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. |
28th March 2015, 18:00 | #1 | Link |
Registered User
Join Date: Jul 2013
Posts: 17
|
Deinterlacing Blu-ray (Anime)
I'm having a really hard time deinterlacing this anime, I tried using things from ITVC to QTGMC but the lines would sometimes look weird.
Here are some samples (I used tsmuxer to split them): https://www.dropbox.com/sh/8qal0ylsz...rYZ-9dMba?dl=0 |
28th March 2015, 21:02 | #2 | Link |
Registered User
Join Date: Sep 2012
Posts: 366
|
Code:
FFVideoSource("EP1.split.1.m2ts").assumefps("ntsc_video") separatefields() #QTGMC() Dup(Threshold=6.0, blend=true) tdecimate(mode=2) NNEDI3_Resize16(1920, 1080) #or not Last edited by ndjamena; 29th March 2015 at 05:03. |
29th March 2015, 10:58 | #4 | Link |
Registered User
Join Date: Sep 2012
Posts: 366
|
The fields are slightly different. I think at least some of the fields were upscaled from SD to HD without being deinterlaced. I can't really tell, maybe it all has. I've being testing out Jagabo's scripts from this VideoHelp Thread: http://forum.videohelp.com/threads/3...cing-and-edges but it gives terrible aliasing and generally looks like crap. The frame rate in this one is even further off, there's fields lost in editing and I'm pretty sure some of it is full NTSC frame rate. -edit- there's no setting in Dup that will merge the fields without destroying frames it shouldn't, I tried fiddling with QTGMC and ReDup to see if I could get something working but after being deinterlaced some of the derived top/bottom field frames have more differences to each other than they do against some of their original frame neighbours. Last edited by ndjamena; 29th March 2015 at 11:03. |
3rd April 2015, 09:32 | #5 | Link | |
Registered User
Join Date: Jul 2013
Posts: 17
|
Quote:
https://www.dropbox.com/s/2y571y7yknkp396/Try1.mkv?dl=1 Also the subtitles: https://www.dropbox.com/s/zz83l7nw5ojo4r0/Try2.mkv?dl=1 Your script probably gave me the worst result, the lines just looked terrible. |
|
3rd April 2015, 11:29 | #6 | Link |
Registered User
Join Date: Jan 2012
Location: Mesopotamia
Posts: 2,587
|
maybe this will help http://forum.doom9.org/showthread.php?t=158696
__________________
See My Avisynth Stuff |
3rd April 2015, 13:27 | #7 | Link |
Registered User
Join Date: Nov 2006
Posts: 83
|
I'm not sure if this will work:
Code:
orig = last _deint = tdeint(emask=orig.TMM(),slow=2) #also try without the emask parameter tfm(slow=2,clip2=_deint).tdecimate(mode=1) Code:
#1st pass tfm(slow=2,pp=1,output="tfm_file.txt") Code:
#2nd pass orig = last tfm(slow=2,pp=1,input="tfm_file.txt") tdeint(hints=true,clip2=orig,slow=2) tdecimate(exPP=1,mode=1,tfmin="tfm_file.txt") |
3rd April 2015, 14:55 | #8 | Link |
Anime addict
Join Date: Feb 2009
Location: Spain
Posts: 673
|
My attempt:
Code:
LoadPlugin("C:\Portables\MeGUI\tools\lsmash\LSMASHSource.dll") LWLibavVideoSource("C:\Users\Isra\Desktop\EP1.split.15.m2ts") complementparity() separatefields() supaa2() assumefieldbased.weave() qtgmc(preset="fast", fpsdivisor=2, lossless=2, sourcematch=2)
__________________
Intel i7-6700K + Noctua NH-D15 + Z170A XPower G. Titanium + Kingston HyperX Savage DDR4 2x8GB + Radeon RX580 8GB DDR5 + ADATA SX8200 Pro 1 TB + Antec EDG750 80 Plus Gold Mod + Corsair 780T Graphite |
3rd April 2015, 20:20 | #10 | Link |
Registered User
Join Date: Oct 2014
Posts: 476
|
Some idea of what file you're working on would be useful. "Try2" tells me nothing.
