Quantcast Avisynth: Motion Adaptive Filtering Now Possible? - Page 19 - digitalFAQ.com Forums [Archives]
  #361  
06-05-2003, 02:52 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by audioslave
If a movie contains a lot of fast action scenes will it be easier to fit on a single disc?
No
If it's an action movie, you only have two choices: Either use a lower resolution and put it on one disk, or go for KVCDx3 (or even 704x480(576) ) but target for two CD's. With the new script and techniques, you should be able to put just about any movie on two CDs. No need for three

-kwag
Reply With Quote
Someday, 12:01 PM
admin's Avatar
Site Staff / Ad Manager
 
Join Date: Dec 2002
Posts: 42
Thanks: ∞
Thanked 42 Times in 42 Posts
  #362  
06-05-2003, 03:01 AM
audioslave audioslave is offline
Free Member
 
Join Date: Mar 2003
Location: Sweden
Posts: 725
Thanks: 0
Thanked 0 Times in 0 Posts
OK, that's what I guessed... I did just that - targeted the movie for two discs and got a CQ value around 63 (I don't remember exactly)... But hey, to fit a movie onto two discs ain't too shabby. Especially when it look almost identical to the DVD!

That clip kwag, that clip... AWESOME!!!
I guess I'll have to start all over with my encode as soon as I get off work. It'll be SO worth it!
I'm amazed at how fast things are moving and changing on this forum. Who would have guessed that a VCD could look this good - ever!
Reply With Quote
  #363  
06-05-2003, 03:03 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by audioslave
Who would have guessed that a VCD could look this good - ever!
VCD What's that
Reply With Quote
  #364  
06-05-2003, 03:04 AM
audioslave audioslave is offline
Free Member
 
Join Date: Mar 2003
Location: Sweden
Posts: 725
Thanks: 0
Thanked 0 Times in 0 Posts
Sorry if I offended you kwag! What I meant was KVCD - of course!
Reply With Quote
  #365  
06-05-2003, 03:06 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by audioslave
Sorry if I offended you kwag! What I meant was KVCD - of course!
Just kidding
Reply With Quote
  #366  
06-05-2003, 03:09 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Makes me wonder, how much time we can really get now with KDVD and this script, and also KVCD ULBR (or LBR )

-kwag
Reply With Quote
  #367  
06-05-2003, 03:12 AM
audioslave audioslave is offline
Free Member
 
Join Date: Mar 2003
Location: Sweden
Posts: 725
Thanks: 0
Thanked 0 Times in 0 Posts
@kwag
Have you slept at all last night? You must have been up testing and tweaking, right? I went to bed at 2 p.m. and when I logged on this morning I saw the goodies! Great!
Reply With Quote
  #368  
06-05-2003, 03:32 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by audioslave
@kwag
Have you slept at all last night? You must have been up testing and tweaking, right? I went to bed at 2 p.m. and when I logged on this morning I saw the goodies! Great!
I really didn't get that much sleep last night. So here I am, in bed typing in my iBook via wireless 802.11b network
And now I'm going to bed. Let the scripts rest (for some centons)

Have fun!
-kwag
Reply With Quote
  #369  
06-05-2003, 03:36 AM
audioslave audioslave is offline
Free Member
 
Join Date: Mar 2003
Location: Sweden
Posts: 725
Thanks: 0
Thanked 0 Times in 0 Posts
Sweet dreams, kwag!
Reply With Quote
  #370  
06-05-2003, 06:27 AM
jorel jorel is offline
Invalid Email / Banned / Spammer
 
Join Date: Aug 2002
Location: Brasil - MG - third stone from the sun
Posts: 5,570
Thanks: 0
Thanked 0 Times in 0 Posts


Kwag my friend,
something is wrong with my sample in the new script.
( June 5, 2003 @05:45GMT )
i got CQ70,297 in 480x480 but the image is very "smoked"(blurred)
and with big staircases in thin lines....
with 704x480 the CQ is too low.


"sniff"
Reply With Quote
  #371  
06-05-2003, 07:32 AM
mfb mfb is offline
Free Member
 
Join Date: Oct 2002
Location: Austria
Posts: 64
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by jorel


Kwag my friend,
something is wrong with my sample in the new script.
( June 5, 2003 @05:45GMT )
i got CQ70,297 in 480x480 but the image is very "smoked"(blurred)
and with big staircases in thin lines....
with 704x480 the CQ is too low.


"sniff"
Hi Jorel,

strange...I did a view samples on 5 different movies and different
resolutions, and they all came out sharp as a knife (maybe almost a little
to too sharp for my eyes)

and hey Jorel don't "sniff", I'am sure the master himself will lend you
a helping hand...

regards, ***mfb***
Reply With Quote
  #372  
06-05-2003, 10:48 AM
girv girv is offline
Free Member
 
Join Date: Sep 2002
Posts: 108
Thanks: 0
Thanked 0 Times in 0 Posts
Send a message via ICQ to girv
Quote:
Originally Posted by asharp 0.95 documents
asharp(float "T", float "D", float "B", bool "hqbf")

T : unsharp masking threshold. 0 will do nothing. (value clamped to [nothing=0..32])
No mention of negative values being blurs!

Using Subtract, I saw that T < 0 blurred more until T=-1 but then made no further differences no matter how negative I made it.

