New "Color Range..." feature along with Luma Range for color based non-destructive layer mask
Implemented
Currently Luma Range option helps to refine mask based on luminosity values non-destructively. Where as to refine further based on color there is not option similar to Luma Range. The only method available is create layer based color first by using "Create Masked Layer From Selection" option from Color Editor tool and then apply Luma Range on that layer to further refine based on luminosity.
- Biggest limitation is that the mask created based on color selection is "Rasterized" mask similar to the mask created by brush. After the layer is created if original color is changed it doesn't get reflected.
Where as if Color Range option is added as shown in below picture similar to Luma Range, color based non-destructive dynamic mask can be created. So, any layer with initial mask created through Filled/Brush/Gradients can be further refined by "Luma range based on luminosity" and "Color Range" based on color.
-
Hi Prasad,
Thank you for these suggestions shared here.
I have forwarded those to the product manager as something to consider when developing future versions of Capture One.
0 -
This would be an awesome feature!!
1 -
Can we expect this feature in 2022?
0 -
When the Luma range feature was introduced this was a happy day for me, as it makes it very easy to adjust certain parts of the image, and really is non-destructive, whereas the implementation to define/refine masks based on color is not, as you cannot revisit and change the pixels based on color again.
This would really be a valuable feature, even in late 2021 :-)
Prasad, C1 will not provide an answer here.
0 -
As color based refinements are not dynamic Color range will bring infinite amount of possibilities when both Luma and Color Range tools...
Final mask can be based on Luminosity + Hue + Saturation ...
0 -
At what point in the process are you proposing to pick the colour range?
At base level with Base characteristics applied?
Or any level, using whatever colour modify adjustments may have been applied (and perhaps later changed) at some previous level?
0 -
SFA - Think "Color Range" tool is same as "Luma Range" but on Color instead Luminosity to refine the Mask Dynamically.
0 -
Hi SFA,
The Luma Range seems to be fairly unaffected by "luminance adjustments" e.g. exposure etc. It seems to have an anchor / base at the actual base development, maybe after the curve is applied.
On the other hand there are examples for tools which are affected by changes in the image processing pipeline, e.g.
In the Advanced Color Editor you can define multiple stakes / wedges of the color cake in succesion, each subsequent adjusment picking up the output of the previous.
There are other tools in C1 which influence each other e.g. exposure>curves, exposure>levels, levels>curves etc.One or the other way there will be advantages and disadvantages (as always...) I think I would prefer if the Color Range mask would be unaffected by other tools color changes.
0 -
Hi BeO,
I agree.
I think the issues about the effects on colours was one of the reasons that various Luma adjustments and the Luma Range influence on masks were introduced.
I would agree that the colour range mask should be "fixed" at the point of definition as far as the colour is concerned. At the moment that implies that after fixing you have defined a mask, based on colour, that can be used to apply further adjustments not necessarily based on colour. Some of those adjustments are likely to move the colour result away from the selected colour. Indeed your example specifically applies that requirement.
However, if one is to add further functionality to the mask creation activity (rather than the "what to do with the mask" activity) the implication is that some sort of clever coding will be applied to, for example, re-interpret the mask based on other change. So the mask becomes dynamic always looking for a specific colour range even as other edits are applied to an image (or set of images if what is promised for V22 is delivered as we might expect it to be delivered.)
Now that is where I think it would be easy to get results that might be unintended. Perhaps also some results in specific situations that might be exactly as intended provided one was very familiar with how things were intended to work.
My understanding of the principle of C1 is that the base reference point is always the original image.
If so the colour that one might pick at some point during an edit would, presumably, be mapped back to the "RAW" value and be used for selection for a new mask? But if you pick the same colour you get the same mask. If you pick what might currently appear on screen as a different colour you should, for a dynamic mask, probably get a different mask.
That's probably OK if making a new selection - but is it OK to dynamically change a previous selection?
If you really want a new mask based on a new selection when previous edits made have caused the re-selection requirement as one now sees the image in progress, how can that be accommodated by a dynamic mask using AI and still produce the results that the person editing can predict to be applied?
My past experiences suggested that, so long as one was not attempting to be too extreme in the changes to be applied, in most cases the results were acceptable even if not entirely as expected. But it was quite easy to get to a point where multiple adjustment layers were active and the adjustments were taking things quite a long way from the starting point in some parts of an image, such that unexpected changes could appear without any obvious way to discover why. Moving the position of a layer adjustment in the stack of adjustments might indicate which changes were and were not having an effect - but that was a long way from knowing where to go back to and adjust what had been done there in order to make things work further up the stack.
I enjoyed using the other applications functionality, though much of it seemed irrelevant for my purposes.
However, C1 back then seemed to deliver a result closer to what I was seeking without doing anything at all.
My time was better-used processing more images than trying to fine-tune some powerful but not entirely necessary editing features that could easily be persuaded to fight each other without realising it.
Just my thoughts.
I'm not a fan of the benefits, if there are any, of extended complexity in everything.
0
投稿コメントは受け付けていません。
コメント
9件のコメント