“…From the view of geometric optics, those methods simultaneously record the two-dimensional (2D) spatial and angular information of the light rays, thus allowing perspective view image generation, refocusing of the scene, and free-glass 3D display [2,5,6]. However, lens array based light field capture [1][2][3][4][5][6][7][8] has to make an intrinsic trade-off between the the spatial and the angular resolution. This is because when the size of the lenslet is large, one of the captured elemental image will have a large spatial resolution, therefore the covered quantity of lenslet that the light rays from the object scene will be small, which leads to less number of elemental images, i.e., low angular resolution.…”