Depth Map Estimation and Restoration of an Improved-quality Image from Multi-aperture Images

Abstract

In this study, we propose a method for estimating a depth map from images captured by the multi-aperture camera, which places CMOS image sensors and lenses in a two dimensional array. By employing a global estimation method, we succeeded in estimating a depth map even from noisy images which are captured in a low light condition. We also succeeded in obtaining an improved quality image by synthesizing the multi-aperture images with disparity compensation.