Basic guide to vibrant colouring in Photoshop
Program(s): Photoshop
Involves: Basic tips to graphic making; improving screencaps; preparing a base; levels, colour balance, selective colouring
Translatable: Yes for the basics, no for the tips on selective colouring
Steps: 4
Difficulty: Pretty basic
Want to make graphics with colours like these? This is your tutorial!
I've received requests for tutorials for the colouring of my Gossip Girl icons and picspams. I don't give .psd's or just lists with numbers, because I firmly believe in learning to do something yourself. Quick fixes are a no-go with me. So instead (also because my colouring is pretty general) I offer you my basic colouring guide for Photoshop CS. It's beginner-friendly, except I won't give you screenshots of everything I explain. This is more about the 'how it works in theory' than the 'how-to'.
Step one - A base
Other than set photography, screencaps provide the best bases, provided they are of good quality. The best I've come across so far are by marishna and luux_lu. For picspams, a screencap that's a little blurred due to movement of the character does not have to be a big problem, since you will be resizing the whole picture. If however you are cropping a certain part of it to make an icon, it follows that your icon will be blurred as well, and we wouldn't want that.
Step two - Decide what you want to achieve
If you are going for a creative banner, natural colours might not be what you're looking for. If, however, you're writing a recap of, say, a Gossip Girl episode and your graphics are meant to go with your comments, you're more probable to favour natural looking colours over edgy, daring effects because you want your post to be a whole, your graphics to be an illustration, a way of making your point.
My goal in making images for picspams is cleaning up the screencap and bringing out the most prominent colours already present in the original screencap, making the image vibrant and colourful, but not unnatural.
Step three - Prepare your base
This is something you usually see in tutorials (at least, you used to see it before those godawful radioactive colouring and hideous effects came into vogue), but for me, it should be written in capitals. My base is the basic part of my colouring.
Screencaps tend to look a little washed out and by 'preparing my base', I mean to fix that. For a specific tutorial on cleaning up screencaps with logos, click here.
I first resize the image to fit my picspam and then I copy the background layer twice. I set one of the copied layers to screen and the other to soft light. I see how the image looks and then play with the opacity settings. The advice really is 'experiment', but that doesn't mean you don't have guidelines. Lighter images won't need the screen layer to be at 100% opacity and images with clear contrast will look better with a toned down soft light layer.
See the difference between an untouched picture and the prepared base I made of it?
-->
Whether the screen layer or the soft light is on top can make a difference as well. Just go with what your eyes tell you.
A few things to remember:
- To bring out colours in the eventual result, make sure you don't 'wash out' the base too much by setting the screen layer to a too high opacity, as well as keep the soft light layer from giving too much contrast so all subtle colours vanish from a picture. Find a balance, it really isn't as hard as you might think.
- You can use several screen and soft light copies from your base if necessary. Just copy the one you already had, play with the opacities and see if your image benefits from it.
- If your picture is really dark, say from a night scene, remember that screen layers will light the image up, but will also decrease its quality. The dark parts of a picture will become grainy and ugly. There are ways to work around this, but it can be a pain. Try to avoid really dark pictures as much as possible if you think they will need a lot of screen light to become balanced.
I almost always use both at least one screen layer and soft light layer for my base. What I do then is select all layers and merge them, so I have one, now enhanced, background. I look if the image needs any sharpening or blurring. (NB: this can also be done at the beginning, so before the copying of the base. If you do that, then you will later be able to go back and fix the settings of your screen layer and soft light layer if necessary. With big, plain screencaps however, you sometimes need to do the first steps of colour first because it will make clear exactly what the image needs.)
When do you use a sharpening filter? When the focus of the image (with screencaps, mostly a person) looks blurred. When do you blur? When the skin of a person in the focus of a graphic (or something else clearly in view in the image) is grainy. Be careful with this, one of the ugliest things on a graphic is when it's oversharpened, especially if it's accompanied by overly blurred skin.
With the screencaps I use, I hardly ever find it necessary to sharpen the picture, but I always blur the person's skin. I do this by selecting the blur tool, choose a soft brush and lowering its opacity to what is appropriate. Women's skin tends to be 'softer' than manly stubble, so for women, depending on whether it's a close-up or a shot from a distance, I set it around 60% or 40%.
This is my personal preference, but like I said: make sure you keep a good balance, keep it looking natural. It's better to take a brush of a lower opacity and brush over the same area twice than have it be too blurry in one stroke. For more information on making people's skin look good, click here.
Step four - Additional colouring
By preparing your base, the colouring of your image will have been enhanced. Usually however, you will want to bring the colours of your picture some more. There are several ways to do that. One good way is to use Colour Fill Layers of different colours and use different Blending Modes to bring out certain colours. In my Gossip Girl graphics however, I don't often do that. The three ways I use the most are:
Layer > New Adjustment Layer > Levels
The levels of an image can be adjusted according to the basic colours. I never meddle with the RGB section, I only change little things about the separate settings (Red, Green, Blue). The middle arrow gives the most effect. Level layers are very useful when used subtly.
