logo Sign In

Color matching and prediction: color correction tool v1.3 released! — Page 31

Author
Time

Awesome! For which movie? What type of print?

Author
Time
 (Edited)

DrDre said:

The debugged color matching tool v1.2 offers much improved color matching capabilities. It’s more accurate, more stable, and much faster.

Just a little late catching up here, but … whoa, that’s an awesome demonstration!

  • Will you incorporate any algorithm to prevent the originally flattened RGB from “stair-stepping” on expansion (looking like acne pot-marks especially on skin colors)?

[EDIT: Actually, most of what finally shows is on the source (ouch, I’ve seen those R-G-B splits!). Only, it is stronger once the RGB spectrums are normalized. That would require pre-processing, which is probably outside of the scope of a color-match project.]

Author
Time
 (Edited)

Spaced Ranger said:

DrDre said:

The debugged color matching tool v1.2 offers much improved color matching capabilities. It’s more accurate, more stable, and much faster.

Just a little late catching up here, but … whoa, that’s an awesome demonstration!

  • Will you incorporate any algorithm to prevent the originally flattened RGB from “stair-stepping” on expansion (looking like acne pot-marks especially on skin colors)?

[EDIT: Actually, most of what finally shows is on the source (ouch, I’ve seen those R-G-B splits!). Only, it is stronger once the RGB spectrums are normalized. That would require pre-processing, which is probably outside of the scope of a color-match project.]

In this case the color match was done using an 8-bit color image. In practise you would use at least a 16-bit color image, which would prevent stair stepping, unless it is part of the original 35mm frame. It’s also interesting to note, that this print has sufficient amounts of color noise.

Author
Time

In practise you would use at least a 16-bit color image, which would prevent stair stepping, unless it is part of the original 35mm frame.

Pardon my ignorance, but what do you mean when you say “stair stepping”?

I am proud to say I remember the 80’s!

http://originaltrilogy.com/topic/Episode-1-TPM-game-sounds-files/id/15201/page/1

Currently working on: Red Dwarf Night 10th Anniversary V2.
http://originaltrilogy.com/topic/Red-Dwarf-Night-10th-Anniversary/id/18056#781639

Red Dwarf Night 10th Anniversary

Author
Time

peter_pan said:

In practise you would use at least a 16-bit color image, which would prevent stair stepping, unless it is part of the original 35mm frame.

Pardon my ignorance, but what do you mean when you say “stair stepping”?

When you perform a color correction at a low color depth, the colors from one pixel to the next may resemble a step function, rather than a smooth transition.

Author
Time

DrDre said:

peter_pan said:

In practise you would use at least a 16-bit color image, which would prevent stair stepping, unless it is part of the original 35mm frame.

Pardon my ignorance, but what do you mean when you say “stair stepping”?

When you perform a color correction at a low color depth, the colors from one pixel to the next may resemble a step function, rather than a smooth transition.

Right, got you.
Can’t you have your time line set to a higher depth so that there is head space for the addition of the colour space?

I am proud to say I remember the 80’s!

http://originaltrilogy.com/topic/Episode-1-TPM-game-sounds-files/id/15201/page/1

Currently working on: Red Dwarf Night 10th Anniversary V2.
http://originaltrilogy.com/topic/Red-Dwarf-Night-10th-Anniversary/id/18056#781639

Red Dwarf Night 10th Anniversary

Author
Time

peter_pan said:

DrDre said:

peter_pan said:

In practise you would use at least a 16-bit color image, which would prevent stair stepping, unless it is part of the original 35mm frame.

Pardon my ignorance, but what do you mean when you say “stair stepping”?

When you perform a color correction at a low color depth, the colors from one pixel to the next may resemble a step function, rather than a smooth transition.

Right, got you.
Can’t you have your time line set to a higher depth so that there is head space for the addition of the colour space?

What do you mean exactly? Could you elaborate?

Author
Time

peter_pan said:

DrDre said:

peter_pan said:

In practise you would use at least a 16-bit color image, which would prevent stair stepping, unless it is part of the original 35mm frame.

