OK, so I’ve seen this posted around on a number of forums, people asking how to get photos to look like this, where it almost looks like a painting. And I always thought I should maybe write a tutorial on it. And well that time has come.
I really have no idea what the proper term is for this style. it is becoming known as the “Dave Hill Look”. Though this is similar, it is different and gives slightly different results.
Essentially what is happening is you are compressing all light into the exposed range. In the real world, there is a lot of visible light. And a lot more non-visible. Now just because it is visible doesn’t mean it is all visible at the same time. Especially to a camera. What this means is that when you look at something dark, anything that is bright is over-exposed or blown out and the details cannot be seen. Likewise, if you are looking at something bright, you cannot see the details in dark things. There is a range of light that can be seen at any one time.
Eyes, film cameras and digital cameras all work the same way, in that they have an iris that opens and closes to let in or block out light which allows for a certain range of light to be captured. Anything darker than that range will be under-exposed and probably just look black. Anything brighter will be over-exposed and look white (and will probably add glows and halos around its edges along with all manner of other issues I won’t be going into here).
However when an artist paints a scene, very rarely will she or he paint areas under or overexposed. Have you ever seen a painting where the sky is all washed out and white? Ok so now some smart-arse is going to find one and email it to me with an abusive message. I don’t care. And if that’s you “Get a life!”.
What the artist is doing is looking at each area of the scene separately and then painting them with visible colours. And this is what we need to do with the photos.
The first step is to take your photo. If you know you are going to want to apply this look, it really is better to know before you do anything. Ok, so not a rule but a suggestion USE A TRIPOD! Now the idea is to capture ALL visible light. Unless you just started reading here you’re probably thinking “didn’t he just say a camera can’t capture all available light…” and you are correct. That is why it is best to take several photos at different exposures. I usually take one at what would be considered the correct exposure and then one 2 f-stops above and below and also 4 f-stops above and below. I’m not here to teach you about f-stops so read your camera manual if you don’t know what they are.
Now, this can vary depending on what you are shooting. When the sun is in shot, I’ve had to close my iris right down and increase the shutter speed to 2000th/sec to get it right. But that’s getting too technical. All I’m saying is experiment to get the best results. Once you have done a few you’ll get the idea.
Sometimes it just isn’t possible to take multiple photos. Especially when something in the frame is moving. This is where camera RAW is a wonderful thing. A lot of digital cameras these days can shoot in RAW format and I would recommend always using it. Essentially what it is doing is recording all the data that hits the microchip in the camera whereas the JPEG format will discard a lot of information in its compression. This is ok if all you want to do is take happy-snaps and print them out at your local Kodak self-help kiosk. But if you want to manipulate your photos in any way, the more information you start with the better the end result will be. Even though RAW format still only captures the light the iris allows through, it has a much better range than a JPEG image but nowhere near as much as taking multiple images at different exposures. However, sometimes it’s all ya got.
I’ll write the rest of this tutorial based on only one photo being taken in camera RAW format. It should be easy enough to adapt to other scenarios.
Ok, I know what you’re thinking… “That’s a whole lotta writing and not many pictures. Isn’t this supposed to be a Photoshop tutorial?”
So let’s start with the original photo. It’s kind of interesting but dull. The sky is blown out although there is a bit of detail there. The dark trees and ground are all kind of murky. I can see detail in the paint but it really doesn’t stand out that much. It’s just a regular photo.
Now because I took this in RAW format there is actually more information than is actually visible. So the first step is to open it up in Adobe Camera RAW which is part of Photoshop. Just open it like you would any other file and Camera RAW should launch automatically.
What we want to do is create several files at different exposure settings to make as much of the light or data captured, visible. In your camera RAW settings, all you need to touch is the ‘Exposure” setting. First set it to 0 and then hit the save button. Create a folder to save all these files in and save it as a PSD or Photoshop file with the name and then 0.
Now set the exposure to +2 and save again, this time with +2 and the end of the name. And now repeat once more for -2. This is all I needed for this image but some may need a + and – 4 as well.
Now we have these images we have to group them together into one image with the full range of light data. This is called an HDR image or High Dynamic Range. This can be done inside of Photoshop by opening all your images and going to File menu, Automate, Merge to HDR. This works but not well enough for the next step.
OK, so this is the step where everyone will probably hate me. I’m going to step outside of Photoshop.
I know, I know it’s outrageous but it has to be done. Photoshop is good for a lot of things but this ain’t one of them. I’m going to use a program called Photomatix Pro, which can be found at http://www.hdrsoft.com/ It is specifically designed for creating high-quality HDR images and toning them. I’m sure all of this can be done within Photoshop but I prefer this method, as it is a lot simpler.
I open all the images in Photomatix Pro and go to the HDRI menu and select Generate HDR. Then I made sure the ‘Use already opened images’ box was ticked and chose OK. This then takes you to another screen where you’ll see why you included the exposure settings in your file name. Set the Exposure Value to be the same as that stated in your file name. It usually does a pretty good job of this automatically but it’s a good idea to check. Once done press OK.
On the next screen, I usually use the default ‘Use standard response curve’ and press OK. But you may want to experiment a little. This will then take a moment to create your HDR image.
and…..
waiting…..
Da daaaaaaa…..
OMG, It looks horrible what have you done!
Well, of course it looks horrible you didn’t think it was going to be that easy did you?
