What is it?

Panorama photography has been around roughly as long as photography itself, although immersive '360x180-degree' interactive panorama photography is rather newer. Interactive panoramas began, more or less, with Apple's QuickTime VR in the mid-1990s, and complete spherical panoramas (where you can look all the way up and down as well as around) arrived later than that.

Photographic panoramas can be created in a number of different ways:

1. Shoot with a wide-angle lens and crop off the top and bottom
2. Shoot with a rotating lens set-up
3. Shoot multiple images and collage them together

1) This method is often used by cheap compact digital cameras, and even print processing companies. It involves throwing away image data, so you effectively downgrade the image in order to achieve the effect. Also, you can only reach ‘around’ as far as your lens can see in a single shot.

2) The rotating lens method is rare, and it requires expensive specialist equipment. (Extremely expensive if you want more than roughly 120° coverage.) It also has a number of limitations.

3) This method is the one used by modern panoramic photographers, although at the basic level it is hardly new. The simple collage technique of overlaying images to create one bigger picture can be done with prints and glue,or in Photoshop. This is sometimes called ‘panographic collage’, and it was used most famously by David Hockney. These panographic images are generally made without serious effort to hide the joins. Shots may have different exposures and other settings; the end result is normally clearly the result of multiple shots. This is fairly easy to do and doesn’t involve any specialist equipment or skill. The images are often used as-is as artworks, and they can be used wherever seems suitable, whether as prints or as digital images used in other projects.

Producing a more convincing apparent single image requires precise overlaying of images with warping, blending and masking used to hide the joins. This takes more work than the simple collage approach, both at the shooting stage and the assembly stage, but the result is a large, high-resolution, seamless composite image created from multiple original images. This is how high-quality panoramic images are made, and it is the main way that I produce original panorama content.

The ‘how on earth?’ page, also known as the Quick Panorama F.A.Q., covering what it is, how it is done and who uses it.

How is it done?

Multi-shot panorama photographs are made by shooting a series of overlapping images and then stitching them together using software that can find matching points in image overlaps and warp the content to get everything to fit properly.

In order for the different photographs to work seamlessly together it is important to follow some basic guidelines.

• Images should have enough overlap to allow enough pairs of matching points to be found for each set.
• The camera should rotate around the optical centre of the lens to avoid parallax errors where the images overlap. This is normally done using specialist panorama heads, either commercial or home-made. (Panos can be shot hand-held, but that’s hard to do well.)
• All camera settings should be on manual, otherwise exposure, focus, white balance and other settings could shift from one photo to the next, causing problems when blending the shots together.

It can help to take the shots fairly quickly if the scene includes people moving around a lot, or similar active content. Shooting with a larger overlap can also be helpful as it provides more scope for masking parts of one image out.

If shots are taken to cover the view upwards (the zenith) and the view downwards (the nadir) then a complete ‘spherical’ panorama that allows complete interactive viewing freedom can be made. This is normally done with fisheye lenses to minimise the number of shots required and specialist panorama heads to keep all the different rotations correctly aligned.

The individual images are ‘stitched’ together using specialist software such as PTGui, Stitcher, AutoPano Pro and others, or using the Photomerge feature in Photoshop CS3 and CS4. (Photoshop CS3's feature only works with shots taken with rectilinear lenses, not the fisheye lenses normally used when shooting for complete spherical panoramas. Additionally, it can only make cylindrical panoramas.)

Once stitched - and checked and possibly tweaked in Photoshop - the result is processed for whatever kind of output is required. For on-screen interactive use this is normally as a QuickTime VR (QTVR) file or as one of the growing number of Flash-based panorama players, although Java-based viewing is also possible. Tools for doing this include Pano2VR, Flash Panorama Player, KRPlayer and others.

For print use this is normally processed to a specific cropped projection of part of the complete image, although mapping techniques can be used to apply the complete image to objects and packaging as image wraps or in panelled sections.

Who uses it?

Panoramas appear all over. In print you'll see them anywhere from billboard posters to corporate annual reports, magazine illustrations and product packaging artwork. Interactive panoramas are viewed on computers. They're normally visited online, but they can also be delivered using CD-ROM and other storage media.

Panoramas can be presented as individual views or as linked ‘virtual tours’, and they are used for an incredibly broad range of things including promotional marketing of all kinds, as well as game development, film production, security reviewing and legal documenting, teaching, and fine art.

If you are interested in having location or event panoramic photography done for print or online work or in doing it yourself, contact me for advice, suggestions or quotes.