Is it just luck / bugs that make T<0 blur the image?
Reply With Quote
  #373  
06-05-2003, 11:19 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by girv
Quote:
Originally Posted by asharp 0.95 documents
asharp(float "T", float "D", float "B", bool "hqbf")

T : unsharp masking threshold. 0 will do nothing. (value clamped to [nothing=0..32])
No mention of negative values being blurs!
I know I found out because I thought it would, and indeed it did
I couldn't use BlockBuster(blur) because it didn't blur enough. I also tried another filter ( forgot now which! ), but only scaled in steps of 1, so that was a no no. So asharp used "in reverse" made our day
Quote:

Using Subtract, I saw that T < 0 blurred more until T=-1 but then made no further differences no matter how negative I made it.

Is it just luck / bugs that make T<0 blur the image?
Another nice "Feature"

-kwag
Reply With Quote
  #374  
06-05-2003, 11:31 AM
girv girv is offline
Free Member
 
Join Date: Sep 2002
Posts: 108
Thanks: 0
Thanked 0 Times in 0 Posts
Send a message via ICQ to girv
Quote:
Originally Posted by kwag
Another nice "Feature"
hehe some day i really must install VC++ and write that MotionAdaptiveLumaBlurWithNoBugs() plugin
Reply With Quote
  #375  
06-05-2003, 11:34 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by girv
hehe some day i really must install VC++ and write that MotionAdaptiveLumaBlurWithNoBugs() plugin
LOL
That's a good one

-kwag
Reply With Quote
  #376  
06-05-2003, 12:29 PM
GFR GFR is offline
Free Member
 
Join Date: May 2002
Posts: 438
Thanks: 0
Thanked 0 Times in 0 Posts
While you're at it, please write a "AvoidMosquitoesAroundCartoonBorders()" filter
Reply With Quote
  #377  
06-05-2003, 12:30 PM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Just made a minimal change to the script:

Changed asharp(2,4,0.25,hqbf=true) to asharp(2,4)
Changed MergeLuma(blur(0.1)) to MergeLuma(blur(0.2))

The result is sharper picture and lower file size.
Apparently "hqbf" method is still not very well optimized.
Also speed increased 5 seconds for every minute.

-kwag
Reply With Quote
  #378  
06-05-2003, 02:49 PM
Jellygoose Jellygoose is offline
Free Member
 
Join Date: Jun 2002
Location: Germany
Posts: 1,288
Thanks: 0
Thanked 0 Times in 0 Posts
Hi kwag...

I made thousands of samples today, and the script got better constantly... still haven't tried the VERY latest script ( ) with asharp(2,4) and MergeLuma(0.2)...
However, as the picture quality gets a lot better, there's still one scene in my movie, where I think Blockbuster noise would REALLY help a lot...
It's a very dark rain scene, and DCT blocks show up, and ugly artifacts all around the fast-moving rain-drops... this doesn't seem to get better, but in case worse, the more we apply sharpening...
any suggestions?
__________________
j3llyG0053
Reply With Quote
  #379  
06-05-2003, 03:08 PM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by Jellygoose
I think Blockbuster noise would REALLY help a lot...
It's a very dark rain scene, and DCT blocks show up, and ugly artifacts all around the fast-moving rain-drops... this doesn't seem to get better, but in case worse, the more we apply sharpening...
any suggestions?
Hi Jellygoose,

Last night I played for about 2 hours, integrating BlockBuster(noise) into the script. I was able to make the variance adaptive , but at 704x480, BlockBuster really doesn't help
This was confirmed a long time ago by SansGrip, and I can confirm the same thing. Somehow, TMPEG's CQ mode can't "see" the noise correctly. You can see the noise clearly by loading the .avs in Vdub, but in TMPEG it's almost gone, and you have to use very high value for variance which then spoils the picture. So we have to find another way to deal with that

Edit: I know this shouldn't make a difference, maybe it does!, but could you code that scene with a higher MIN bit rate and see if the DCT blocks dissapear This is something I've had on on my queue list to try, but haven't got to it yet

-kwag
Reply With Quote
  #380  
06-05-2003, 03:20 PM
Jellygoose Jellygoose is offline
Free Member
 
Join Date: Jun 2002
Location: Germany
Posts: 1,288
Thanks: 0
Thanked 0 Times in 0 Posts
I'll give that a try right now, thanks for the tip kwag!

Edit: However, since it was last tested a long time ago, I'd like to give that a try too! remember CQ_VBR did beat CQ a long time ago too... TMPGEnc has changed maybe, so if you don't mind sending me the part of the script with adaptive Blockbuster variance, I'll give it a shot!
__________________
j3llyG0053
Reply With Quote
Reply




Similar Threads
Thread Thread Starter Forum Replies Last Post
Avisynth: Dynamic Linear Adaptive Filtering and Scene Change Detection supermule Avisynth Scripting 3 09-15-2006 01:45 AM
Avisynth: Message from linear motion adaptive filtering? holgerschlegel Avisynth Scripting 7 08-27-2003 04:08 AM
Avisynth: Help with Linear Motion Adaptive Filtering pitoman Avisynth Scripting 2 08-05-2003 12:51 PM
Avisynth: Motion adaptive filtering good enough? bicho_visacoso Avisynth Scripting 6 06-15-2003 06:30 AM
Avisynth: AVS 2.5x Script, Motion Adaptive Filtering problems? Bchteam Avisynth Scripting 15 05-31-2003 12:38 PM

Thread Tools



 
All times are GMT -5. The time now is 01:42 AM  —  vBulletin Jelsoft Enterprises Ltd