Panoramic photography

Panoramic photography gives the photographer more creative options when creating an image. Sometimes elongated feels better than rectangle or square. I am not really talking about gigapixel images that allow you to zoom in on a town square or an office window from the top of another building. I am thinking purely from a creative point of view, the compositional aspects of panoramic photographs rather than the billions of pixels aspect. I am also talking about digital photography and seeing the final image in the minds eye before pressing the shutter. And that means multiple images that are stitched together later in Photoshop,  or Hugin or something like that.

Panoramic photography in a way harks back to the so called ‘good old days’, in that if you take multiple images to take a panoramic photograph using a digital compact or DSLR, then you won’t be able to see the final image until you are back at the computer. That process will force you to slow down and really think about your composition. 

This was a composition that sort of jumped out in panoramic form. The London Underground sign is situated on a very narrow rounded corner with the two streets either side invisible to each other.

Temple Station, London.  This composition popped out of nowhere and screamed panorama...

If I was to use a conventional rectangle or square format the composition would be different.But this is what I saw in my minds eye, before I started thinking about the panoramic format. It was taken on a Saturday so (unfortunately) there are not many people around. What I would have loved would have been someone leaning against the red telephone box talking on a mobile phone. But that never happened! This is a composition where it is necessary to put the subject in the middle of the frame and create a symmetry with each street either side of the underground sign.

The composition of a panoramic photograph is no different from any other photograph whether square format or rectangular. The big difference are the number of possibilities that are available when taking panoramic photographs. And remember, with digital photography, regardless of the pixel count and format of your cameras sensor, you can create any format of photograph at any resolution if you are prepared to use multiple images stitched together. At the risk of making huge generalisations, with a square format it is more common to put major elements in or near the center of the frame. With rectangular format the choices open up a bit to allow more room for the off-centering of major elements, whilst allowing the image to fan out sideways and beyond of the major elements. With panoramics there are other creative possibilities. For example you can center a tree or the corner of a building and show different things happening either side of the main element. Or you can emphasize the size of an object using objects of different sizes (trees/people for example) coupled with a long vertical panorama.

You should still feel inspired to take a panoramic photograph and you should (eventually) see panoramic compositions intuitively. There is nothing wrong with cropping in photoshop at home on the computer, and you can use that process as part of the learning curve, but it’s challenging and rewarding to see unique compositions whilst you are out taking photographs, and then have a go at creating them as panoramics back home, after taking mulitple images.

This forest image was a panoramic composition that presented itself very clearly to me. I initially saw the root system of the tree when walking along a public footpath. In front of the main tree is a large crater shaped dip that I am standing in, it’s as if a tree had fallen long ago and then removed, leaving a crater that had been weathered smooth over the years. That meant that I was lower than the surrounding ground level, as well as being on a steep hill.

Oldbury Hill, Kent, UK

I noticed that the trees spilled down the side of the hill and that the line that separated the brown earth and the green shadowy forest created a nice, diagonal graphic element that drew the eye from the footpath on the left to the dark shadows of the forest as the trees spilled down the hillside. The subject is the exposed root system of the tree, the rest is the environment in which the tree is situated. Had I used a more rectangular frame, I would have had much more of the trees in the frame, showing off their height much more with the increasing brightness from the sky above having much more emphasis. That is not to say that a rectangular or square frame would be unsuitable, but it would have created a different feeling and would have emphasised different aspects of the forest.

multiple digital panoramic images can take any shape and size, you are not confined by format anymore. You have more flexibility than with any other format of photography, and you choose your resolution. But now it’s time to talk a bit about how to frame the shot.

First you see something that you think will make a good panoramic photograph, you need to stop and look at the potential composition and roughly work out the shape of the frame and where the edges will be. This means working out what to exclude from the frame and what to include. What I then do is take a point at the farthest left edge that I want included in the composition and go a little beyond that with my framing. Then I look right and find the right most element in the frame, I go a little beyond that, and that is my right edge. Then I need to work out how high the image will be, in other words, where is the top of the frame, and where is the bottom of the frame. I do all of that with sight only, I don’t look through the viewfinder, but in my minds eye I will now know where the edges of the frame will be. Sometimes I will hold my arms out and use my hands as the edges of the frame to get some idea as to what the overall composition will look like.

Then, handholding the camera I look through the viewfinder to the left edge that I want included in the frame and position the the viewfinder roughly where each frame will be (without taking the photo), going left to right for each row that I need to achieve the desired height. Overlap by at least 30 percent. You need to overlap by a significant amount so that the stitching software can join everything up.

