基于区域融合的单视点图像深度信息提取

摘要

Generating three-dimensional (3D) scenes from two-dimensional (2D) scenes is one of the important methods for 3D multimedia serv ices application. During 2D - to -3D conversion, depth estimation from a single - view image is probably the most challenging task. In this paper, a new depth estimation method is proposed for single - view image based on region merging. Firstly, the image is segmented to several regions using re gion merging, a prior hypothesis of depth gradient is used to assign depth to this regions, and a depth map is gotten using region merging at last The re sults show that the 3D image can be gotten with high visual quality using the proposed system.%将2D转换为3D是目前3D多媒体成功应用的重要解决方法之一.而在2D至3D转换过程中,单视点图像中深度信息提取是最具挑战的任务.提出一种基于区域融合的单视点图像深度信息提取方法,利用区域融合对单视点图像进行区域融合,用先验假设的深度梯度图对区域进行深度分配,得到基于区域融合的深度图.实验结果表明通过所提方法得到的深度图可绘制出具有强烈视觉效果的3D图像.