These Man-Horse beasts lived in Thessaly, fed on meat and were given to riotous revelries. They came to symbolize the dark, unruly forces of nature. They were usually depicted as drunken followers of Dionysus, except for Cheiron who was the tutor to several heroes…

In this tutorial you will learn some basic extraction techniques. How to use blend modes and masks to create deep atmospheric effects. I will also show you how two meld two creatures together to make a mythical half man half horse Centaur.

Step 1

First of all we are going to cut out the elements we are going to use for the main composition. Open ‘horse-23434-12.jpg’ and grab the pen tool ( P ). Being as careful as you can, slowly plot a path around the horse.

When you get the head of the horse don’t worry too much about being neat, this part of the horse will be masked out later on.

Step 2

Right click in the middle of the path and select Make Selection. With a selection around horse, Cut the horse ( CTRL + X ) and create a new Document for our main composition. Any size will do as long as we have enough space to play around in. Mine was 25cm x 25cm. Paste the horse on to the new canvas ( CRTL – V ).

Step 3

Once the horse in in your man canvas make sure you remove any remainder background using pen tool.

Step 4

Open ‘man-katana.jpg’ and using the Pen Tool ( P ) with same method above cut out the man and the sword, make sure you remove the sun glasses on his head.

Cut and Paste the Man and his Katana into the main canvas.

Step 5

Make sure the images you cut out are nice and clean. To Remove the dog tags from mans torso use the  Spot Healing Tool by sampling with the ALT key then painting it out with single clicks. Constant Re-sampling next to where the pixels you want to replace will give the best results.

Step 6

Use Edit > Free Transform > Scale to make the images fit together. Making the horse smaller and tilting the man so it looks like the horses front legs and model are one.

Using the lasso tool ( L ), create a selection around the horses head. Making sure you leave an inch between the models waist and the horses neck to help with blending. Lowered the opacity of the model layer to help. Use a mask or delete the head off the horse.

Step 7

Pick a soft edge brush with around 20% opacity and color set to black. Slowly paint / merge the horse and man together at the waist.

Take the Lasso Tool ( L ) and make a selection on the horse where the torso meets the horses neck. Copy the area you selected ( Control -C ) Then past it on to its own layer naming it ‘ Horse Skin Blend’

Step 8

Change the horse skin blend layer to an Overlay Blend mode and create a mask ( the icon and the bottom of the layers palette that has  a square with a circle in it ).  Grab a soft edge brush and set the opacity to around 10-15%.

Take your time and blend the horse skin into the human skin. Take note of the skins contours and shadows.

Press CTRL – ALT + E to combine these layers into a new layer. Call this layer ‘ Centaur’.

Step 9

Time to adjust the shadows and highlights. Go to image > Adjustments > Shadows / Highlights.

Set the Shadow amount to 12%  Highlights to 23% and The Midtone Contrast to +34. Click OK.

Now open up the Levels ( CTRL + L ) and input

10| 1.00 | 220

Duplicate your Centaur layer by dragging the centaur layer to the Create a new layer icon on the bottom of the layer palette.

With the Centaur copy layer active go to Image > Adjustments > De-saturate and open the levels panel ( CTRL + L ). Input these settings.

32 | 1.00 | 200

Step 11

Inverse the gray scale layer ( CTRL + I ) and select Filter > Blue > Gaussian Blur with a Radius of 10 pixels.

Step 12

Apply an Overlay Blend mode to the layer at 58% Opacity, and 58% Fill. With the Centaur copy layer selected, control click the Centaur layer below and press CTRL – SHIFT + I to inverse the selection. Press delete to remove any blur on the outside of the centaur.

Step 13

Merge the two layers and grab the burn tool. Using a soft brush with the opacity set to about 15-20% with midtone range selected, apply to all the shadows on your centaur.

Now grab the dodge tool and set highlights at 15% and apply to the highlighted areas like the top of muscles.

Step 14

Duplicate the layer again and go to Image > Adjustments > Hue / Saturation ( CTRL + U ).

Set the hue and saturation value to 30. Click OK

Apply an Overlay blend mode with an opacity of around 50-60%.

Merge these into a new layer.

Step 15

Now we are going to apply some Tattoos / war paint. Create a new Layer called Tattoos.

