基于OMAP3530的2D到3D视频自动转换系统  

System of 2D-to-3D Video Auto-Conversion Based on OMAP3530

作　　者：黄转娣[1] 农镇铭[1] 赖豪[1] 刘佩雯[1] 王卫卫[1] 朱跃生[1] 

机构地区：[1]北京大学深圳研究生院信息工程学院,深圳518055

出　　处：《数据采集与处理》2012年第6期670-676,共7页Journal of Data Acquisition and Processing

摘　　要：研究和实现了一个基于OMAP3530的2D到3D视频自动转换系统,重点研究深度图获取和深度信息渲染等主要核心技术及其实现。该系统利用OMAP3530其特有的双核结构,进行系统优化:由其ARM处理器完成系统界面的设置及控制、显示等功能,其DSP处理器实现图像分割和渲染等核心算法,实现了1个高效的2D到3D视频的视频转换系统。2D转3D技术可把现有的大量传统2D视频转换成3D视频,简化和节省了用摄像机直接捕捉3D场景的复杂性和高成本,可以极大解决3D片源不足的问题。The project studies and implements a system for converting 2D to 3D videos auto- matically based on OMAP3530. It focuses on the algorithms of depth-map extraction and depth-image-based rendering （DIBR）. Using the dual-core structure of OMAP3530＇s to opti- mize the system, the functions like setup, control and display are implemented by ARM pro- cessor, and the algorithms of image segmentation as well as DIBR are implemented by DSP. Therefore, a system converting 2D videos to 3D videos with high efficiency is implemented. The conversion of lots of traditional 2D videos to 3D videos is one way to alleviate the predicted lack of 3D content in the early stages of 3D-TV rollout. The project has great development prospects since it reduce the complexity and money cost compared with the tradition 3D video production using video cameras directly.

关 键 词：深度图获取 深度信息渲染 平面视频转换立体视频 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]