I use Level layers especially when the light in a scene I use a picture from is very dominantly blue or yellow or whatever other colour. This is because Level layers can easily, but naturally, bring a more normal look back to the image. This goes especially for infusing yellow in overly blue pictures.
Layer > New Adjustment Layer > Colour Balance
Colour Balance layers have a more subtle effect than the direct approach of Level layers, according to basic colour theory. To learn more about colour theory, I'll link you to this brilliantly lucid explanation. I'm not going into detail, but let's just say Colour Balance works in a more general way than Selective Colouring. In Colour Balance, you immediately effect the whole image when you move an arrow one way or the other instead of Selective Colouring, that only effects the colours already present in a picture. You have three options in Colour Balance: working in Midtones, Shadows and Highlights.
When you're working with a decent quality image and going for an enhanced look of colours already in the picture, you will be focussing on the Midtones section. I tend to move arrows only a tiny bit to give a certain colour (say red) just a different tone and depth. In the Blair at the table image, for example, I infused the image with some more magenta, red and yellow.
-->
Layer > New Adjustment Layer > Selective Colouring
As I said, basic colour theory can be complicated, so for our purposed: just keep in mind the question of whether your image would benefit from more or less of a certain colour and move the arrows accordingly. I
In the picture of Chuck, you can see already in the original base the red lining on both his sweater and the cushion he's leaning against. So you try to bring out the red and what sort of red you want it to be. If you select the Red section in the drop down box of your Selective Colour layer and then move the arrow a little towards Magenta, you'll see the red become more magenta - it really all speaks for itself.
-->
A lot of people are complaining about a general abuse of Selective Colouring and I agree. I think Selective Colouring should only be used to enhance colours already there, making them look more vibrant but never unnatural.
So always be moderate when using Selective Colouring, as a rule, never go past the -50 or +50 point when natural enhancement is what you're after.
For Gossip Girl images, I do more with my base than with additional colouring layers, so that when I finish with my base, a single Selective Colouring layer to bring out colours that are already present in the image can be enhanced just a little is enough. If not, duplicating that selective colour layer and setting it to a low opacity (sometimes at the normal blending mode, sometimes at soft light) does the trick of making the image look vibrant but not over-the-top.
Well, that's it, basically. Hope this was in any way helpful!
Please don't take these images without asking me first, thanks!
If you have any questions, don't be afraid to ask!
Involves: Basic tips to graphic making; improving screencaps; preparing a base; levels, colour balance, selective colouring
Translatable: Yes for the basics, no for the tips on selective colouring
Steps: 4
Difficulty: Pretty basic
Want to make graphics with colours like these? This is your tutorial!
I've received requests for tutorials for the colouring of my Gossip Girl icons and picspams. I don't give .psd's or just lists with numbers, because I firmly believe in learning to do something yourself. Quick fixes are a no-go with me. So instead (also because my colouring is pretty general) I offer you my basic colouring guide for Photoshop CS. It's beginner-friendly, except I won't give you screenshots of everything I explain. This is more about the 'how it works in theory' than the 'how-to'.
Step one - A base
Other than set photography, screencaps provide the best bases, provided they are of good quality. The best I've come across so far are by marishna and luux_lu. For picspams, a screencap that's a little blurred due to movement of the character does not have to be a big problem, since you will be resizing the whole picture. If however you are cropping a certain part of it to make an icon, it follows that your icon will be blurred as well, and we wouldn't want that.
Step two - Decide what you want to achieve
If you are going for a creative banner, natural colours might not be what you're looking for. If, however, you're writing a recap of, say, a Gossip Girl episode and your graphics are meant to go with your comments, you're more probable to favour natural looking colours over edgy, daring effects because you want your post to be a whole, your graphics to be an illustration, a way of making your point.
My goal in making images for picspams is cleaning up the screencap and bringing out the most prominent colours already present in the original screencap, making the image vibrant and colourful, but not unnatural.
Step three - Prepare your base
This is something you usually see in tutorials (at least, you used to see it before those godawful radioactive colouring and hideous effects came into vogue), but for me, it should be written in capitals. My base is the basic part of my colouring.
Screencaps tend to look a little washed out and by 'preparing my base', I mean to fix that. For a specific tutorial on cleaning up screencaps with logos, click here.
I first resize the image to fit my picspam and then I copy the background layer twice. I set one of the copied layers to screen and the other to soft light. I see how the image looks and then play with the opacity settings. The advice really is 'experiment', but that doesn't mean you don't have guidelines. Lighter images won't need the screen layer to be at 100% opacity and images with clear contrast will look better with a toned down soft light layer.
See the difference between an untouched picture and the prepared base I made of it?
-->
Whether the screen layer or the soft light is on top can make a difference as well. Just go with what your eyes tell you.
