satisfy, for my source). yup. |
24th August 2016, 11:00 | #94 | Link |
Registered User
Join Date: Mar 2015
Posts: 775
|
maybe you could use my VD plugin (link)
Dont know how it compares to binormal implementation. There is both blur and USM. Compiled x64 binary is inside the vd packs.
__________________
VirtualDub2 |
24th August 2016, 11:52 | #96 | Link |
Registered User
Join Date: Mar 2015
Posts: 775
|
x64 pack is called VirtualDub64_pack_###.zip
__________________
VirtualDub2 |
Thread Tools | Search this Thread |
Display Modes | |
|
|