I have used the beta release version of this feature with reasonably good success. There are many approaches to sensing depth cues in 2D video that can be used to create a 3D image. One of the simplest is when the camera or the subject are moving. By looking at one frame, and then another frame offset by a small number of frames (depending on speed of movement), one frame can be used for the “left eye” view and the other for the “right eye” view. Other tricks can be used to create a “depth map” that attempts to position objects within a 3D depth.
Click on the viewer’s 3D button and choose the 3D option that works for you and watch the video on Youtube, here. (The 3D viewing option does not appear on the embedded video, below).