Select a Chalk Brush at about 14px and zoom right in. Select the color you want and while applying the tattoo reduce the size of your brush as you use it ( Use the [ ] keys to do this ).

Once your happy with the application, change the blend mode to Color.

Step 16

Its at this point that I realized I didn’t like the sword that came with the stock model. So we will change it !

Open up Japanese-swords.jpg and using the pen tool, draw a path around the blade. Create a selection as we did with horse and model layer.

Cut and paste it into the main composition . Do the same for the hilt of the sword and paste them into their own layers. Name them accordingly.

Step 17

Select the Sword layer, and using the Free transform tool rotate the sword and scale it down to fit in place where the old katana used to be. Do the same with the hilt.

Create a mask for sword and hilt layers. Then paint black with a hard edged brush to mask out any overlapping out place sword or hilt.

Create a mask on the centaur layer and mask out the old sword. While also using the burn tool to create a new shadow on the chest right under the new sword, doing the same for the hilt as well.

After this we will merge the centaur one more time into a new layer ( CTRL – Alt + E ). To help clean things up we will create a folder to put all our centaur aspects in to a group.

And there is our complete Centaur. Its now time to create a background.

Step 18

Open ‘greek-ruins.jpg’ and place it behind the Centaur layer and move it to the top left hand comer of the canvas.  Name the layer Ruins.

Hide the centaur layer and using the pen tool we are going to create a path to remove the background from the background…. Yes I did say that!

Right click in the center of your path with the pen tool still selected. Click Make Selection. Then Press delete to remove the backgrounds background thus giving us room for the sky.

Step 19

Open ‘sky-4345.jpg’ and cut and paste it in to your composition. Place it below your ruins layer. Name the layer sky.

Now press CTRL + T and re size the sky layer so it fits in to your comp as shown below Making sure you leave about 1 inch underneath the ruins layer for blending.

Step 20

Select Photo Filter from the Create new fill or adjustments layer icon at the bottom of the layers palette. Make sure its above the ruins layer. Apply a Deep Red color and unclick Preserve Luminosity.

Apply a Screen Layer Blend mode with an opacity of 75% on the photo Filter Layer.

Step 21

Select the sky layer and open the hue and saturation box ( Control – U ). Apply these settings.

Add a gradient form the add a layer style menu at the bottom of the layers palette. Apply the gradient in the picture. It’s the standard orange yellow orange preset.

Step 22

Now would be a good time to check and make sure you background is where you want it be in the final image.

Select a soft edge brush with black as your color and opacity of about 12%, select the ruins later and create a mask with the Add Layer Mask button at the bottom of the layers palette. Start softening / blending the edges with the sky layer. You should hide the Centaur in this step to make it easier to see.

Step 23

Now once you are happy with the background,  it would be a good time to Crop your image into a more manageable size.

Grab the CROP tool and crop the image so the scene looks complete.

Once again merge all of the background layers into one layer ( CTRL – Alt + E ) name it Background.

Step 24

Duplicate your background layer and set a multiply blend mode with an opacity of 80% to darken the background.

Step 25

Merge the 2 background layers again ( CTRL + Alt + E ) and duplicate the layer. Set the Blend mode to Multiply with an opacity of 20% – This adds more Atmosphere to the clouds

After this merge all the background layers into one layer.

Step 26

Create a mask on the background layer, and with a soft black brush and around 15% opacity paint some highlights in the sky. Try and paint over any blues and purple.

Step 27

One last time we are going to Duplicate the background layer, and give it an overlay blend mode with a 40% opacity.

Merge the background layers.

Step 28

We are now going to create a shadow for the centaur on the ground.. Create a new layer under the centaur and name it Centaur Shadow. Grab a soft edged black brush with an opacity of 10% and slowly made a shadow.

Keep in mind where the light is coming from. In this case it’s the top left hand corner.

Once you have finished, lower the opacity to around 75% or until it looks natural.

Step 29

Now to finish the manipulation create a fill layer, select Layer > new Layer Fill > Solid Color. Click OK

Select a solid black and click ok.

Set the opacity to 75% and with a white soft edge brush start painting in the light.

Step 30

Apply a Warming photo filter above all layers with a density of 45%

Step 31

For the last step I decided to add some rays of light…

Grab the pen tool and on a new layer named heavens roughly map out a ray of light.

