Filming in 3D generally requires two cameras that are separated laterally, to create the parallax effect needed for stereoscopic vision. Fortunately, videos shot from Earth orbit can be converted to 3D without a second camera, because the camera is constantly in motion.
An observer in low earth orbit travels at five miles per second, or roughly 1/5th of a mile (1,000 feet) per frame at 24 frames per second. So, we can create a virtual 3D camera simply by time shifting the video to create two channels, one shifted a few frames behind the other. Better yet, we can make this separation arbitrarily large, from as little as 1,000 feet to several miles, and thus bring fine surface features such as mountains and clouds into fine relief, as well as the curvature of the Earth.This method can also be used for still photos, simply by shooting several photos in rapid sequence.
If the camera is stationary, with a fixed viewing angle and no near field objects in view, no post-processing should be required, just a simple time delay between left and right channels.It’s a simple enough idea. If you produce something interesting, hit me up at bsmcconnell at gmail or @brianmsf.
UPDATE: this method will work best for video where the point of view is perpendicular to the direction of travel (to the side or downward). It won’t work so well for videos shot fore or aft the direction of travel.
UPDATE: according to Alex Tolley this method was also used in aerial reconnaissance during WWII (thanks Alex).
UPDATE: I learned about the Hyperlapse project shortly after I published this article. In principle, this method should work here as well.