Fujitsu is developing a smart way to allow cellphones with only one rear camera to jump onto the 3D bandwagon. The concept is based upon an attachment which uses mirrors to send two discrete stereo images to the camera's sensor. The user then shoots their movie and uploads it to Fujitsu's servers for processing, where it's aligned, has the perspective created based on the difference between the two images, and is returned as a true polarized-3D video.
Using mirrors to create 3D images with one lens isn't a new idea — a paper by Dr. Donald Simanek from Lock Haven University demonstrates how easily you can do it yourself — but this simple attachment and the use of cloud processing could popularize the technique. One real issue we can see is that each image only takes up around 25 percent of the sensor, so the output videos will be nowhere near as high-quality as the 2D ones that your cellphone's camera can shoot natively.