After all that, I still haven’t taken the photos yet. If you have a zoom lens where you can adjust the focal length you can determine more precisely the final resolution, by either zooming in and taking more individual photos, which will create a higher resolution image or you can zoom out, go a bit wider and take fewer individual images which will create a lower resolution image. The ‘practice run’ that I described above can be used to test different focal lengths if you do have a zoom. Using a zoom lens, and after becoming used to the technique described above,  I find it quite easy to create a 20,000 pixel image (longest edge) or, for the same composition take a 40,000 pixel image. For the most part though, I take multiple image panoramics for reasons of composition and rarely go above 25,000 pixels along the longest edge, as this is a convenient size for sorting the individual images out on the computer as well as processing time and final image size.

These are the single images that I took to create a panorama. I had already worked out the composition before taking any of these photos. You should be able to see in this example that all the photos overlap, and that there are two rows. Also, all the photos are horizontal/landscape format. I find positioning the camera in a portrait/vertical position is awkward and cumbersome, So I always construct my panoramics from a series of horizontal format images even if I am creating a vertical shaped panoramic.

I saw a lone tree in the distance set against frost covered trees.  Once I had carefully worked out mu composition I photographed each of the images necessary to make the panoramic

Next I got the photos home, put them in a folder on the computer and used the free software program Hugin to stitch the images together. Hugin is as good as any other software package that I have ever used. Before you pay good money for ‘pro’ software, check out Hugin first. Also Hugin is better than the photostitching software supplied with Photoshop. Photoshop struggles to join lots of intricate straight lines accurately, especially if you are using a wide angle lens, and bare branches of forests and intricate brickwork can also be a problem. All stitching software will occasionally struggle with joining up certain elements, and Hugin is no different, and if that happens there is no alternative but to join up the bits and pieces manually in Photoshop or whatever photo software you use.

Once the panoramic has been stitched you can then crop it down to the image you visualised at the time.  With this image I didn

This picture was created during a walk in the forest and seeing a lone tree in the distance (it’s difficult to see at this size, but the tree that I saw is in the center of this frame beyond the two foreground trees. Click the image for a larger version).  Most of the trees had branches covered in frost but this one didn’t, but was set against against trees that did. I noticed some lone trees in the foreground and walking under them I noticed their bare branches filled the boring grey sky. The branches from two of the trees seemed to reach over towards each other and if I positioned myself appropriately I could make the branches frame the lone distant tree. In my minds eye, the top half of the frame was filled with these branches, but not the tree trunks they were connected to. So I set about taking the individual images, photographing beyond my initial visualised composition to give myself some wriggle room back in photoshop.

So once I had the big panoramic formed from the individual images, I just need to crop away the uneven edges and do some post-processing. The tree in the middle distance is the subject of the image, and I used the panoramic format to place it in context in it's environmentAll I did to this image was to adjust contrast and brightness, and I also added a vignette just to push the viewers eye into the frame. For the final cropping, I went left and right as close to the tree trunks as possible without actually showing any of the tree trunk. I really liked the graphic effect of the branches, but didn’t feel that the whole tree added to the composition.

There are some other technical things that you need to do during this process. Like how to set the camera up prior to taking the images that will make the panorama. I usually use a tripod for nearly all of my landscape photography, and that’s how I take panoramics too. But you don’t need to do that, it is easy to take panoramics without using a tripod. And there is no need to have a ‘panoramic tripod head’. I am not going to explain what the ‘parallax effect’ is, Wikipedia does that much better than I can, but panoramic tripod heads are designed to reduce/eliminate the parallax effect. That’s all that they are designed to do, they don’t exist to make the panoramic process easier or more convenient. And how much the parallax effect is reduced is dependent on you setting the ‘panoramic tripod head’ up correctly, which is a job in itself. And just to add another nail in the panoramic tripod heads coffin,  stitching software handles the parallax effect very well (not 100 percent all of the time, but  not far off), so there is absolutely no need for an expensive tripod head that would probably be more useful as a door stop, and, if like my own experience, may actually prevent you from taking panoramics because of the fiddlyness and time consuming nature of attaching a whole load of largely irrelevant metal to the top of your tripod.

