binocularity.org
Knowledge about all aspects of 3D displays and their application.
Home3D DisplaysHuman 3D VisionCreating 3D ImagesResources

Drawing stereo pairs
Toed-in cameras
Parallel cameras
3D photography method
..3D photography equipment
3D filming
Stereoscopic OpenGL
..Toed-in OpenGL example
..Parallel OpenGL example


Drawing stereo pairs
A stereoscopic image pair can be easily created by drawing the two views using a bitmap editing tool (such as Gimp or Adobe Photoshop) or a vector image drawing tool (such as Inkscape or Adobe Illustrator). The image below was created using Inkscape and stored in the scalable vector graphics format before being converted to a bitmap in png format for display on this web page. If you can free fuse stereo images you should be able to see the blue shape is in-front of the yellow shape and the green shape is behind the yellow one.

Stereoscopic image pair drawn using vector graphics.

Here the left eye image is on the left and the right eye image on the right. The process for creating these images is simple and was well known to Wheatstone when he created the first ever stereoscopic image pairs.

A stereo image frame

The frame of reference around each image is important to help establish a zero depth plane, this is perceived to be in the display plane.

Stereo image zero depth object

Shapes in the the same depth plane as the frame of reference, such as the yelow square above, should be in the same horizontal position in each image. There will therefore be no stereo disparity and no perceived depth relative to the frame for these shapes.

Stereoscopic image with a near object

To make a shape appear in front of the display plane, such as the blue square above, its relative horizontal position should be moved to the right in the left image and to the left in the right image. This produces a crossed stereoscopic disparity so that when the images are viewed binocularly the shape appears in-front of the display plane.



For the green square which appears behind the display plane its relative horizontal position should be moved to left in the left image and to the right in the right image. This produces uncrossed stereoscopic disparity and the shape is perceived to be behind the display plane when viewed binocularly.

It is only necessary to introduce small amounts of horizontal disparity to achieve a 3D effect, as little as one pixel of horizontal disparity can be visible as depth, and values up to 20 pixels of crossed or uncrossed disparity will produce a significant effect. Beyond 20 pixels is often too much on desktop 3D displays.

Home3D DisplaysHuman 3D VisionCreating 3D ImagesResources