Photoshop Smart Objects for Photographers – Part 2, Gradient and Selective White Balance
Welcome to Part 2 of the Smart Objects for Photographers tutorial. This is a step-by-step tutorial on processing images using smart objects, in particular, selective white balancing.
This straight out of camera image below is what we will be working with today. It is an image of Bronte Baths in Sydney, NSW Australia.
Trying out Adobe Lightroom
Lets check out our image in Lightroom. The skies are a little overexposed, but not too overexposed, as we still have enough detail in the RAW file to pull back. We use Lightroom’s gradient adjustments to reduce the exposure (darken) of the sky to obtain a bit more detail in them. If we had shot in JPEG mode, then we would not have been able to do this.
Now we need to adjust the white balance to something a little more suitable and representative of what it really was that evening. I remember the skies was an orange-red colour, and the water was very deep green-blue in colour, but the in-camera auto white balance is lying! So we need to make an adjustment.
Lets try and get the skies right. I find it really helpful to make the image really vibrant (using the vibrancy adjuetment), as it enhances all the colours, making the white balance much more sensitive. Once you find the right white balance, you can reduce your vibrancy back to something a little more tame.
The skies here now look close to what it really was. But the water is now the wrong colour. So now we adjust the WB to about 8700K and the water looks a lot better.
Lightroom’s Gradient Tools
But we have a problem, how do we do both of these adjustments, one to the sky, and one to the water or foreground. We look at Lightrooms gradient tools, which are very handy indeed.
We see can make gradient adjustments only to the following:
- color filter
But we can’t do it with white balance! (you could do it with color filters, but that’s a little different all together). The solution to this is Smart Objects in Photoshop. Lets see how we can do this.
Open your image in Photoshop as a Smart Object.
Open in Photoshop
The little icon in the bottom right of the layer thumbnail shows that it is now opened as a smart object. Double clicking this will bring up the Adobe Camera Raw dialog box, which pretty much has all the adjustments available the same as on Lightroom.
We want to make 2 smart objects. As we want to make separate white balance adjustments (and maybe some other adjustments as well, as we see fit) to each of the copies.
We now make the desired white balance adjustments on each of the individual smart objects, one for the sky, and one for the foreground. You make any other adjustments you like as well, such as exposure etc. Don’t forget to pull that vibrancy slider back if you were trying out my previous white balance adjustment tip. For my image, I’m going to name my layers to something more description as well. I always like doing this.
Now that we have made the Camera RAW adjustments to the 2 smart objects, we apply a layer mask to the top layer so that only the skies layer adjustments show through in the sky section; while the foreground adjustments only apply to the foreground. We can control how hard the transition is with our layer mask, by the rate of change in our gradient which we will explain shortly.
I apply a black to white gradient mask. The side that is black is the end which will mask away the adjustment of the skies on that layer, i.e it will show the foreground adjustment. Black conceals, white reveals. Therefore the adjustment needs to be black towards the bottom, white to the top.
Now we see we have our graduated adjustment now. Now for a few other things I love about smart objects. We can also work in LAB colour mode non destructively. And we can now soft proof and show gamut warning. Unfortunately the gamut warning and soft proofing isn’t “live”. Here are some extra adjustments, and with gamut warning turned on (proofing to FujiFlex paper).
We make a few little final adjustments, to get rid of the gamut clipping. Flatten the image, rasterise the smart objects and rotate the image a little to fix the crooked horizon.
So there you have it folks, perhaps it wasn’t the best image to do an example on, but its more of an exercise to show the advantages of using smart objects. A serious landscape would have a lot more work done it such as sharpening, dodging and burning, more careful color adjustments etc. But try it out.
When else would this be useful?
There are many situations where this type of graduated adjustment performed as smart objects could be useful. For example:
- Combined lighting situations, say with combined natural light and artificial lighting such as tungsten or fluorescent lighting. As long as as the lighting isn’t too mixed together, say for example a light in a window, we can use custom white balancing for different portions of an image.