Moon runes can be a problem because it detects them as combed. Try turning off postprocessing. tfm(pp=0).tdecimate(mode=1,hybrid=0) You can play around with santiag for the residual combing. A deinterlacer is the wrong tool here. This is just telecined. Last edited by kuchikirukia; 4th April 2015 at 02:49. |
3rd April 2015, 23:03 | #11 | Link | |
Registered User
Join Date: Nov 2004
Location: Poland
Posts: 2,843
|
Quote:
Resize it to half resolution vertically and upscale back with nnedi3. |
|
4th April 2015, 06:28 | #12 | Link | |
Registered User
Join Date: Dec 2011
Posts: 354
|
Quote:
|
|
4th April 2015, 06:37 | #13 | Link |
Registered User
Join Date: Jul 2013
Posts: 17
|
I'll upload it now. It's from another episode, I just wanted to test out to see if it works with other episodes.
Not sure if this info would be useful: Code:
General ID : 0 (0x0) Complete name : D:\Videos\Gintama\FullDisc\GINTAMA_BD_BOX1_1\BDMV\STREAM\00001.m2ts Format : BDAV Format/Info : Blu-ray Video File size : 5.29 GiB Duration : 24mn 55s Overall bit rate mode : Variable Overall bit rate : 30.4 Mbps Maximum Overall bit rate : 48.0 Mbps Video ID : 4113 (0x1011) Menu ID : 1 (0x1) Format : AVC Format/Info : Advanced Video Codec Format profile : High@L4.1 Format settings, CABAC : Yes Format settings, ReFrames : 4 frames Codec ID : 27 Duration : 24mn 56s Bit rate mode : Variable Bit rate : 27.7 Mbps Maximum bit rate : 40.0 Mbps Width : 1 920 pixels Height : 1 080 pixels Display aspect ratio : 16:9 Frame rate : 29.970 fps Color space : YUV Chroma subsampling : 4:2:0 Bit depth : 8 bits Scan type : MBAFF Bits/(Pixel*Frame) : 0.445 Stream size : 4.82 GiB (91%) Audio ID : 4352 (0x1100) Menu ID : 1 (0x1) Format : PCM Format settings, Endianness : Big Format settings, Sign : Signed Muxing mode : Blu-ray Codec ID : 128 Duration : 24mn 56s Bit rate mode : Constant Bit rate : 1 536 Kbps Channel(s) : 2 channels Channel positions : Front: L R Sampling rate : 48.0 KHz Bit depth : 16 bits Stream size : 274 MiB (5%) Text ID : 4608 (0x1200) Menu ID : 1 (0x1) Format : PGS Codec ID : 144 Duration : 24mn 53s Delay relative to video : 1s 502ms Last edited by Taizenshū; 4th April 2015 at 08:25. |
4th April 2015, 11:54 | #14 | Link |
Registered User
Join Date: Sep 2012
Posts: 366
|
No, it's not. None of us are miracle workers and your source is borked. Anyone assuming it's simply telecined or hybrid isn't paying attention and any script that doesn't involve either blending or downsizing is wasting your time.
If I could finish ReDup blending may be an option, but since that's not happening you have no choice but to downsize or live with the messed up lines. |
4th April 2015, 17:43 | #16 | Link | ||
Registered User
Join Date: Jul 2013
Posts: 17
|
Quote:
Quote:
|
||
6th April 2015, 08:24 | #18 | Link |
͡҉҉ ̵̡̢̛̗̘̙̜̝̞̟̠͇̊̋̌̍̎̏̿̿
Join Date: Feb 2009
Location: No support in PM
Posts: 712
|
These are typical artefacts from an Interlaced SD master → Imperfect IVTC/Deinterlacing → Upscaling → Interlacing/Pulldown process. The problem is that a “smart” deinterlacer is used, trying to detect which parts of the picture should be field-matched and which parts should be deinterlaced. Sometimes the detection fails and the wrong processing is applied.
The best thing to do is to IVTC without caring about the residual combing (just TFM/TDecimate with basic parameters) then reduce the resolution to something close to the original so the combing is removed, leaving just slight ghosting, or run DAA3 on the result if you want too keep the upscaled resolution.
__________________
dither 1.28.1 for AviSynth | avstp 1.0.4 for AviSynth development | fmtconv r30 for Vapoursynth & Avs+ | trimx264opt segmented encoding |
Thread Tools | Search this Thread |
Display Modes | |
|
|