Abstract—Integral Photography (IP) has been extensively studied in recent years. As a representative problem of IP, there is a problem that if the rays from object are recorded as inverted images of objects, the depth inversion occurs in the displayed stereoscopic image. As a shooting method to solve this problem, a method of shooting with a two-dimensional concave lens array and recording an erect image of the object has been proposed. In IP, the size of individual lenses of the lens array used in the recording procedure and display procedure must be the same. Therefore, when using a concave lens for recording, it is necessary to align the size with the convex lens used for display, but finding the lens array of the same size is hard. So there is an approach making the concave lens array which is the same size to the convex lens array, but with the hand-made lens array the processing accuracy is not good. Therefore, we propose a method to create a concave lens array in CG using ray-tracing software POV-Ray and shoot IP image in CG. With this method, it is possible to easily produce a lens array with high processing precision according to the size of the lens array for display. In experiments, it was confirmed that IP images could be taken and stereoscopic images could be displayed even in photography in CG.

Index Terms—integral photography, integral imaging, photographing, concave lens array, CG, POV-Ray

Cite: Shotaro Mori and Yue Bao, "A Photographing Method of Integral Photography Using a Concave Lens Array in CG," Journal of Image and Graphics, Vol. 6, No. 1, pp. 74-79, June 2018. doi: 10.18178/joig.6.1.74-79