Somewhere I don't remember, I heard about the Brenizer method.
After reading a few sites, there is no consistent information about what it is exactly, or what the procedures are to apply it, so:
I hope you can explain it to me a little more. Thanks!
It's essentially a way of using image stitching so you can use a long lens to shoot a wide-angle image.
As with a panoramic image, you shoot a number of pictures that will eventually become one image. The idea is to use a relatively long lens, like a "portrait length" telephoto or longer, so that you can get a very shallow depth of field on your main subject. You then shoot a number of images of the subject's surroundings, keeping all of the settings -- including the focus point -- the same as the original image. This is a whole lot easier to do if everything, including focus, is manual.
Ordinarily, the primary subject is a person or a couple (Brenizer is a wedding photographer), and people move, their clothing shifts, etc. You want, then, to get a very tightly-framed picture of all (or at least most) of your main subject with the first shot. Sometimes you can cut them off at the feet and fill in the rest later, but if you're shooting people "dressed up" outdoors, the chances are that dress hems and trouser cuffs will go astray. Sometimes stitching makes up the difference, but not all of the time.
The next shots will be of the surrounding area. Make sure there's a generous amount of overlap between images, and take enough pictures to completely cover the same field of view as the wide-angle lens you are trying to simulate.
The rest is post-processing: panoramic stitching and whatever retouching you may need to do to make the image seem like it was all-of-a-piece in the first place.
When you are done, you will have created an impossible picture -- the equivalent, say, of shooting the image with a 24mm f/0.2 lens. That is, you'll get the field of view that a wide-angle lens will give you, but a razor-thin depth of field that a wide-angle lens can never give you. (Well, not any wide-angle lens that can be made with the materials and knowledge of optics that we're likely to have in the next century or so.)
The optimum for ease of shooting would be one or two people and a static environment. The more moving parts there are in the scene, the harder it gets -- people will have to hold a pose anyway, and the more people there are, the less likely it will be that everyone will hold still while they're being shot. Dogs and kids? Maybe not a good idea most of the time.
What is its goal?
The main goal of the Brenizer method is to mimic the shallower depth of field you get with a medium or large format camera without using a medium or large format camera.
While larger formats do not, inherently, give you shallower DoF, the fact that larger formats yield a wider field of view tends, if you try to compose an image identically as you would with a smaller format, cause you to either get closer to your subject and/or use a longer lens, and both of those changes can reduce the depth of field.
You typically shoot this type of panorama with a telephoto or short telephoto fast portrait lens (say, a 135/2, or 70-200/2.8 on full frame, or an 85/1.8 on crop), with the aperture set near to wide-open to get the shallowest DoF possible for each member image.
Is it a merely post-processing job?
Yes, and no. The panorama stitching, certainly is a post-processing job. But given that some frames may be entirely out of focus, this also requires a bit of planning and setup on the shooting as well. You may want to use a specialized panorama head, not so much for rotating around a no-parallax point of the lens, but more to track the coverage of the scene, given the narrow angle of view that longer lenses yield.
And you certainly want to have exact focus, locked in with manual focus--and you have to have subjects that can hold still while you shoot ALL the member images. You also need to be experienced enough with panorama shooting to make sure you have enough coverage for the scene you want with no gaps, and enough coverage through time to be able to edit out ghosts or clones that can easily by caused by anything moving through the background.
Can it be applied with any lens?
It can, but the effect is unlikely to be realized without using a long/fast lenses, since wide and slow lenses tend not to exhibit out of focus blur as easily unless used at very close subject distances, and the whole point here is to get a shallower depth of field at normal or slightly farther subject distances.
Is there a preferred scenario?
This is most often used for portrait photography. Typically it's done outside with a picturesque view surrounding the subjects, and the shallow depth of field is used to emphasize the couple/subjects against the backdrop. But any time you wish you could have shallow depth of field, but your lenses, sensor format, and working distance prohibit it will pretty much work for this--as long as you have a lens and aperture combination that can achieve background blur at the distances you want to work.
