Star Wars GOUT in HD using super resolution algorithm (* unfinished project *) — Page 44

It says so on the Infognition website, where they explain the method with some Angelina Jolie pictures. Going from native 1920x1080 to 1440x1080 naturally involves a downscale, where you lose vertical resolution. It's not as bad as a full downscale in both directions, but still noticeable.

DrDre said:

That may certainly be true. On the other hand these photos have been the basis for a large number of color correction projects on this forum. However, it would of course be better to compare directly to scans of the print. 

zee944 said: said:

You mean people used those photos taken in the theater as reference?What a horrible idea. I don't know what they were thinking. 

DrDre said:

... Michael Kaminski, who atttended the screening, ... said these photos are the best reference for how the film is supposed to look ... but they are said to to be in the ballpark. 

 I would say not even in the ball park for this reason (and the post after it) ...

http://originaltrilogy.com/forum/topic.cfm/The-Matrix-35mm/post/672964/#TopicPost672964

I really, REALLY like your colour correction, except for the crushed blacks

Having matched the colors of Team Negative1's preservation (the video sample) to the GOUT, we can now directly compare it to SRV13:

http://screenshotcomparison.com/comparison/138443

Granted, the GOUT colors reproduction is not as great, but the amount of detail it has compared to Team -1s frame is stunning. It even has better contrast in some areas.

Zyrother said:

Granted, the GOUT colors reproduction is not as great, but the amount of detail it has compared to Team -1s frame is stunning. It even has better contrast in some areas.

It's not that comparable, as their work is much softer. You can make it much sharper which will bring out a lot of details. And halos too (of which GOUT has plenty).

Yes, but the sharpening will also introduce artifacts. If the print is in good enough condition, and the quality of the scanner is good enough, no sharpening should be required. 

The compression probably removed a lot of details, and the compression in these samples is truly awful, so a fair comparison can only be made if we get some uncompressed frames. 

DrDre said:

Having matched the colors of Team Negative1's preservation (the video sample) to the GOUT, we can now directly compare it to SRV13:

http://screenshotcomparison.com/comparison/138443

 Stunning!

The greens issue is weird. In the -1 frame there is too much green behind Vader. In the SR frame there is too much green on the left edge of the frame, but it looks much better behind Vader?

Well, after reviewing a video sample of SRV13 I've come to the conclusion that the slow moving grain can be reduced somewhat, but at the expense of detail and clarity, as was clear from the screenshots. All in all, I think the grain is just something we're going to have to live with. It's part of the source, and gets enhanced with everything else. I will continue rendering SRV12.

DrDre said:

All in all, I think the grain is just something we're going to have to live with.

(old) Obiwan: "... [maybe] You can't win, but there are alternatives to fighting [it] ..."

How about .... start all over?  :O  ... with a better picture right off the hardware?

http://originaltrilogy.com/forum/topic.cfm/Unofficially-Official-LD-Decode-thread/topic/17924/  ...

quote from Happycube:

I hacked up a video capture card to work as a direct ADC, then capture the raw RF signal from a test point on the Laserdisc player. ... This is the best direct comparison I could come up with between the X0 and [the] ld-decode (using a CLD-V2800, a rather low end player!): http://screenshotcomparison.com/comparison/128845

Spaced Ranger said:

quote from Happycube:

I hacked up a video capture card to work as a direct ADC, then capture the raw RF signal from a test point on the Laserdisc player. ... This is the best direct comparison I could come up with between the X0 and [the] ld-decode (using a CLD-V2800, a rather low end player!): http://screenshotcomparison.com/comparison/128845

 Perhaps we should just keep waiting on Team XO to deliver. Perhaps the problem is that they have not been paypal'ed enough dollars?

Hello, I'm new here. I have been following this thread for some time now and I am really impressed of the results so far.

I just tried the SRV10 script on my copy of the GOUT(pal).

I played around some in QTGMC and tried to apply some denoise.

The result is quite good I must say.

Spline64 (original gout frame) vs SRV10 + denoise and sharpen

http://screenshotcomparison.com/comparison/139412

@DrDre

- Will you share with us the SRV13/12 script any time soon?

Looks quite good indeed! I will be sharing the script soon.

DrDre said:

Looks quite good indeed! I will be sharing the script soon.

 Great. I'm looking forward to it.

Edit:

I have a question though. You have probably answered it previously in the thread, and in that case I must have missed it.

Will you be trying to stabilize the image and de-shake it? Maybe implement g-force's stabilization script?

Possibly, I will do some tests first to see how it works out.

Kurosawa10 said:

DrDre said:

Looks quite good indeed! I will be sharing the script soon.

 Great. I'm looking forward to it.

Edit:

I have a question though. You have probably answered it previously in the thread, and in that case I must have missed it.

Will you be trying to stabilize the image and de-shake it? Maybe implement g-force's stabilization script?

 If I recall, isn't g-force's script dependent on additional filtering? (denoising? I forget...). Maybe it would affect the results?

If Premiere is going to be used to add the subtitles (last I checked, it might have been the plan) and encode an end-user version, some gateweave removal can be done at that stage. Though, it's picky and in my experience is just better using on individual shots that are particularly problematic. It would be nice to at the very least fix the intro and crawl in this regard. 

DrDre said:

Possibly, I will do some tests first to see how it works out.

towne32 said:

Kurosawa10 said:

DrDre said:

Looks quite good indeed! I will be sharing the script soon.

 Great. I'm looking forward to it.

Edit:

I have a question though. You have probably answered it previously in the thread, and in that case I must have missed it.

Will you be trying to stabilize the image and de-shake it? Maybe implement g-force's stabilization script?

 If I recall, isn't g-force's script dependent on additional filtering? (denoising? I forget...). Maybe it would affect the results?

If Premiere is going to be used to add the subtitles (last I checked, it might have been the plan) and encode an end-user version, some gateweave removal can be done at that stage. Though, it's picky and in my experience is just better using on individual shots that are particularly problematic. It would be nice to at the very least fix the intro and crawl in this regard. 

 I agree, fixing at least the intro and crawl would be great.

Another video sample will be ready in three days. I believe I've managed to get rid of the intrusive grain, but at a price of course. However, I believe it can be best judged in motion. I will post a sample for both SRV12 and SRV13 based on towne32's cleaned up GOUT, and then it's time for critique ;-).

Yes, I'll clean up the script and post it along with the sample.

The video sample is done. I just have to add sound to it, and convert it to mkv, so it should be online this evening (European time).