Open Photographic Society -
OPS Front Page 3
Low Tide, Bali, Indonesia
Dawn over Bryce Canyon, Utah, USA
Bryce Canyon Hoodoos, Utah, USA
http://www.openphotographicsociety.org/images/front_page_images/front_page_images_3/bryce_13-2.jpg
Rice Farmers Walking to Work at Dawn, Burma
Ponte Vecchio, Florence, Italy
People on a Ferry, Gambia, West Africa
Ghanaian Fishermen, West Africa
Children Playing, Gambia, West Africa
Dunn's River Falls, Jamaica
The London Eye, UK
Gentoo Penguins, Antarctica
Tribesman, Highlands of Papua New Guinea
http://www.openphotographicsociety.org/images/front_page_images/front_page_images_3/png_01-06.jpg
Reflected Beauty, Prague, Czech Republic
Church on the Island of Santorini, Greece
Plaza de Espagne, Seville, Spain
Roof of the Esplanade, Singapore
Sparrowhawk, UK
Ploughing, UK
Western Reef Egrets, Gambia River, West Africa
Columns, Tunisia
Articles
What is 3D Imaging?
Photographers often speak of the depth in a printed image although a two-dimensional image has, by definition, no physical depth. The illusion of depth can however be created by skilled photographers by manipulating perspective, tonal gradation, contrast and so on. The viewer knows by experience that a large object such as a building which appears smaller in an image than a person must be much further away. The brain interprets such information in extremely sophisticated ways without us even realizing what has happened.
True 3D imaging, sometimes known as stereoscopic imaging, relies upon a pair of slightly different images which simulate the inputs received by our two eyes. A person with normal vision perceives the world simultaneously through two different lenses (eyes) separated horizontally by a few inches. The images received by the two eyes are different principally because of the different locations relative to the subject from which they are captured. In the extreme example shown in the diagram it is easy to see why the red ball is perceived by one eye but not the other.
The two images captured by our eyes from different viewpoints are combined by the brain to create a three-dimensional image which gives human beings their spatial sense. People with normal sight are consequently well able to perceive a true three-dimensional world. Those with eye defects, or who use only one eye, may have more difficulty with spatial perception but nevertheless develop effective means of coping with the world through experience, the perception of perspective and so on. These techniques are much the same as those used by normally-sighted people when viewing a two-dimensional photograph.
The two images required to produce a 3D or stereo image can be produced in a number of ways. The simplest approach, although not the best, is to use a normal camera to capture two similar images of the subject from slightly different viewpoints. The subject must remain motionless because the two images are by necessity taken at different times. Another basic approach uses two adjacent camera whose shutters are released simultaneously.
A more practical approach to 3D or stereo photography uses a custom-made stereo camera fitted with two lenses. This is designed to duplicate the views perceived by human beings, with the two lenses separated by a distance comparable with the separation of a pair of eyes. When the two images are then presented to two eyes, only one image being visible to each eye, the brain combines them in to a single spatial image which provides a real impression of depth.
Getting Started
Photographers wishing to make a start in stereoscopic photography are probably well advised to begin by reading the extensive material available on the subject. Those who wish to experiment with the concept without too much commitment might choose to create basic stereoscopic images using a single camera and use them to achieve an initial understanding of the subject. Others might reasonably opt to use digital facilities to capture and mix pairs of images which can then be viewed through special glasses.
However, photographers who are serious about producing good quality stereo images should consider purchasing or creating a specialized stereoscopic camera equipped with two lenses. Ready made film-based and digital products (above and below) are available, but a cheaper alternative may be to join together in an appropriate manner two conventional single-lens cameras. These are then equipped with a common shutter release mechanism which enables the two exposures to be made simultaneously.
An excellent source of detailed information on 3D photography may be found Stereoscopy.com.
