GOUT image stabilization - Released — Page 27

dark_jedi said:

So I assume you made the appropriate changes to the PAL subs version to?

Yes.

dark_jedi said:

also would you say this is it,no more modding,so now we can partner up and do the Japanese Special Collection "Take 2" and the 97 SE Trilogy?

That's it. I'm only going to make changes here on out if there is something that we learn from filtering the Special Collection. The Japanese Special Collection is #1 priority in my book.

So, when will I have something to work on DJ?

-G

g-force said:

So, when will I have something to work on DJ?

-G

what type of file would you need to start getting this rolling,I have both the Japanese Special Collection Star Wars and the 97 SE Star Wars on my hard drive now,I was going to do both at the same time LOL,well at least try to anyways.

I would suggest taking one at a time. I suspect that the SE will require less filtering, and we will probably learn a lot from the JSC, so I would suggest starting with that one. I really don't have as much interest in the 97 SE, but could definitely help with that too.

You could send me a .m2v to get us started. Preferrably already de-interlaced and IVTC'd, as I wouldn't know what to do there. Also, the scene in Obi's hut has some of the more difficult material to filter, so maybe just that scene?

-G

samples are going up in the "Take 2" thread.

http://originaltrilogy.com/forum/topic.cfm/Star-Wars-Trilogy-JSC-Take-2/topic/10521/

Thank you to g-force and everyone for all your efforts developing and refining these scripts.

I've been meaning to try an anamorphic conversion on my NTSC discs for a while, and I have finally started some short tests on ANH. I followed Oldschooljedi's guide to get to the point of running the script and I'm loading the .avs into VirtualDub, then encoding to Lagarith YV12.

I just tried a very short clip with v5.02 and then v4 and I am seeing the results below. I wonder if this is the typical output that everyone else is getting. I did see some of the discussion on the multi-image trails, but this seems worse than some of the previously posted screenshots.

Is this the output I should expect or do I have a problem with my settings? Does v4 of the script work better with older versions of the avisynth plugins (older than dark_jedi's most recent set)?

Are dark_jedi's most recent additions included in v5.02 or will there be another update including those changes?

v4 (split-screen, resized only on left, v4 script on right):






v502 (entire frame):

g-force said:neebis, I'll double check from home later, but those look frames look really messed up. I seem to remember that other people that have had weird artifacts re-installed AVIsynth and the problems went away. Worth a shot. Which version of AVIsynth did you use?

Thanks for the reply. I didn't think this could possibly be the output everyone else was getting. I was using the version of AVISynth I had already installed for earlier cleanup tests on other videos. Version() shows: 2.5.8 tsp MT version 5 April 6, 2009.

What is the recommended version that everyone has been using for the latest tests? Does v4 of the script need a different version than v5.02?

Actually, I see Avisynth 2.5.8 included in the most recent filter pack posted by dark_jedi. I'll try again with a fresh install of that and see what happens.

just get the latest version of everything and you will be good to go.

Moth3r said:

Mielr said:

Just for clarification- the script used for these discs stabilized the gate-weave and improved the aliasing as well (like in the PWNAGE edition)? The reason I ask is because in addition to everything else, the aliasing seems improved too.

Yes, g-force's script includes an anti-aliasing stage (it uses the newer NNEDI2 filter, which wasn't available when I did the Pwnage Edition).

 

"Minor correction (something I wrote in another thread): upscaled with EEDI2 and NNEDI! My method has some of the benefit of EEDI2's ability to connect more lines than NNEDI, but making sure not to introduce "blur bubbles" (nor blur the picture). And yes, in case anyone is wondering, NNEDI2 makes no difference with this method."

I think DJs DVDs were made with a version that used NNEDI, not NNEDI2. I hadn't incorporated it yet, as I found it had no effect with my method of anti-aliasing, so I didn't bump the revision. Then I decided to change the subtitles, and just threw in NNEDI2 for giggles.

