Back in 2008, an extraordinarily talented photojournalist and wedding photographer named Ryan Brenizer was off traveling and wanted to create a landscape shot while isolating some of the elements in the scene, the way one can with the shallow depth of field offered by a fast portrait lens. He figured that if he used his portrait lens and manually plugged in his settings and focus, then shot pieces of the scene wide open, he could stitch them together and get the look he wanted. And it worked.
The technique, now known as the Brenizer method (or bokeh panorama or bokehrama), provides a wide-angle image with an extremely shallow depth of field, isolating the subject from the background in a dramatic way. It basically creates the look one can get with a medium format camera but it can be accomplished using a regular old portrait lens on any old camera. The finished photograph can look like it was shot with a 20mm lens at f/0.6, or maybe an 18mm lens at f/0.7 -- in other words, it looks like it was shot with a lens that doesn't actually exist anywhere in the world. The technique has become popular among photographers mostly for wedding and engagement portraits, and other artistic and unique portraits of human subjects.
I found out about the technique randomly about two weeks ago and did a search for its use. Interestingly, I haven't really seen it used in animal photography or even pet portrait photography. In the past I've seen some stitched images in wildlife shots (some that are extremely good and award-winning) but these are only two or three images stitched together -- a far cry from this purposeful panoramic stitching technique that can use as many as 50 or 60 images to capture the scene. My only guess on why it hasn't been used much in professional animal photography is because it's helpful to have a subject that holds relatively still in order to capture shots to overlap in post-processing, and that can be difficult to get with an animal. Difficult... but not impossible.
I decided to test it out and gather up some tricks for pulling it off with pets and hopefully encourage pet photographers to try out this fun and interesting technique. Plus it is useful to have in our hip pockets for photo shoots when we want to get a shot with a little extra oomph. Any dog with a solid stay command -- or any calm cat, bird, horse and many other pets -- can potentially be a great subject for photos using this technique. And the results can be absolutely stunning. It is a great tool for photographers who want to minimize how many lenses they use on a shoot, but maximize the variety of images they create.
To try out the technique and figure out some best practices, I created three images with my own dog. I quickly became intrigued with it, so I roped in some friends to let me use their dogs as models to test it out more. I'm now entirely smitten with the technique but admit that it is certainly something that takes time in post-processing while you get used to it, and has a learning curve for really nailing it. Let me say now, I don't claim to be any sort of expert in the technique (yet) -- I've only tried it out for a week. But there are some strategies specific to working with pets that I think will be helpful for anyone wanting to use it with animals, and I cover everything I've learned so far. At the end of this tutorial, I link to excellent resources that will help you master using it. So here we go.
What you'll need:
- A location where elements in the background will provide the impact for the depth of field. Having foreground objects as well as and background objects is a great way to provide extra contrast. Locations such as like a trail or path, trees in a forest or park, crop rows, city streets with buildings in the background, or even a living room with furniture will all work.
- An animal that will hold a sit-stay, down-stay or stand-stay in one place 10-15 seconds or longer.
- Light that isn't changing. You want your exposure to be the same in all frames so you don't want clouds moving across the sun as you shoot, trees blowing in the wind and shifting the light on your subject, and similar issues.
- A lens with manual focusing or a camera with an autofocus lock button.
- A portrait lens that is 50mm or longer, with an aperture of f/1.8 or faster. An 85mm f/1.2 lens works really well, as does a 70-200mm f/2.8 lens zoomed out to 200mm. I used a 50mm f/1.4 lens for all images in this article.
Steps for taking the image:
The most important step for getting started on your shot is composing the image in your mind. You want to know where the corners of your image will be so that you know how far around your subject to shoot. It's easier than you think to not shoot enough of the background and you end up stitching together a photo with a whole corner missing. Even if it means bringing along an 8x10 matting to hold up so you can more easily visualize your composition, take the time to know where the four corners of your final image will be before beginning.
Select your manual settings. Set your aperture at wide open or just one or two stops down. The wider the better, but be sure it's an aperture where your camera is sharpest, which is often one or two stops down from wide open anyway. If you're using a telephoto zoom, such as a 70-200mm f/2.8, use it fully zoomed out for the most minimal depth of field. Then select your ISO and shutter speed for the best exposure.
Maintaining an even white balance is also important. You can do manual white balance on location, but I've found it easier to just sync the images with the right white balance in Lightroom during post-processing.