Pardon my ignorance, but what do you mean when you say “stair stepping”?

When you perform a color correction at a low color depth, the colors from one pixel to the next may resemble a step function, rather than a smooth transition.

Right, got you.
Can’t you have your time line set to a higher depth so that there is head space for the addition of the colour space?

I don’t think that would work in this case: If you take a frame from the blu-ray for example, which is 8-bit color, and you bring that into your After Effects timeline and set the color depth to 16 bit, it won’t magically double the number of colors in the image you imported.

The way I interpreted Dre’s comments was that the images he was working with were 8-bit color, and this is why the example images are not so good as they might be if he had 16-bit versions of the same images to play with.

TheStarWarsTrilogy.com.
The007Dossier.com.
Donations always welcome: Paypal | Bitcoin: bc1qzr9ejyfpzm9ea2dglfegxzt59tys3uwmj26ytj

Author
Time

Williarob said:

peter_pan said:

DrDre said:

peter_pan said:

In practise you would use at least a 16-bit color image, which would prevent stair stepping, unless it is part of the original 35mm frame.

Pardon my ignorance, but what do you mean when you say “stair stepping”?

When you perform a color correction at a low color depth, the colors from one pixel to the next may resemble a step function, rather than a smooth transition.

Right, got you.
Can’t you have your time line set to a higher depth so that there is head space for the addition of the colour space?

I don’t think that would work in this case: If you take a frame from the blu-ray for example, which is 8-bit color, and you bring that into your After Effects timeline and set the color depth to 16 bit, it won’t magically double the number of colors in the image you imported.

The way I interpreted Dre’s comments was that the images he was working with were 8-bit color, and this is why the example images are not so good as they might be if he had 16-bit versions of the same images to play with.

Exactly! 😃

Author
Time

The only thing you can do in this isntance is add a little noise to the image. This stops the stepped-gradient effect happening when you start adjusting things. The noise can then mostly be removed afterwards.
It isn’t perfect, but can stop colour banding from occuring when manipulating 8 bit files.

Donations welcome: paypal.me/poit
bitcoin:13QDjXjt7w7BFiQc4Q7wpRGPtYKYchnm8x
Help get The Original Trilogy preserved!

Author
Time

poita said:

The only thing you can do in this isntance is add a little noise to the image. This stops the stepped-gradient effect happening when you start adjusting things. The noise can then mostly be removed afterwards.
It isn’t perfect, but can stop colour banding from occuring when manipulating 8 bit files.

This was my first solution in the tool as well. Currently, I’m using a different approach, where a smoothing parameter is used to decide the smoothness of the gradient. This reduces stair stepping and similar artifacts.

Author
Time

poita said:

The only thing you can do in this isntance is add a little noise to the image. This stops the stepped-gradient effect happening when you start adjusting things. The noise can then mostly be removed afterwards.
It isn’t perfect, but can stop colour banding from occuring when manipulating 8 bit files.

Building on this idea: check out f3kdb: https://f3kdb.readthedocs.io/en/latest/

A picture is worth a thousand words. Post 102 is worth more.

I’m late to the party, but I think this is the best song. Enjoy!

—Teams Jetrell Fo 1, Jetrell Fo 2, and Jetrell Fo 3

Author
Time
 (Edited)

DrDre said:
In this case the color match was done using an 8-bit color image. In practise you would use at least a 16-bit color image …

Yes, if one starts in 16-bit to finally end in 8-bit, it would fudge a solution. But the problem is, with faded film, the scanned R-G-B layers no longer have that 16-bit (or 8-bit) spread. They might have (for example, in 8-bit) only a 7 or 6 or 5-bit spread. As soon as one color-corrects it (expands it back to 8-bit), steps are introduced in that original workspace. Same for working in 16-bit.

poita said:
… add a little noise to the image. This stops the stepped-gradient effect happening when you start adjusting things. The noise can then mostly be removed afterwards.