Right click with in the path and click make section with a feather of 5px

Fill this with white and press CTRL + D to remove the selection.

Apply Filter > Blur > Gaussian Blur with a radius of 10. Drop the opacity of the layer to 28%.

Then move the Heavens layer below the photo filter layer.

Step 32

Take this opportunity to mask in more shadows or highlights where you think necessary! Create a black boarder or add some extra background elements. Experiment to make your centaur unique.

Here is how i created this simple yet stylish image .

Step 1

First i created a new document of 500×300 pixels that I filled with white ( you can chose to create yours on a larger canvas or smaller its up to you) . Create a new layer and select the lower part of the canvas. Then Fill it with black.

Step 2

Now using brushes that i had in my collection, i painted grass and some flowers in black across the top of the black line. You can find brushes like the ones i used at deviantart.com or a similar place.

Step 3

Now create a new layer and place it under the black layer ( with the grass and flowers ). Fill the layer with white and drag the gradient below from top to bottom.

The gradient will look similar to the below picture.

Step 4

Create a new layer in between the gradient background layer and the grass layer. With the polygonal lasso tool i create rough mountains as below.

Then apply the gradient below to the mountain layer.

Step 5

Next decorate you composition with some objects, i chose a boy jumping ( i found a silhouette brush on deviantart.com ).

Step 6

Now brush in some clouds with the standard brushes that come with adobe photoshop starting with darker colors, and layering up to lighter like below in the Final image! Easy huh ? do not be scared to fool around with the colors and brushes. I would love to see what you come up with, so if you follow this tutorial or any tutorials on this website make sure you add them to the gallery under each tutorial..

This will work on any version of photoshop from 6 up. Obviously CS4 is recommended and if you haven’t tried it out you should!

Step 1

First of all ensure your back ground is set to black, it is important as the end result does vary quite considerably depending on the choice of color used. You will need to brush out a fiery shape with a white brush set to 100% opacity and 100% thickness. Having some debris flying around randomly does help add to the effect as sparks.

Try different brush setting to get interesting effects. Also having some premade brushes will help you. Search deviant art or the freebie page on simplypsd.com. But to give you a rough example, check the image below.

Step 2

The next step is still simple, duplicate you initial brushed fire shape layer, and set a new layer to overlay. Then hit Ctrl+U and check colorize using the following settings:

• Hue = 35
• Saturation = 25 (default)
• Lightness = -50

This should give you a similar effect to the image below.

Step 3

Another simple step is to just duplicate your overlay layer and set it to color burn as you can see in the image below.

Step 4

In this step we will be going over what weve done before. So follow steps 2 and steps 3 through one more time, duplicating your layer, overlay, hue, duplicate, burn, new group, merge down etc. Just follow 2 and 3 again with this new layer and all is well. Hopefully you should achieve this messy object, which is all you want really as seen in the image below. The following step will show you the finished product!

Now you could leave it there, if you want to ( merge all layers, except black layer, and set to hard light), however I continued. Since the effect can be enhanced.

Create a layer set ( group ) using the folder icon at the bottom of the layers box. In the correct order, put the three layers you have into the set. Select the set, and hit ctrl+e thus making one new layer.

Step 5

Set your final layer to hardlight and thats pretty much it, the mess vanishes, and you have yourself a nifty explosion. It works well for space effects, impact scenes etc (image below).

To give an example of how you may use it, check the image below.

The Final ImageClick To Enlarge

Step 1

The first stage which forms the starting point for most of my digital work is a sketch done on paper. This is where I assemble any unformed ideas and decide on how the composition will be structured and try and give myself a template to begin working from. In picture below you can see a pencil sketch that I scanned in which would form the basis of my final painting. You can download the tutorial files to follow along.

For this tutorial I wanted to combine a natural looking rock face that incorporated some element of architecture that had been carved into the stone.

Step 2

The next stage was to get rid of the glaring white of the image and so I traced around the drawing and on a separate layer which I named Background, I used the Gradient tool fading from a pale to a darker green. On another new layer set to Hard Light I used a dark grey and filled in the area that makes up the rock itself and then to establish the light source I simply erased parts of this layer to reveal the white underneath, see the picture below.