Get your focus using either manual focus, or dialing in with autofocus and pressing the autofocus lock button on your camera if you have one, and holding it down as you shoot.
The distance between you and your subject is up to you, depending on the composition and the effect you want. I've found I get the best results when my subject's face fills a good portion of the frame.
The easiest way to capture images is, as Brenizer says, to start with the moving parts first, which means starting with your subject. So, once all your settings are complete and you're ready to shoot, lock in your focus and shoot for the expression you want in your final image. It's okay if this means taking a handful of photos of your subject until you have just that perfect look for your final image.
After nailing that perfect eye contact or head tilt, start to move around your subject, capturing the scene in a systematic way moving outward so you know you have everything. I found it easiest to start with my subject's face, then move to just above the subject's head and begin a clockwise expanding spiral, overlapping the images by 30-50 percent.
While you don't need to overdo it with the number of frames you take, be sure to get plenty of overlap -- it would be frustrating to find out in post-processing that you're missing a portion of the scene that you can't stitch, and you have a blank spot in your image that may not be fixable. For the image example above, I visualized where I wanted my edges to be, began shooting the frames, and went a little bit beyond where I wanted to end just so that I was sure to be able to get the crop I wanted after stitching.
If the animal moves their head or feet around a bit while you're photographing the overlapping parts of their body, don't worry about it. It doesn't necessarily mean you have to start again. Just know you'll probably have to manually blend the layers rather than using Photomerge to be sure you get all the right pieces in the frames for their natural body. On the other hand, if the animal completely shifts position while shooting their body, well, there's no fixing that. You'll want to start over. However, if you already have all you need of the subject and you've begun working to capture then scene, and the animal moves or even gets up and leaves the shot, then just make sure they're fully out of your scene and keep shooting what you need of the edges. They wouldn't be in these frames anyway so it's no big deal.
Again, I can't stress enough to remember to photograph the full scene, especially the top corners. You'll want to crop down the image later and you need to have enough material to work with. The image below was my third try at the technique and realized after stitching with Photomerge how much of the scene I didn't capture when I thought I did. It's not that I needed it for the final crop that I wanted, but I thought I shot it and didn't. What if this happened and the image was for a paying client with expectations? So, consider your edges a safety net.
Shooting plenty around the borders of your composition also simply gives you choices later on for how much of the scene to leave in and how much to crop in to your subject depending on the look you want. For example, here is a complete image image on the left, using 29 images and getting the full scene as I envisioned it on location. On the right is a cropped-in version that uses only 9 frames. Because I shot more than enough, I was able to decide in post processing how much of the scene I really wanted after all. That kind of flexibility is invaluable. Think of it like writing: It's easier to edit down a too-long first draft than to build up a too-short first draft.
The effect works best when close to your subject, filling the frame with their face. But you don't have to strictly follow this guideline. If it is easier for you to be at a distance from your subject, or if you just want to get more of an environment shot, stand back as far as you want. Just know that the dramatic isolation of your subject that this technique is famous for happens best when you are close. Here is an example of when I stood about 15 feet back from my subject.
Another thing to note is that you don't have to take 50 or 60 images. Start small with 9 or 12 images as you get the hang of the process. My first attempt was 28 frames, then 39, the 73. I was a little over-exuberant and learned after a bit that you can often get the same interesting effect with just a dozen or so images while cutting down your work time significantly. The more of the scene you want to capture, the more images you'll take, but the more time and effort you will have to spend mucking around in Photoshop. So start small, get a handle on it, and expand from there.
Steps for readying your images for stitching:
Once you have all your images, the real work starts. There are many ways to do this and a lot of software options you can utilize including programs dedicated to stitching panoramic photos. Here's just one option for how to get your final image.
Put all the frames you'd like to use in a single folder and upload the folder to Lightroom. Select the first image. In the Develop panel, check the "Enable Profile Corrections" box in the Lens Corrections tab. Make sure that it is recognizing the lens you used. Also, adjust the vignetting slider to remove any vignetting. For my Canon 50mm f/1.4 lens, I usually slide this to about 120 or so.
Then go up to Basic adjustments and adjust your white balance. These are about the only adjustments you'll need to make at this stage, though if you need to adjust exposure or something else, go for it. Just don't worry about in-depth adjustments quite yet.
Select all the images in the folder and hit the "Sync" button at the bottom of the Develop panel. This will apply your adjustments to every image in the folder.