-G

These were made with version 4.

Thanks again g-force and dark_jedi. It wasn't the Avisynth version (although I am now using the one you posted), but it was a few filters that needed older versions to work. I had two or three filters which were newer versions, so I reverted to all the same versions as included in dark_jedi's post.

I did some testing of short clips throughout the film to adjust the settings a bit and now I'm running my first full encode to Lagarith.

If it works out, I'll encode with HCenc, then I'll just have to decide which audio to use. I think I'll start out with the Belbucus DC48 audio, then maybe I will try hairy_hen and Satanika's 5.1 upmix. I was thinking about trying my own upmix with the unedited DC48 and hairy_hen's LFE, but I'll have to do some more reading and testing to see if I can replicate Satanika's process.

Hey neebis,

If you're using HCenc, I have a custom Matrix that works really well. Interested?

I'd be curious to see what you tweek the values to. Always looking for creative input. Also, any chance you could re-post those split-screens now that you have the script working?

Thanks,

-G

I just let Imgburn set it and go,I have no idea how to set it to the retail DVD,and to me,I really could care less,if it works,it works.

g-force said:

If you're using HCenc, I have a custom Matrix that works really well. Interested?

I'd be curious to see what you tweek the values to. Always looking for creative input. Also, any chance you could re-post those split-screens now that you have the script working?

Thanks. Sure, please paste or send me a link for your custom matrix. I've only done a few full encodes with HCenc (nothing archival quality) and I usually end up using one of the Fox matrices or one of the defaults.

My full encode is still running, but I'll try to post some screens when it's done. I tried to use MT and Avisynth multithreaded, but I don't really know much about it and I was afraid since MT splits the frame that a lot of the filters and functions wouldn't work right or cause odd artifacts down the middle, so I went back to the standard single-core build.

I only tested and tweaked on my LCD PC monitor so far, but if the full encode works out, I'll test it on my projector.

I went through and used Trim to look at different short sections of the film with different settings. On my LCD screen, the default script looked a little yellow and although many scenes looked fine, the level adjustments were too dark causing a loss of detail in a few shots. Also, on my small screen at least, I felt the grain/noise reduction was a little high for my taste. I'll have to wait and see what I think of it on the big screen with my Blu-ray player upconverting the DVD.

My tweaks are probably more a question of personal taste combined with the (mis?)calibration of my screen rather than much of an improvement over your settings. The settings I've played with so far were Levels, Tweak, sigma and flit sharpen (in the stabilization/degrain stage).

I wasn't really sure about all the values to control the level or noise reduction and sharpness. I tried lowering the sigma to 3 then 4, which is what I'm using. Is that the right variable to lessen the degrain effect?

What about the values below:

source.MVDegrain2 thSAD=400

last.MVDegrain2 thSAD=300,thSCD1=150,thSCD2=75

Will these or other values in that section of the script allow me to fine tune the degrain effect? I plugged in different values (mainly higher), but I didn't get much of an effect that I wanted, so in the end I left them as is.

I also see that Contra is used in the sharpening stage. I really couldn't figure out how to tweak that one, so I didn't touch it. Can you give me a hint about which values to change to increase or lessen the sharpening filter?

neebis said:

I wasn't really sure about all the values to control the level or noise reduction and sharpness. I tried lowering the sigma to 3 then 4, which is what I'm using. Is that the right variable to lessen the degrain effect?

What about the values below:

source.MVDegrain2 thSAD=400

last.MVDegrain2 thSAD=300,thSCD1=150,thSCD2=75

Will these or other values in that section of the script allow me to fine tune the degrain effect? I plugged in different values (mainly higher), but I didn't get much of an effect that I wanted, so in the end I left them as is.

I also see that Contra is used in the sharpening stage. I really couldn't figure out how to tweak that one, so I didn't touch it. Can you give me a hint about which values to change to increase or lessen the sharpening filter?