Positioning in 3D Imaging
A pair of stereoscopic images must simulate as closely as possible the two images perceived by a pair of normal human eyes. Eyes are separated by a distance that varies from person to person but which is normally in the range 58mm to 68mm. This distance, known as the pupillary distance, determines the extent of the small separation in viewpoint recorded in the perceived images. The separation of the eyes is also along the horizontal plane. The eyes are designed to cope with this lateral separation and are capable of independent horizontal movement of their axes. However, they cannot move independently in the vertical plane and cannot cope with the perception of two vertically shifted images. This is simply outside the boundaries of what they are equipped to do. The alignment of a pair of stereoscopic images must therefore be arranged in a similar fashion. The two images must be perfectly aligned in the vertical plane and separated horizontally by an appropriate distance.
The horizontal separation of a pair of stereoscopic images viewed in a stereoscope determines the distance at which the merged spatial image is formed. A larger separation between the individual images moves the spatial image back and a smaller separation moves it forward. The plane occupied by the slide frames or the print borders of the individual images is generally know as the stereo window, but the spatial image can be formed either in front or behind this plane. In most cases, it is desirable for the spatial image to be formed behind the stereo window. The size of the spatial image also varies as its position relative to the stereo window changes.
Focus for 3D Imaging
Human beings endeavour to see the world around them in the clearest possible manner, and may even feel uncomfortable when this is not possible. Our survival instinct demands as much information as possible to achieve physical safety. Our eyes therefore focus automatically upon every perceivable detail of a subject, and the smallest details are often required to enable the brain to assemble a spatial image from the inputs provided by two eyes. Those who suffer eye defects such as short or long sight therefore prefer to wear spectacles with correction lenses.
When viewing an image, the eyes focus in the same way to extract as much detail as possible. If the image is blurred or out of focus it may be difficult or uncomfortable to view. If a pair of stereoscopic images are soft and generally lacking in detail it will be more difficult for the brain to merge them. Sharp focus and extensive depth of field are therefore important elements of stereo photography.
In practice this means maximizing depth of field by using a small aperture - typically f/8 or smaller. With correct focus, the depth of field envelope can then be arranged to cover the whole depth of the image - from the close foreground to the far distance.
Stereo Pairs
Three-dimensional effects effects are most simply achieved by capturing two images of a subject from two slightly removed positions corresponding to the two locations of a pair of human eyes. Such images are typically known as stereo pairs. The optimum distance between the two camera locations is dependent upon the extent of the stereo effect required and the distance to the nearest objects in the frame.
To achieve the required result, a photographer must define a projection plane, and angle the two cameras inward so that the centre lines of the two lenses intersect at this plane. Objects that are closer to the camera than the projection plane appear to be closer to the viewer, and those more distant appear to be further away from the viewer. In general, the separation of the two camera locations should be around 3 - 7% of the distance from the camera to the projection plane.
Once the stereo pair or images has been captured, they must be presented in a suitable manner to the viewer. A number of techniques are available. The key requirement is that each image should be viewed independently by a single eye. This mimics the real-life situation where a normally-sighted person perceives two slightly different views of a subject with their two eyes and the brain combines the images to give perception of depth.
Some normally-sighted people are able to force their eyes to focus on a point beyond the plane of the images and hence see, without additional assistance, a separate image with each eye. Others are able to cross their eyes and see the left image with their right eye and vice versa. "Cross-eye" stereo pairs are intended to be viewed in this way. However both these approaches require some practice.
Another technique involves the use of red/blue glasses to view the two superimposed images, which are presented in the appropriate colour to allow separation by the coloured lenses. Images of this type are known as anaglyphs. A basic limitation of this technique is that only greyscale images can be viewed. An alternative but similar approach uses two projectors equipped with right and left circularly polarized filters to superimpose the two images on a suitable screen and view them through correctly polarized glasses.
|
|
| Crossed pair of the eyes of a carpenter bee - Image by kind permission of Ian Jacobs via Flickr. See also. |
More sophisticated computer-based techniques are also available. Typically, these involve changing the monitor display rapidly between left-eye and right-eye images, and equipping the viewer with synchronized shuttered glasses controlled by the computer. The shutter on the left side is then closed when the right-eye image is displayed and vice versa.