It is important to define the light source as this is paramount in how the viewer reads the image and interprets the forms. Rock can be a difficult subject to paint as it varies tremendously, not only in its coloration but more importantly in its structure and the types of formations it adopts. Because its surface is so uneven with so many undulations and crevices it is easy to loose sight of the overall form and get too involved with details.

In order to keep control of the painting process I find it is best to work on a larger scale to begin with and gradually work your way down to the finer detail. In other words, decide on a light source and block in the main highlights and shadows and carve out the general structure and form of your image before concentrating on the numerous incidental passages that will help add interest and detail to the final piece.

Step 3:

In Fig03 you will notice that I have added a new layer which I called Main and begun to rough in some of the areas that fall into shadow and started to shape some of the crevices. I used warm tones ranging between those seen on the left using a standard soft round airbrush with a varied diameter.

I set the opacity to between 65 and 100% and the flow to around 50% using pen pressure to control it as shown below the brush pre-sets. It does not matter too much how rough your marks are at this stage as the painting will go through a process of refinement until it is finished.

Quick and seemingly random marks are perfectly suitable and may even inadvertently suggest part of the final outcome so be bold and carefree!

Step 4:

At the moment the light areas on the rock face appear far too light. I want to create quite a smooth and reflective surface but white should only be reserved for the extreme highlights and not used across the entire picture. In order to tone it down Ive added a new layer set to multiply at 71% which I called Overlay, and with a very pale brown filled in some warm tones to soften the contrast and reduce the white to certain areas see below.

Step 5:

Because there will be a strong light source in the upper left of the picture with some very bright highlights it is only fitting that there will be some strong shadows to match. So on another layer which I named Dark areas, I used a dark brown as seen in Fig05 to accentuate the forms and add volume to the rock face.

Step 6:

The image is now starting to take shape and we can see how the eventual rock face will look. The two vertical uprights on the left and right are meant to almost form a right angle but at the moment the rock in between appears to be on a similar plane and so in order to remedy this I am going to add a Shadow layer set to Hard Light at about 50% opacity and using a dark grey, block in a large area that signifies shadows cast by the sun, see below.

At this stage we have the main elements firmly established and we can now begin the final phase which is that of refining what we already have and painting in some details across the rock face. You can either do this on separate layers, work on the existing ones or even flatten the PSD file and then continue  it is up to you. It is wise not to continually create new layers as the file will become very large and navigating through it will become a chore  just preserve the key layers if need be.

Also do not be afraid to change anything if you feel it is wrong no matter how advanced the painting is; it is never too late to re-work problem areas!

Step 7:

To the details then! You will have noticed from the image up above that I had painted in some cracks, most notably the one running up the flat face on the left. However the overall surface still looks flat and uninteresting and so needs far more fractures and undulations. I started by erasing parts of the original grey layer (part2) to create highlights exactly as before.

Knowing that the light is being cast from the top left I then traced a darker line to the left of these lines on the Dark areas layer where the shadows would reside. With the large shadow layer above all other layers it means that when erasing the grey it does not reveal a white anymore, see the image below.

Step 8:

In the picture below you will see that Ive added a lot more detail in the form of cracks and crevices as well as darkening the shadows around the top and bottom of the picture. The best approach is to be quite loose and free about the way you apply marks to begin with if they are wrong you can simply erase them (if you are erasing from a layer such as the grey one, just add in more grey to undo it).

There is no logical pattern or set of rules to follow here, just use your intuition and add detail where you see fit. This is when I use some of the initial marks made in stage 3 to suggest how and where to place the details. You can see when you compare this image with step 6 that there is now more clarity and definition and the whole image looks less sketchy.

Step 9:

The final phase of the tutorial involves improving the detail further and using a customised brush to add a subtle pattern across certain areas to break up the surface somewhat.

tarting with a default Dry brush I then opened up the brush presets and chose a suitable Dual brush and altered the Scattering settings until I ended up with a result similar to that shown below.

You could do this on separate layers if you wish and set the blending mode either to Soft light or Multiply depending on the area and desired effect it is up to you. You can see in the picture where I have applied the marks, numbered 1- 4. With a bit more work and a few tweaks I arrived at the final result as shown below.

