Shooting with a shallow depth of field has become so popular in the last few years that it’s almost become a cliché. But it remains something that’s very much in demand. Fast f/1.4 and f/1.2 lenses can be extremely expensive, though, and so very difficult for people to achieve with their f/slow-slower kit lenses, especially if working with APS-C or Micro Four Thirds sensors.
In this video, Unmesh Dinda from PiXImperfect shows us a way to simulate shallow depth of field in Photoshop using the Iris Blur filter on a smart object, with a neat tip to offer a lot more control over the Iris Blur filter than you might’ve realised it offered.
Unmesh shows two different examples in the video to demonstrate the technique. The first is a fairly basic introduction to the process, and the second is a little more complex with body parts overlapping other parts further back. And while both of the images shown were shot in the studio, you can apply these principles to location shots, too.
Watch the video above for the full technique, but here’s a basic overview.
The first step to making it look realistic is to figure out your plane of focus. Unmesh goes into some about exactly what this is, for those who’ve not come across the term before, but it’s essentially everything that’s roughly the same distance from the lens. You can’t just focus on what you want to be in focus and ignore the other things that would be in focus had you actually shot it with a wide aperture lens.
The areas highlighted in red are on the same plane of focus. These are the areas that should remain sharp in the final image.
After figuring out which areas need to stay sharp, Unmesh then duplicates the background layer and converts the duplicate into a smart object. This allows him to apply filters and then go back in and adjust filters after he’s applied them. He first applies the Iris blur to the face. It works by keeping the center part sharp, and then blurring it out as it gets further away from the centre, defined by four inner markers and an outside ring.
The inner points define where the blur begins. Everything inside them remains sharp, but everything outside of them starts to blur. The outer ring is the point of maximum blur. But the feature Unmesh mentions here is that you can actually hold Alt (or Option on the mac) and click on the four points to customise their individual position. You don’t have to have it be a perfect circle or oval. So, you can redefine it to the shape that you need. In this case, the subject’s face.
This lets you get a lot of control over how the blur fades in as it gets away from the in-focus areas as not everywhere will fall off at the same rate from the plane of focus. The hair on top of the head, for example, will often fall back out of the plane of focus much more quickly than the hair on the sides that is hanging almost straight down.
The results of this process are very effective, even in the short time Unmesh takes to achieve them in this video. With some practise and time, you can create some very realistic looks.
So, if you’ve been struggling with slow kit lenses or small sensors and can’t quite get the shallow depth of field look that you’re after, have a watch of the video and get practising. It may not get you identical results to actually using a wide aperture lens on a full frame camera, but it’ll be quite convincing if you do it well.
