Scene depth extraction from Holoscopic Imaging technology

Abstract

3D Holoscopic Imaging (3DHI) is a promising technique for viewing natural continuous parallax 3D objects within a wide viewing zone using the principle of "Fly's eye". The 3D content is captured using a single aperture camera in real-time and represents a true volume spatial optical model of the object scene. The 3D content viewed by multiple viewers independently of their position, without 3D eyewear glasses. The 3DHI technique merely requires a single recording that the acquisition of the 3D information and the compactness of depth measurement that is used has been attracting attention as a novel depth extraction technique. This paper presents a new corresponding and matching technique based on a novel automatic Feature-Match Selection (FMS) algorithm. The aim of this algorithm is to estimate and extract an accurate full parallax 3D model form from a 3D Omni-directional Holoscopic Imaging (3DOHI) system. The basis for the novelty of the paper is on two contributions: feature blocks selection and corresponding automatic optimization process. There are solutions for three main problems related to the depth map estimation from 3DHI: uncertainty and region homogeneity at image location, dissimilar displacements within the matching block around object borders, and computational complexity.

3D-Interactive-depth generation and object segmentation from Holoscopic image

“…The method is better than the depth-through-disparity algorithm as it overcomes errors within the algorithm. A recent paper presented a new approach to estimate and extract from a 3D Omni-directional HI system a full parallax 3D depth map [14]. This approach is a great improvement on the current optimal 3D object estimation and segmentation techniques.…”

Recognition of Holoscopic 3D Video Hand Gesture Using Convolutional Neural Networks

“…A feature match selection algorithm is presented by [13], with an aim to extract and estimate an accurate full parallax 3D model form from a 3D omni-directional holoscopic imaging (3DOHI) system. The novelty of the paper is based on two contributions: feature block selection and its corresponding automatic optimization process.…”