DrDre said:
Currently, … a smoothing parameter is used to decide the smoothness of the gradient. This reduces stair stepping and similar artifacts.

My thought was to use integration – “an integral assigns numbers to functions in a way that can describe displacement, area, volume, and other concepts that arise by combining infinitesimal data” – or was it differentials (the reverse of integration; my advanced calculus is very foggy now) – to create an entirely new spread based on the old, remaining samples. It’s probably vector calculus (using both), but I never studied that.

Does this sound like a right approach (vectors) and/or what do pro-software incorporate for their solutions?

Author
Time

Regarding the SOTS 35mm reference to 16mm target … I noticed that all matching/prediction attempts produced colorized tree-trunk shadows for the 16mm. (Such poor distributions are allot easier when viewed on the LCD screen I’m using, simply by changing my head position – never before thought of that as a plus.) I’ve come across something similar when using HSL color transferring between same pictures.

When the L (lightness) of the 2 pictures have differences (contrast, brightness), the transferred HS (hue,saturation) “colors” are altered by the destination’s different L, and sometimes unacceptably so. I’ve found that pre-matching the lightness of the destination to the source (very easy to do on the B&W L splits) produces a proper color transfer result.

Enough of your excellent matches show such anomalies that they cause you constant push-here-pull-there touch ups. Would some sort of R-G-B pre-normalization step, to create a better distribution for your limited samplings, save you such re-adjustments? (I would suggest using a larger sample pool, but the increased storage and processing requirements for such diminishing returns might be prohibitive.)

Author
Time

Spaced Ranger said:

Regarding the SOTS 35mm reference to 16mm target … I noticed that all matching/prediction attempts produced colorized tree-trunk shadows for the 16mm. (Such poor distributions are allot easier when viewed on the LCD screen I’m using, simply by changing my head position – never before thought of that as a plus.) I’ve come across something similar when using HSL color transferring between same pictures.

When the L (lightness) of the 2 pictures have differences (contrast, brightness), the transferred HS (hue,saturation) “colors” are altered by the destination’s different L, and sometimes unacceptably so. I’ve found that pre-matching the lightness of the destination to the source (very easy to do on the B&W L splits) produces a proper color transfer result.

Enough of your excellent matches show such anomalies that they cause you constant push-here-pull-there touch ups. Would some sort of R-G-B pre-normalization step, to create a better distribution for your limited samplings, save you such re-adjustments? (I would suggest using a larger sample pool, but the increased storage and processing requirements for such diminishing returns might be prohibitive.)

The latest version of the tool does better in this respect, but some anomalies are unavoidable. I will take a look at the method you’re suggesting, and see if I can incorporate it in some way.

Author
Time

You may recall that some time ago TN1 was experimenting with a very faded Eastman Reel 5, using Histograms and HSV color spaces to try to restore the colors. Anyway, the end results were not really usable, some color was restored and some of it even looked ok, but there was also a lot of noise and a lot of artifacts, so ultimately the reel was put aside.

However, I just rediscovered it while trying to free up some hard drive space, and figured I would run it through Dre’s tools to see what would happen. Now I only have access to an 8-bit color version of the reel so these aren’t as good as they could be… But, well, see for yourself:

Here is the original frame (actually it’s just a nearby frame, I didn’t make note of the frame number when I started, but it’s close enough to illustrate the faded colors of the original):

Imgur

First I used the color balance tool, which gave me this:

Imgur

Next I used the Match tool and matched the balanced version I just created to the same frame from the SSE:

Imgur

Not too shabby! And look how clean the reel is. Makes me wish we had more than just the one reel.

Dre: Why hasn’t Hollywood come knocking and offered you millions of dollars for these tools?!

