A new imaging model of Lytro light field camera and its calibration

Huixian Duan, Lin Mei, Jun Wang, Lei Song, Na Liu

Abstract: Due to the unique light gathering and post-capture processing, light field cameras with microlens array are being paid more and more attention. Especially, commercial light field cameras, such as Lytro and Raytrix, have been released. But calibrating light field cameras is still an open and challenging problem. In this paper, we present a new imaging model and a calibration method for Lytro camera, which makes its imaging process be easy to understand. Firstly, according to the image formation for pinhole camera and the theory of light propagation, we model the imaging process of Lytro light field camera, and drive a homogenous intrinsic matrix, which represents the relationship between each pixel on the image plane and its corresponding light field. Then, based on the derived intrinsic matrix, a calibration algorithm using the checkerboard grid for Lytro light field camera is proposed. At last, extensive experimental results have demonstrated the correctness of our imaging model of Lytro light field camera and the effectiveness of our calibration algorithm.