OK, what we have created is an HDR image. This means there is way more information within it than can actually be seen or shown at any one time. What we now need to do is bring all of that information within a range that it is all visible at once just like a painting. This is called Tone Mapping.
Go back to the HDRI menu and select Tone Mapping. A window will open with a whole new bunch of toys… Err um tools to use. This is where you need to experiment. I’m not going to go into each setting and what it does cause I’ll be typing for the next six months and then there’ll be a new version with different tools and I’ll have to start again and it’ll all be bad. So have a play and see what happens. But this image shows the settings I used for this image.
Once done click OK and your image will be generated.
Now you can see all the light that the camera took in one visible image. It’s looking better but still doesn’t quite fit the mood that I wanted. Let’s save it by going to the File menu, Save As and I usually save as a TIFF file.
OK Back to Photoshop. I knew you were uncomfortable being away from it for so long. Open your new tone mapped image up. The first thing I do is duplicate the background and label it. I like to label each layer and generally keep my projects organised so I can go back to it any time and know what it is I’ve done. I labelled this layer B&W. I’ll give you one guess why.
We want to make the layer black and white. Make sure the layer is selected and go to the Image menu, Adjustments, Desaturate. Ok now duplicate that layer twice and place all three Black and white layers in their own Group by selecting all three and dragging them down onto the folder icon at the bottom of the layers window.
First, turn the top two layers off and change the blend mode of the bottom one (which should be called “B&W”) to Overlay. This will make it very rich and dark but don’t panic. With the second layer, labelled “B&W copy” set the blend mode to Screen. I thought this was a touch too bright so I lowered the opacity to about 90%. This has made the image a touch brighter and more contrasty.
With the top layer, better known as “B&W copy 2” I’m going to bring out some of the fine detail by sharpening it. Go to the Filter menu, Sharpen, Unsharp Mask. In the dialogue box that pops open, I put the following settings.
This will look way too sharp but that’s ok. I lowered the opacity to about 53% and kept the blend mode as normal.
I should mention that the size of the image I am working on is 3072×2048 pixels roughly a 6.3 Megapixel image. I say that as sharpen and blur settings can vary depending on the size of the image you are working on.
Now that’s beginning to look better. Although the effect of those three layers is probably a bit too strong. This is the beauty of placing objects in groups. Instead of changing the opacity of all three layers, I can drop it just for the group layer. I put it at around 62%. Now you should have something looking a little like this.
It was at this point that I noticed the clouds were looking a little sharp so I added a layer mask to “B&W copy 2” so that they would stay soft. There are a million tutorials out there on selections and layer masks, so I won’t go into it here.
To my eye, there’s still too much colour so I added an adjustment layer and selected Hue/Saturation. In the window that opens I turned the Saturation down to around -35%. Much better.
The idea behind this photo was to draw people’s attention to the car, which you’d probably think I have achieved by the way it takes up most of the frame. But I’m still finding the surrounding trees a little distracting. To fix this I’m going to get rid of most of the colour and leave the blue to stand out.
To maintain the detail of what we’ve done so far I want to copy all the layers into one new layer. You can do this by selecting the top layer, which should be “Hue/Saturation 1” and press Command+Option+Shift+E on a Mac or Control+Alt+Shift+E on a PC. You may need two hands for this one. Now rename this new layer to “Tint – Grey”. And you guessed it we’re going to make it black and white again. So head up to the Image menu, Adjustments, Desaturate.
Of course, this will make everything greyscale. To get the blue to be visible through it we’ll need to double click on the layer to bring up the Layer Style palette. In the blending options, you’ll notice the section called Blend If. What we want to do is see through this layer, wherever there is blue beneath it.
We do this by selecting the blue channel and then adjusting the slider for Underlying Layer. You’ll notice that as you move the slider it creates a hard edge where it is seeing through. To soften this, hold down the Option key for Mac or Alt key for PC and drag one side of the little arrow. You’ll see that it splits into two giving you the ability to select over a range. The settings I have here gave a nice look.
OK now to add that sepia tone. Duplicate the layer “Tint – Grey” and rename it to “Tint – Sepia”. To change it to a sepia colour open up the Image menu, Adjustments, Hue/Saturation and apply the following settings.
To adjust the tint slightly I opened up the layer styles palette and changed the Underlying Settings of the Blue channel from 33/232 to 50/235. This is subtle but it makes the sepia blend into the blue slightly.
The final step for this image was to create a vignette. I did this by creating a new layer and naming it “vignette” then with the circular marquee tool I held the Shift key to constrain the shape and drew a circle a little smaller than the image.
Then filled it with black and deselected. Because the image isn’t a perfect square I pressed Command or Control + T to transform the circle and stretched its width out and pressed return to commit.
I then selected Filter menu, Blur, Gaussian Blur and pushed it all the way up to 250. Again this will vary depending on your image size. I then selected the blurred circle by holding down Command or Control and clicking on the thumbnail of the layer. By clicking on the eye next to the thumbnail I hid the layer and then created a new one. I then went to the Select menu, Inverse to invert the selection and then filled it with black. I lowered the Opacity to around 70% and hey presto, you’re done!
2 Responses
Interesting tutorial, thanks. A couple quick questions for you:
Thanks!
Such as a good tutorial on how to make that painted, contrasty, photo looky thing to do.