DrDre

I think the general consensus is that the bluray is too orange, the WOWOW is more reddish, while the theatrical grading was more yellow. Although the 35 mm stills are too saturated, it's still interesting to see if the regrade is closer to those colors, than the bluray. Here are some comparisons (order: 35 mm, bluray, WOWOW, regrade):

The sky just doesn't look right in either the bluray or the WOWOW. It's supposed to be blue, like in the regrade. Also, particulary in the last example, the colors in both the bluray and the WOWOW are flat, too orange in the bluray, too red in the WOWOW. The regrade has a full spectrum of colors, and looks much more natural. 

Ronster said:

yep that is more like it :)

Belloq is Brilliant from WOWWOW, but your still overdoing something look at the Hat.

Like this perhaps?

Regardless the WOWWOW has Blue in this so it is the choice and the BLU-Ray is no good.

The first Image does not need to be altered really it's more or less correct as it is.

 I don't necessarily argree, one of the nice things to come from the still is the greater dynamic range. This is most obvious in the frame of Indy running in the grass, but the reference as well. The bluray colors are very monotone. The colors you use are more muted. These scenes were shot in extremely warm conditions in Egypt, so I think the regraded frame reflects that pretty well.

This what you get if you use your regrade as a reference for my reference frame.

Ronster:

DrDre:

I prefer the latter, but I admit that I generally prefer higher dynamics.

It looks absolutely perfect! Thank you very much for the hard work!

After being convinced by a few members that the Raiders regrade is too saturated, I decided to use a different reference, namely a 1981 publicity still:

Bluray:

Bluray matched to reference (slightly corrected for blue shift): 

The sky is now a natural blue, and the colors are very balanced. Interestingly, predicting the frame that was matched to the former reference frame leads to a result that is close to the former reference, without being oversaturated:

35 mm frame:

Bluray:

Bluray corrected:

More in the Raiders of the Lost Ark regrade thread:

http://originaltrilogy.com/forum/topic.cfm/Raiders-of-the-Lost-Ark-35-mm-regrade/

Here are a two more comparisons to the WOWOW (top WOWOW, bottom regrade)

Here's how the color correction model would color grade a number of other frames based on the new reference (bluray top, correction bottom):

I think these are close to perfect...

I agree the regrade is too saturated. I decided to use a different reference, namely a 1981 publicity still:

Bluray:

WOWOW:

Bluray matched to reference (slightly corrected for blue shift): 

The sky is has a more natural blue color than in the bluray or the WOWOW. More screenshots to come...

nightstalkerpoet said:

That's awesome you figured it out. New release coming soon then?

Small request - could you include your release number in the Program Name at the top of the tool, and keep this thread title updated with the current release.

ie DrDre Color v1.2 and Color matching between two sources - GUI v1.2 released

also, would this help you set up video input support? http://stackoverflow.com/questions/15538730/reading-and-displaying-video-file-frame-by-frame

 Sure, on that point, how do you change the title of a thread?

nightstalkerpoet said:

Is this due to your highlight compensation?

 No, one of the stabilization parameters I use was set a little too high, resulting in a lower accuracy.

Multi color space model predictions are now just as robust as single color space models. Since both the single and multi color space model are able to match the Raiders reference frame, their predictions should be virtually identical.

Single color space model prediction:

Prediction of multi color space model:

I think I might have come up with a better solution for the crushed blacks/whites. Going back to the The Empire Strikes Back frame with a color correction model I calibrated on a frame of Star Wars, I had the following problem giving the frame a darker grading:

Semi-Specialised Edition V2.2:

This is what I got:

The source frame suffers from crushed whites, so the model could not correctly estimate the colors in those areas, resulting in artifacts.

This time I decided to upgrade the image with a new algorithm I developed to estimate what an image with a higher color depth would look like, and use that for the color correction as a source frame. The source frame looks identical on the surface, but when I color correct it now I get this:

I think this looks much better.

Well there's a new hope afterall. Of George says it can happen, it probably will happen... Yippieeeeee!

Ronster said:

What about this reference...

http://www.youtube.com/watch?v=dprxDRIq23I

seems like it's closer to the Blu-ray to me than this color matching program. Not saying it's bad but to me I think there is something you are failing to take into account... Common sense on skin colors and so fourth. Over saturated Blues and Reds should be very apparent to you by now in your output... You can see that it's just not a natural range of color.

 The fact that it's closer to the bluray is not surprising. It is a showing of a new 35 mm print, probably based on the same master as the bluray, so it's colors will be pretty much identical. So, it's not a very good reference for how the film looked in 1981. The reference I'm using is from an actual 1981 print. Whether it's accurate is of course questionable, but the colors are a lot more saturated than the bluray, and closer to what comes out of the color matching algorithm. 

The artifacts in C3PO's leg are greatly reduced, although there's always going to be some artifacts, because of crushed whites. No method at this point can predict what isn't there, at least not without introducig problems in other areas.

Bluray:

Bluray regraded (using old reference, not Tech IB print):

nightstalkerpoet said:

That's what I'm getting with prediction based on the alley shot.

 That's what I got as well, when working in Ycbcr color space. In RGB you get this:

DrDre said:

I believe I have solved the problem of crushed whites, and blacks for the CollorCorrect GUI v1.2. While trying to improve the grading for a frame of The Empire Strikes Back with a color correction model I calibrated on a frame of Star Wars, I had the following problem:

Semi-Specialised Edition V2.2:

This is what I got:

The source frame suffers from crushed whites, so the model could not correctly estimate the colors, resulting in artifacts.