Changing the settings in the FFT filter won't have as much effect as you are hoping for. The FFT filter is not "really" used, it is for finding the motion vectors, and then as a check against errors. I would suggest NOT chaging that. Changing the values in MVDegrain2 won't do what you want either. They are also used for error prevention.

Okay, as far as the degrain goes, If you want less, change both the lines with MVDegrain2 to the following respectively:

source.MVDegrain1(bw_vec1,fw_vec1,thSAD=400,idx=2)

last.MVDegrain1(bw_vec1,fw_vec1,thSAD=300,thSCD1=150,thSCD2=75,idx=3)

This will use less mo-comped frames for the degraining. Unfortunately, I've found that if you remove less grain, what remains is too temporally stable (probably from the DVD compression) and tends to look less like real random grain, and more like a "veil" of grain that slowly waves. Also, you will not get rid of as much "local motion", meaning that objects that should be stationary will seem to move in relation to the rest of the frame. Try it out, and see what you think. I'm definitely open to making this thing run faster, which reducing the MVDegrain radius would do.

Okay, if you want more or less sharpening, the best way would be to change the function "contra".

this line:

MT_MakeDiff(last,last.RemoveGrain(12,-1).RemoveGrain(12,-1))

for more sharpening would change to:

MT_MakeDiff(last,last.RemoveGrain(12,-1).RemoveGrain(12,-1).RemoveGrain(12,-1))

or less sharpening to:

MT_MakeDiff(last,last.RemoveGrain(12,-1))

If those don't get you where you want, then I would start changing the "sharpen" value in the FFT.

Let me know what works best for you.

-G

g-force, thanks very much for the detailed explaination. My first encode still has 18 hours left... so I'll start testing your suggestions after that's finished.

Please do send me send me your HCEnc matrix when you get the chance. Thanks again.

neebis said:

Please do send me send me your HCEnc matrix when you get the chance. Thanks again.

 It's at home, so I'll send it tonight.

Well, my first encode to Lagarith (YV12 lossless) finally finished after about 50 hours. Here are my settings changes and the resulting screenshots. Remember, these are Lagarith intermediate screenshots, so there will be another generation of MPEG2 encoding after this. These shots are meant to show the effect of the script processing, not the final re-encoded DVD results.

The only values I changed are below.

Levels(7,1.14,255,0,255)
Tweak(sat=1.14,hue=-2)

sigma   = 4

Scanning through the film on my PC, I'm quite happy with the results. There is the expected slight loss of detail and cropping at the edges of the frame, but that's the tradeoff for the correction of gate-weave/frame wobble and anti-aliasing. I don't know if it's possible to coax a bit more detail and sharpness out of the script without causing other problems.

Many shots clearly benefit from the level changes (Leia with blaster below) with improved contrast in the washed out blacks and other shadow areas. I changed the level less than the default script because of the problematic shot of Luke below. If not for that shot, I probably would have used a Level setting of 8 or 9. Actually, scanning through more of the film now, I found quite a few more shots that are also at the threshold of acceptable darkness at 7. Ideally, a shot by shot correction would be allow me to use larger level adjustment on most shots, but I don't think I want to invest the time in that, considering the source.

previously posted shots (without trails or ghosting):




problematic shot (degrain and smoothing loses detail)




source sized only:


problematic shot (black level loses detail on dark half of face, higher than 7 goes too dark)


source sized only:


processed (improved anti-aliasing along with loss of detail):


apt irrelevance comparison frames:









source sized only:


improved contrast (slight loss of detail)








improved contrast

neebis,

Thank you for the constructive criticism. You have inspired me to try out some things I've been kicking around. I have updated my script to 6.00. Turns out that I've nearly eliminated the problems you're showing with the Stormtroopers in smoke and the Rebel pic immediately following it, and it seems to have not hurt the stabilization too much! Please try it out and let me know what you think. (screen caps are nice BTW)

-G