摘要由于3D视频可以给用户提供更好的视觉体验，在观看时获得更强的临场感，所以它越来越受到市场的青睐。为了使3D视频的应用称为可能,多视点加深度格式（(Multiview Video plus Depth，MVD)的3D视频格式被提出，它主要由多视点视频及其对应的深度图序列组成。本文主要对MVD格式中深度图的编码进行研究。和传统图片不同，深度图并不是直接用于观看的图片，其中的值表示的是场景中物体到摄像机的距离，主要应用于视点合成。本文主要介绍了在传统视频编码标准HEVC（High Efficiency Video Coding）基础上发展而来的新技术3D-HEVC。和HEVC相比，3D-HEVC对于多视点视频及其对应的深度数据更加实用。对于二者，本文都做了详细说明，并且分别利用其所对应的仿真软件对多个3D视频进行编码。主观和客观结果都表明，3D-HEVC和HEVC相比，节省了50%的比特流。68594

毕业论文关键词 ：3D视频编码  多视点视频加深度  深度图  HEVC  3D-HEVC

毕业设计说明书（论文）外文摘要

Title: Depth Map Coding In The 3D Video System

Abstract

Since the 3D video can give users a better viewing experience and the users can get a stronger sense of reality when watching ,it is increasingly favored by the market. To make the application of the 3D video become possible, the multi-view plus depth 3D video formats are presented, which mainly consists of multi-view video and corresponding depth map sequences. This paper focuses on the research of depth map coding in the MVD. Different from the traditional images, depth map is not directly used for viewing, where the value represents the distance of the object in the scene to the camera, mainly used in the synthesis of perspective. This paper describes the traditional video coding standard（High Efficiency Video Coding ,HEVC）and an extension of the HEVC standard for coding of multi-view video and depth data, which is known as 3D-HEVC.In contrast to HEVC, 3D-HEVC is more practical for multi-view video and corresponding depth data. For both of two, this paper has given a detailed description and used simulation software to encode multiple 3D video according to their different characteristics. Objective and subjective results are presented, demonstrating that the proposed approach provides 50% bit rate savings in comparison with HEVC simulcast.

Key words : 3D video coding  multi-view plus depth 3D  depth map coding    HEVC  3D-HEVC

目录

1.  绪论 1

1.1  课题研究背景 1

1.2  课题研究的目的和意义 2

1.3  文章安排 4

2.  HEVC编码流程概述 5

2.1  HEVC编码流程简介 5

2.2  HEVC混合编码的特点 6

3.  基于3D-HEVC 结构的深度图编码技术 10

3.1  3D-HEVC介绍 10

3.2  基于3D-HEVC的深度图编码技术 12

3.2.1  基视点编码 12

3.2.2  非基视点编码 12

3.2.3  深度图编码 13

3.2.4  深度图的帧内编码 16

4.  实验仿真 18

4.1  实验概述 18

4.1.1  仿真软件 18