A few things to remember:
- To bring out colours in the eventual result, make sure you don't 'wash out' the base too much by setting the screen layer to a too high opacity, as well as keep the soft light layer from giving too much contrast so all subtle colours vanish from a picture. Find a balance, it really isn't as hard as you might think.
- You can use several screen and soft light copies from your base if necessary. Just copy the one you already had, play with the opacities and see if your image benefits from it.
- If your picture is really dark, say from a night scene, remember that screen layers will light the image up, but will also decrease its quality. The dark parts of a picture will become grainy and ugly. There are ways to work around this, but it can be a pain. Try to avoid really dark pictures as much as possible if you think they will need a lot of screen light to become balanced.
I almost always use both at least one screen layer and soft light layer for my base. What I do then is select all layers and merge them, so I have one, now enhanced, background. I look if the image needs any sharpening or blurring. (NB: this can also be done at the beginning, so before the copying of the base. If you do that, then you will later be able to go back and fix the settings of your screen layer and soft light layer if necessary. With big, plain screencaps however, you sometimes need to do the first steps of colour first because it will make clear exactly what the image needs.)
When do you use a sharpening filter? When the focus of the image (with screencaps, mostly a person) looks blurred. When do you blur? When the skin of a person in the focus of a graphic (or something else clearly in view in the image) is grainy. Be careful with this, one of the ugliest things on a graphic is when it's oversharpened, especially if it's accompanied by overly blurred skin.
With the screencaps I use, I hardly ever find it necessary to sharpen the picture, but I always blur the person's skin. I do this by selecting the blur tool, choose a soft brush and lowering its opacity to what is appropriate. Women's skin tends to be 'softer' than manly stubble, so for women, depending on whether it's a close-up or a shot from a distance, I set it around 60% or 40%.
This is my personal preference, but like I said: make sure you keep a good balance, keep it looking natural. It's better to take a brush of a lower opacity and brush over the same area twice than have it be too blurry in one stroke. For more information on making people's skin look good, click here.
Step four - Additional colouring
By preparing your base, the colouring of your image will have been enhanced. Usually however, you will want to bring the colours of your picture some more. There are several ways to do that. One good way is to use Colour Fill Layers of different colours and use different Blending Modes to bring out certain colours. In my Gossip Girl graphics however, I don't often do that. The three ways I use the most are:
Layer > New Adjustment Layer > Levels
The levels of an image can be adjusted according to the basic colours. I never meddle with the RGB section, I only change little things about the separate settings (Red, Green, Blue). The middle arrow gives the most effect. Level layers are very useful when used subtly.
I use Level layers especially when the light in a scene I use a picture from is very dominantly blue or yellow or whatever other colour. This is because Level layers can easily, but naturally, bring a more normal look back to the image. This goes especially for infusing yellow in overly blue pictures.
Layer > New Adjustment Layer > Colour Balance
Colour Balance layers have a more subtle effect than the direct approach of Level layers, according to basic colour theory. To learn more about colour theory, I'll link you to this brilliantly lucid explanation. I'm not going into detail, but let's just say Colour Balance works in a more general way than Selective Colouring. In Colour Balance, you immediately effect the whole image when you move an arrow one way or the other instead of Selective Colouring, that only effects the colours already present in a picture. You have three options in Colour Balance: working in Midtones, Shadows and Highlights.
When you're working with a decent quality image and going for an enhanced look of colours already in the picture, you will be focussing on the Midtones section. I tend to move arrows only a tiny bit to give a certain colour (say red) just a different tone and depth. In the Blair at the table image, for example, I infused the image with some more magenta, red and yellow.
-->
Layer > New Adjustment Layer > Selective Colouring
As I said, basic colour theory can be complicated, so for our purposed: just keep in mind the question of whether your image would benefit from more or less of a certain colour and move the arrows accordingly. I
In the picture of Chuck, you can see already in the original base the red lining on both his sweater and the cushion he's leaning against. So you try to bring out the red and what sort of red you want it to be. If you select the Red section in the drop down box of your Selective Colour layer and then move the arrow a little towards Magenta, you'll see the red become more magenta - it really all speaks for itself.
-->
A lot of people are complaining about a general abuse of Selective Colouring and I agree. I think Selective Colouring should only be used to enhance colours already there, making them look more vibrant but never unnatural.
So always be moderate when using Selective Colouring, as a rule, never go past the -50 or +50 point when natural enhancement is what you're after.
For Gossip Girl images, I do more with my base than with additional colouring layers, so that when I finish with my base, a single Selective Colouring layer to bring out colours that are already present in the image can be enhanced just a little is enough. If not, duplicating that selective colour layer and setting it to a low opacity (sometimes at the normal blending mode, sometimes at soft light) does the trick of making the image look vibrant but not over-the-top.
Well, that's it, basically. Hope this was in any way helpful!
Please don't take these images without asking me first, thanks!
If you have any questions, don't be afraid to ask!