I like to export these images at the highest quality so I can use the final stitched image for large prints, but I also need images small enough to work with in Photoshop without making my computer slow to a crawl. The happy medium for file size for me seems to be to export them as jpgs in AdobeRGB at 300ppi and 100 percent quality, and resizing them to be 1500 pixels on the long edge. Then I export the batch to a new folder.
Next up, Photoshop.
Steps for stitching:
Stitching can be a breeze, but often it isn't. Photomerge is an incredible tool in Photoshop and I adore it. But that said, it's not perfect. There will most likely be layers that don't blend quite right and you'll be going in to do some clean up. The amount of clean up needed can make Photomerge your best friend, or completely not worth it. So I'll go over steps for automating but also steps for manually blending images.
To use Photomerge, select File -> Automate -> Photomerge.
In the Photomerge window, select the "Reposition" option. Add the images you'd like to use in the panoramic. Then click OK.
Photoshop will then start working on automatically layering your images based on content, and then blending your images together. The results can be, um, mixed. Sometimes it works like a dream, and sometimes there are some wonky errors. The errors happen especially when you have multiple images of the subject when the subject has moved positions (so the legs, ears, face and so on don't align), and when you have straight lines such as trees or railings.
For example, in the image below, Photomerge worked almost perfectly. There are only a couple places where the tree trunks don't align. I can simply go in and select the specific layer or layers that are showing for these areas and either adjust the masking on them, or delete them and add the frame I want to use for that area as a layer and manually blend it to correct the problem.
But sometimes, Photomerge just looks janky. Below is an image that that was blended together and makes little sense. There's an entire chunk missing from it because the images weren't blended quite as intuitively as one might hope. When something like this results, your best option is to manually add each layer, mask, blend, and add the next layer. While this is certainly more time consuming, it gives you complete control over what is shown from each layer and you can make the scene look exactly how it is supposed to look. I trashed the Photomerge file below and started afresh, and my result was the image up toward the top of this post, an image with which I'm extremely happy. It took awhile but it was worth every minute.
To use manual blending, all you need to do is create a new canvas (depending on how many 1000x1500px frames I'm using, I start out with a canvas between 4,000 and 8,000px square) and open your exported images. Drag and drop you first image, the one of the subject's face with the perfect expression. Then drag and drop the next frame of your subject, the ears or whatever it may be. Mask and blend that, and continue to build out from there, dragging the next frame, masking, blending, and dragging in the next frame. I layer in the same expanding clockwise circle in which I shot to make things easy.
Like I said, it is time consuming but you have total control of your image and frankly, sometimes it's the only way to get a decent result. For instance, the photo at the very top of this post was virtually impossible in Photomerge -- there's too much going on with the rocks at different angles and different depths, the trees, the twigs that are in focus on the right of the subject and so on. The only way to get everything to align properly was to do it by hand, frame by frame. But when I finished it, printed it and handed it to the dog's owner, the smile on her face was worth every minute spent squinting at the screen.
Once you have your final image completed in Photoshop, save it as a psd (because you'll inevitably find little mistakes and micro-adjustments later on that you'll want to fix), and then save it as a jpg. I then import the jpg to Lightroom and make my adjustments for print, deepening the blacks, bumping up the vibrance or whatever is needed to make the complete whole. And there you have it: a unique photograph with a stunning look that you couldn't have captured with any lens available on the market.
I'm excited to get out and test this technique not only with more pets but also with wildlife. It will be interesting to start with calm elk or deer that will stay put for at least a couple seconds while I'm relatively close and see how it goes. And of course getting more practice and learning more tricks for when this technique is used to its best effect, and how to make post-processing more fluid, will be fun.
Ultimately, I hope it will be a technique that I can use every so often when I come across the perfect scene and subject that will benefit from this super-shallow-depth-of-field-wide-angle goodness that Ryan Brenizer was kind enough to bring to the world's attention.
While I've included all I know up to this point -- and I will continue to add more as I experiment and have advice to share -- you can learn so much more about perfecting this technique by watching this video. And there are of course about a bazillion articles on how to pull off the technique and you can learn countless tips and tricks from the internet at large. I found this video from Adorama and this tutorial from PhotographyLife helpful in getting started. There is also a handy dandy calculator for figuring out the lens you "used" for your shot. It not only is interesting but also gets you really excited about the notion that you just shot a photo with a lens that doesn't even exist.
If you try out the Brenizer method (aka bokeh panoramic, aka bokehrama) in your animal photography, please share your work in the comments! I'd love to see examples of it used for pets and wildlife, and learn more from you!