TheStarWarsTrilogy.com.
The007Dossier.com.
Donations always welcome: Paypal | Bitcoin: bc1qzr9ejyfpzm9ea2dglfegxzt59tys3uwmj26ytj

Author
Time

That looks amazing! Glad to see the balancing tool is put to good use. 😃

Author
Time
 (Edited)

Good news, I may have access to more of these red faded Eastman Star Wars reels, though condition may vary…

Using just the one set of LUTs generated for the shot above (one LUT for balance, the other for Match) I applied them to the whole sample that I have (about half of Reel 5) and saved it as a quick xvid AVI:

https://we.tl/X2CS5GnF61

Generating a separate set of LUTs for each scene would improve things even more, but I’ll wait until I have a better source to play with before I do that.

Jedit: And here is the Before:

https://we.tl/BQzrgH02Eg

Dre: Sorry to keep bumping your thread, if I am able to get hold of more Eastman scans and decide to do something with them then I’ll create a new thread for the project. Just posting this here because I believe it perfectly illustrates just how astounding your tools are. Thank you so much for sharing them with us.

TheStarWarsTrilogy.com.
The007Dossier.com.
Donations always welcome: Paypal | Bitcoin: bc1qzr9ejyfpzm9ea2dglfegxzt59tys3uwmj26ytj

Author
Time
 (Edited)

That’s a really great result Williarob! I was hoping this would be possible, and I’m glad you were able to show it. 😃 There’s also suprisingly little artifacts, and like you said, it’s a very clean print.

Author
Time

A public question to Dr. Dre or anybody of the beta testers of the tool - in December a new box set will be released containing various clips containing alternate / deleted scenes etc. of classic Star Trek.
2 clips have been released on the official CBS youtube channel
https://www.youtube.com/watch?v=qzwp4nr5zcs
https://www.youtube.com/watch?v=JINWsaK0QM8

The footage will not be restored and as seen in the clips it’s mostly red faded. Could somebody please try the tool on some randomly chosen faded pictures from the youtube vids?

Author
Time

Williarob said:

Good news, I may have access to more of these red faded Eastman Star Wars reels, though condition may vary…

Using just the one set of LUTs generated for the shot above (one LUT for balance, the other for Match) I applied them to the whole sample that I have (about half of Reel 5) and saved it as a quick xvid AVI:

https://we.tl/X2CS5GnF61

Generating a separate set of LUTs for each scene would improve things even more, but I’ll wait until I have a better source to play with before I do that.

Jedit: And here is the Before:

https://we.tl/BQzrgH02Eg

Dre: Sorry to keep bumping your thread, if I am able to get hold of more Eastman scans and decide to do something with them then I’ll create a new thread for the project. Just posting this here because I believe it perfectly illustrates just how astounding your tools are. Thank you so much for sharing them with us.

Damn. I gave your corrected sample a little contrast boost, and it still looks better than the GOUT.

Seeking only the most natural looking colors for Star Wars '77

Author
Time
 (Edited)

pittrek said:

A public question to Dr. Dre or anybody of the beta testers of the tool - in December a new box set will be released containing various clips containing alternate / deleted scenes etc. of classic Star Trek.
2 clips have been released on the official CBS youtube channel
https://www.youtube.com/watch?v=qzwp4nr5zcs
https://www.youtube.com/watch?v=JINWsaK0QM8

The footage will not be restored and as seen in the clips it’s mostly red faded. Could somebody please try the tool on some randomly chosen faded pictures from the youtube vids?

Here you go:

Author
Time

If I use this on transformers, will hot rod become red again? Instead of pink?

Author
Time

DrDre said:

pittrek said:

A public question to Dr. Dre or anybody of the beta testers of the tool - in December a new box set will be released containing various clips containing alternate / deleted scenes etc. of classic Star Trek.
2 clips have been released on the official CBS youtube channel
https://www.youtube.com/watch?v=qzwp4nr5zcs
https://www.youtube.com/watch?v=JINWsaK0QM8

The footage will not be restored and as seen in the clips it’s mostly red faded. Could somebody please try the tool on some randomly chosen faded pictures from the youtube vids?

Here you go:

Actually this is great. I think I know what I’m gonna do for several weeks after the thing gets released 😃