I only really got into panoramic photography after my standard ball and socket tripod head disintegrated whilst on an extended photography trip. So I went into the nearest camera shop and they only had a 3 lever pan/tilt tripod head, it wasn’t what I wanted but at the time I had no choice. Then I noticed that with a pan/tilt tripod head I could position my camera straight, tighten the three levers to hold the camera in place, then loosen the left-right lever enabling me to swivel the camera left and right on the same plane. So all I had to do was level the tripod (easy, as I have  a spirit level on my tripod), then level the camera (easy, as I have a spirit level on my tripod head). With everything level I can move the camera left and right creating a strip of overlapping images. Then I can loosen the up/down lever, tilt the camera upwards for the next row (still overlapping the image below) and then I can photograph another strip of overlapping images from right to left. I could repeat this forever creating whatever shape multiple image photograph that I liked.

If I then combine the ability to move the camera on my tripod in this way with the handheld technique I used to help me visualise and frame my panoramic image, I can take a series of photographs that can be very easily stitched in stitching software.

The technique that I described can be done without levelling the tripod and camera, and can be done with a ball/socket tripod head, but you have to be much more aware of capturing everything that you have visualised. You don’t want any missed gaps. And if your camera moves across at a downward angle without you realising, then the final panoramic will have sloping  edges, which may ‘eat into’ your composition. That is why you should frame your shot beyond what you actually need. Then you crop the image back on the computer. As long as you are aware of this then you can use any tripod head, or even ditch the tripod altogether, just make sure you have enough light in order to handhold the camera without blurring the images.

Some other last minute tips for making panoramic images; set your camera to a fixed white balance, like cloudy or sunny etc. If you set your camera to auto white balance, it is common for a camera to adjust its white balance as you move the camera, so that one half of your finished panoramic photo has a warm look and the other half has a cold look.  If you shoot RAW images (as opposed to jpg), then you can correct this easily in post processing, but it’s still better to get it right beforehand. 

It is not essential to use the exact same exposure value for all individual photos. But my advice is to use manual exposure and change exposure values  as necessary, as you go – but in manual mode. If you shoot in manual mode you can ensure that most of the single images are well exposed. One single manual exposure value for all individual images may result in some parts of the final image being very underexposed or very over exposed. If I am taking a landscape panoramic and the sky is significantly lighter than the ground, I will change the exposure accordingly. This would result in a dark dividing line in a two row panoramic that could be easily fixed in Photoshop.

 Here is another example of a batch of single images. This time I used a significantly different manual exposure value for what will become the top half of the panoramic image. This was because the white clouds beyond the tree branches threatened to overpower the branches, making the thinner branches disappear altogether.  But why not just use an automatic exposure mode, like Aperture priority (A), or shutter priority (S)? Well, I wouldn’t use an automatic exposure mode because that may produce significantly different exposures that serve no purpose. But I will nearly always use different manual exposures to create a panorama because of ‘hotspots’ of dark and/or light that I don’t want. But, of course, you may want the final picture to fall into complete darkness or brightness in some areas. So, as you can see, the whole process is flexible, and you may be different. So don’t be afraid to experiment.

It starts to get complicated when the number of images increases.  Work out exactly what the composition is beforehand

This panoramic is a little more complicated as it’s three rows in height. When you start using this many single images you really need to employ the process I described earlier so that you don’t get empty gaps by mistake. It’s very easy to lose track when composing a panoramic image with so many individual images. What really matters is the discipline of pre-visualising the final image before pressing the shutter(s) in order to eliminate errors later on. You do not want to get back to the computer only to find a big empty squareish gap in the middle of your otherwise finely crafted photograph.

Changing the exposure between individual images will create differently exposed areas in your image, but as long as it's not to extreme, it's easily fixable in Photoshop.

In the stitched panoramic above you can see that the bottom half is lighter than the top half. It’s really easy to correct in Photoshop. If I hadn’t underexposed the top half, a lot of the branches would have been obliterated by the bright cloudy sky.

And the final image.  All the uneven exposure has gone, and the composition is how I originally envisaged.

And here is the cleaned up version. The rough edges from the stitching process have been cropped away, and I have re-exposed the image so that the exposure is even, plus I tweaked the contrast and brightness. Nice and simple.

If you have never taken panoramic photographs before, you can download and install Hugin for free (it’s freeware and only 20Mb), then just go outside and take 3 or 4 overlapping images, handheld if you so wish……load them into Hugin (check out their tutorials ) and watch a whole new world of photography open up…

3 thoughts on “Panoramic photography

  1. Pieter Navis

    Amazing.. this is something to work with, cause with none of the three panorama’s in the forest i expected the result you’ve created. Really made me look at the composition from a fresh perspective!

    Thanks for the inspiration Tony!

    Cheers
    Pieter

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>