- - SUPER-RESOLUTION DEPTH MAP GENERATION FOR MULTI-CAMERA OR OTHER ENVIRONMENTS

摘要

A method according to an embodiment of the present disclosure includes, by using at least one processor, obtaining a first input image frame and a second input image frame associated with a first image plane and a second image plane, respectively; obtaining, by using at least one processor, a first depth map associated with a first input image frame; and performing the following steps (a) and (b) one or more times to generate a second depth map: (a) (i) a depth map and (ii) a second image plane from the first image plane. projecting the first input image frame onto a second image plane, using the at least one processor, to generate a projected image frame using the information identifying the transformation to ; and (b) using the at least one processor to adjust at least one of the depth map and information identifying a transformation from the first image plane to the second image plane.