An Efficient Method for 2D to 3D Conversion using Grouping of Image release_ndsy3xwzqffy5a4nd7bmrex3fe

Abstract

In recent years 3D display has become a trend in visual processing field.3D display provide better visual quality than conventional 2D displays. The three-dimensional (3D) displays requires the depth information which is unavailable in the conventional 2D content. This work presents an algorithm that automatically converts 2D images into 3D ones. The proposed method first divides the image into object groups by choosing an effective grouping method. Based on a hypothesized depth gradient model the depth map is assigned. Next, the depth map is assigned by cooperating with a cross bilateral filter to diminish the blocky artifacts. A Multiview image can be readily generated by processing the filtered image using a depth image based rendering method.
In text/plain format

Archived Files and Locations