Disclosed herein is a method for volumizing and encoding a plurality of two-dimensional images of a subject. The method may comprise obtaining a plurality of two-dimensional (2D) images of the subject, wherein each 2D image of the plurality of 2D images is captured at a different angle from the subject. The method may further comprise generating at least one volumizing hierarchical grid corresponding to the plurality of 2D images. The method may further comprise generating a 3D representation from the at least one volumizing hierarchical grid. The method may further comprise generating an encoding hierarchical grid corresponding to the 3D representation to form a plurality of image blocks. The method may further comprise individually processing the plurality of image blocks to form a plurality of processed image blocks.