How you wish to scale your images will determine what method you use to change their dimensions if you have a strict size that you require a document to be, then the Crop Tool is the only way forward. It might sound a little obvious, but you would be surprised at how many ‘Photoshop Professionals’ we’ve seen who create a new document, copy and paste the image in and then proceed to use the Transform function to scale down the image to fit within the doc. When the dimensions don’t fit the image. they then go on to totally distort the proportions forcing it to fit. This is clearly very amateurish, time-consuming and shows a lack of understanding of the power of our beloved digital imaging package. It’s a lot like owning a Ferrari and only using two of the six gears.

Quick crops

So with that pet peeve out of the way, we’ll continue. You can also use the Crop Tool to quickly change the size of a document if you only require either the width or height – simply enter one of the required values and crop away. This approach can be a little long winded for one image, but if you have a lot you need to get through then this will save you bundles of time.

The cropping process can be automated if you aren’t that fussed about the crop. Open the Actions palette, hit Record and use the Image Size function to reduce the height and width, then the Canvas Size to crop the image. This will maintain the proportions, but you’ll lose control over what portion of the image is lost, so it can be a bit of a gamble. Still, it’s better than the ‘Photoshop Professional’ method.

When using the Image Size function, you’re most likely to put in the required amount and then hit Enter. But there’s a little drop-down menu that’s often overlooked which contains three main options:  Bi-cubic , Bi-linear and Nearest Neighbor.

Usually you won’t have to change these and leaving it on Bicubic will do the job, but mastering them will improve the quality of your images whether you’re sampling up or down. When an image is re sampled, an interpolation method is used to assign color values to any new pixels it creates, based on the color values of existing pixels in the image. The more sophisticated the method, the more quality and detail from the original image is preserved.

The image you’re working on will determine what option needs to be employed. Bicubic is used when a smooth tonal range is required, which is why it’s set as the default – it’s perfect for most photographs. When used for blowing up images, a smooth blurring effect will appear around the edges. Bilinear is much the same as Bicubic but reduces the quality more. Again, when blowing up, blurring will occur but a stepping effect will be produced. Nearest Neighbor is best used for images with solid colors such as illustration; as the sampling process preserves hard edges, this method can result in jagged effects. This becomes apparent when distorting or scaling an image or performing multiple manipulations on a selection. The last method is the only one that doesn’t produce an anti aliased effect on the edges.

In CS there are two additional options: Bicubic Smoother and Bilinear Smoother – these are pretty straightforward and self-explanatory.

Crop The Image

The Crop Tool can be used in a variety of ways. It can constrain proportions while changing the size of an image, reduce the height or width while taking full control over the resolution, and rotate and crop at the same time

Image Resolution:

The rule of thumb with image resolution is quite simple: the more info, the better the quality of the final image. Camera phones take pictures at low resolution, so upscaling is unavoidable. Digital cameras also take at a low resolution, but the document size is bigger so quality is maintained, because when the dimensions are reduced the resolution is increased.

Testing

A fantastic way to see these methods at work is to create a new document at 100x100px and 72dpi, and draw (without using the Vector Shape Tool) a black square in the center of the document leaving a white border. Now open the Image Size function and increase the size to 2500 x 2500px and 72dpi, then hit Enter. Zoom in on the edges to see what effect this has. Do this for each of the re sampling methods to see how they work.

When using the Transform Tool to scale an image, the Bicubic method is employed to re sample the image. As mentioned, this is pretty standard, but if you know another re sampling method that produces better results, first scale the image to the desired dimensions using the Transform Tool. Make a note of its new dimensions, then transform the image in its own document using the Image Size function and correct sampling method.

Resampling

As a rule, blowing up or re sampling images up is a bad idea, because there’s never enough information in the images to maintain the quality. But in this age of camera phones and the like, the occasional re sample is unavoidable, because the quality of a printed photograph at 72dpi will appear blocky, whereas a photograph sampled up to 300dpi will look better. A good way of re sampling is to do it in steps, IE from 72dpi to 200dpi then on to 300dpi. This method gives Photoshop the opportunity to fill in smaller gaps and therefore more of the quality is maintained.

Digital cameras also work at 72dpi but they overcome the resolution problem by having a higher document size, that is also referred to as resolution, such as 2500 x 1875px and 72dpi. As the document is larger, there’s more information contained within it, so when it comes to printing, the document is reduced while the resolution is increased. For example, printing an image with these values at 6 x 4 inches will increase the resolution to over 430dpi.