In portraiture and fashion photography many photographers like the extremely shallow depth of field that comes from shooting at very wide apertures, such as f/2.8 or even more extreme wide apertures like f/1.8 or f/1.4. It can be a dramatic look, as the subject is the only thing in focus and really jumps out from the out of focus background. But the look and technique are not without some drawbacks.
Here’s an image shot at f/1.8. Notice the very shallow depth of field. One eye is sharp. The other is slightly soft. And even the tip of the nose and the hair id out of focus. It’s a great look, but hard to shoot.
The shot below is at f/2.8 and also the face is more parallel to the plane of the sensor, so both eyes are in sharp focus. Notice that at f/2.8 and close focusing depth of field is still very shallow.
The first problem is the cost of those very fast (wide aperture) lenses. A Canon 200mm f/2.8 lens costs about $650 and a Canon 200mm f/2.0 will cost about $6000 if you can even find one. That’s ten times more expensive for just a 1 stop faster lens. Maybe that’s OK if you are shooting indoor sports at the Olympics with a major magazine footing the bills. A Canon 85mm f/1.8 is around $350, while the 85mm f/1.2 is $1400. Again that’s four times more for only a 1 stop gain.
The second problem is the extremely shallow depth of field at those wide open apertures. Yes, that’s the look, but there are also problems inherent in that shallow depth of field. In tight portraits, it’s not unusual for one eye to be in focus while the other eye isn’t. Or the eyes are in focus, but the lips aren’t. And when depth of field is so shallow, just breathing, either by you or your model can cause the image to be out of focus. Shooting tight portraits wide open tends to produce a lot of rejected images where the focus just isn’t quite where it needs to be.
The third problem is if something isn’t in focus, but you wish it was, there is not much you can do about it.
So I’m going to show you a way to fake that shallow depth of field look, while shooting your original image at a smaller (and cheaper) aperture and adding the shallow depth of field by blurring parts of the image. I’m going to show you how to do this in Photoshop, but you can do a similar effect in Lightroom, but you don’t have as much control.
This is the image we’ll start with. It was shot at f/11 in late afternoon sunlight. Notice how sharp the road and hills behind her are. Either open the image in Photoshop, or send it to edit Photoshop from Lightroom.
Create a duplicate of the background layer with Ctrl-J (Cmd-J for you Apple types). Select the subject that you want to remain sharp. It doesn’t have to be a super precise selection for most images, and here quick select did a perfectly fine job.
With the subject selected, use Ctrl-J to put just the subject on a new layer. With that new layer selected, go to selection/load selection/ from transparency.
With that selection in place, select the middle layer and then delete to remove the model on the middle layer. At this point, nothing looks different because we still have a copy of the model.
With the middle layer selected and the selection of the model still in place, go to edit/fill/content aware fill. What we’re doing is getting rid of the model so that when we blur this layer there won’t be an image of the model that will blur and create a halo since blurring tends to expand the image.
Here’s the result of the content aware delete. Ctrl-D to deselect.
Make sure the middle layer is selected, and go to filter/blur/Gaussian blur and select a fairly strong blur. Here I used 38, but we’ll adjust the final effect that adjusting the opacity of the blurred layer. Rather than Gaussian blur you can also use blur/blur gallery/field blur which will give more options and make fake lens blur a bit better.
Adjust the blur to taste by changing the opacity of the blurred layer. When it looks good to you, go to layer/flatten image and save the image.
Here is the final image with the blur effect.
If the deleting the model and content aware steps seem like useless extra work or you don’t understand the importance, here is the image if we don’t delete and fill the model. Notice that when the model is on the blurred layer, she blurs outward and creates a halo around the sharp image.
The exact same procedure can be used on other subjects like products or architecture.