I realized where the problem is with crushed whites and blacks, namely that it introduces errors in the color distribution, implemented a solution, and this is what I got:

Pretty neat, I think, and the colors look pretty good as well... :-)

 Turns out the solution for crushed whites and blacks reduces the accuracy of the match. So, back to the drawing board for this problem.

Laserschwert said:

What I noticed in your regrades though (including the SW shots in the color matching thread) are the occasional blown out highlights. In the shot of Indy and Sallah by the map room, the sky has some areas where at least red and green are at their full value of 255 (which you should try to never reach in an image), while the original image still has some variation in there.

The dark areas show similar problems with some details just being swallowed in black. It's not by much, so flattening the gamma curve just a bit could help with these problems.

Apart from that I feel the images look a bit oversaturated, but I'm not sure if this is a tendency 35mm material has.

 In principle this method is mainly aimed at reproducing the color properties of the reference. If the reference has blown out highlights, and is saturated, so will the regrades be. 

nightstalkerpoet said:

Which color space are you working in for single match mode? I'm still getting (imo) better results with my gimp process - take a look at my steps and see if there's anything about it that can be implemented in the gui.

http://screenshotcomparison.com/comparison/143328

 You should try your method on the skeleton frame. I've not had good experience working with YCbCr. When I use it, the blue in the background is muted.

Also in this frame, the sky is more green, and the wires on the right have virtually disappeared.

CatBus said:

DrDre said:

CatBus said:

DrDre said:

I believe I have solved the problem of crushed whites, and blacks for the CollorCorrect GUI v1.2.

I have to say, this might change everything for me.  Back when you were color correcting the Blu-ray exclusively, I was inclined to believe that contrast boosting on the Blu was destroying your highlight detail.  But then I saw the same thing when you corrected the GOUT and I thought something must be wrong with the method in general.  I didn't complain because frankly I felt you were catching quite enough flak already, and were still doing interesting work in spite of the problems.

Now... things look very promising indeed.  Keep up the good work, and I'd like to see more captures using this new method.

 Could you explain with an example?

The stormtrooper on the far right's chest:
http://screenshotcomparison.com/comparison/139124

Artifacts in the highlights on C-3PO's leg:
http://screenshotcomparison.com/comparison/140781

Lots of highlight detail lost here:
http://screenshotcomparison.com/comparison/142085

Seeing those shots redone with the new method would be very interesting to me.

 The first two were actually done with a different model, than the one I've been using for a while now. The last example is actually due to lack of color depth and noise in the GOUT. I will try to show some more examples, but the new method was used since page 6 of this thread. 

CatBus said:

DrDre said:

I believe I have solved the problem of crushed whites, and blacks for the CollorCorrect GUI v1.2.

I have to say, this might change everything for me.  Back when you were color correcting the Blu-ray exclusively, I was inclined to believe that contrast boosting on the Blu was destroying your highlight detail.  But then I saw the same thing when you corrected the GOUT and I thought something must be wrong with the method in general.  I didn't complain because frankly I felt you were catching quite enough flak already, and were still doing interesting work in spite of the problems.

Now... things look very promising indeed.  Keep up the good work, and I'd like to see more captures using this new method.

 Could you explain with an example? Contrast boosting generally results in crushed highlights. There's pretty much nothing you can do about this. The detail in de mid range is enhanced, at the cost of the highlights. 

nightstalkerpoet said:

So just to clarify,  the single frame technique corrects all resulting frames in a single color space? 

If so, is there a chance to implement a third option, where the original image is corrected, saved, and then the multiple color space model analyzes the source image and uses the new image as the reference? I'm just thinking that some images in the film may require a little more force to push the correction. Or is that assumption incorrect?

 That would not work, because the multiple color space method would essentially immediately converge to the single color space method, because the initial correction effectively removes all the color information in the other spaces.

Laserschwert said:

Do you think you can make the algorithm avoid crushed whites on the result, even if the source frame has them?

This is exactly what it does, it corrects for crushed blacks and whites in both the source and the reference, to a certain extend.

I believe I have solved the problem of crushed whites, and blacks for the CollorCorrect GUI v1.2. While trying to improve the grading for a frame of The Empire Strikes Back with a color correction model I calibrated on a frame of Star Wars, I had the following problem:

Semi-Specialised Edition V2.2:

This is what I got:

The source frame suffers from crushed whites, so the model could not correctly estimate the colors, resulting in artifacts.

I realized where the problem is with crushed whites and blacks, namely that it introduces errors in the color distribution, implemented a solution, and this is what I got:

Pretty neat, I think, and the colors look pretty good as well... :-)

In the Raiders color timing thread it was remarked that the bluray was an attempt to approach the theatrical color grading, and that the original color grading was more yellow, especially in the desert scenes, as is evident from this 35 mm frame:

Well the color correction model predicts the following for desert shots:

Bluray:

Bluray corrected with single frame color correction model:

I think the color correction model nailed it.

In the Raiders color timing thread it was remarked that the bluray was an attempt to approach the theatrical color grading, and that the original color grading was more yellow, especially in the desert scenes, as is evident from this 35 mm frame:

Well the color correction model predicts the following for desert shots:

Bluray:

Bluray corrected with single frame color correction model:

I think the color correction model nailed it.