- Mar 26, 2014
My understanding is the lens projects two 17.5mm tall (circular) images side by side. The real world view overlaps, but each image gets a separate part of the sensor to be projected on. Software is used to combine the two images into a 3D model of the scene. AFAIK, the model's resolution can't exceed that of each image, which is well under 8K.I agree I can’t see how they can get 8K from half the sensor. Given the left and right will overlap when rendered.
Unless they are using half pixels as whole ones. I.e. left and right sides of the dual pixel to double the horizontal resolution. It would be the first time such a thing had been done, to my knowledge.
It would surely darken the image as only half as much light would hit each half pixel. interesting idea though.