Dissertation > Excellent graduate degree dissertation topics show
Algorithms for Depth Map Generation From Monocular Videos
Author: ZhangPeng
Tutor: WangJia
School: Shanghai Jiaotong University
Course: Communication and Information System
Keywords: Stereo video Depth map Color depth adaptive recursive filter Keyframes Confidence mapping Two - way bilateral filter
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 305
Quote: 0
Read: Download Dissertation
Abstract
|
With the rapid development of science and technology , the transition to three-dimensional stereoscopic display technology more and more attention of experts and scholars from the two-dimensional plane . This paper introduced three-dimensional video background and development began to gradually introduce the latest three-dimensional video display technology , and details monocular video fusion depth information about the three-dimensional display technology . In this technique , the depth image generating essential. The paper discusses in detail the method of acquiring an image depth information from a monocular video sequences , the direction of the study provides some feasible solutions to the depth of three-dimensional video images drawn , as well as provide some valuable ideas for the development of three-dimensional video technology . This paper presents two semiautomatic depth map generation methods: First, by using the adaptive modification of the color depth of recursive filter to generate a depth map , the method through the sub- sampling to improve the processing efficiency of the filter , using the adaptive window is reduced by the motion compensation error using median filter to improve the overall quality of the depth of the image . Second, the proposed algorithm based on the semi-automatic depth map confidence , the keyframe depth information and improved two-way bilateral filtering algorithm to generate the depth information of other frames . The algorithm uses the confidence mapping technology to mitigate the unoccluded region blur , when large foreground objects and not blocked after the King of regional color differences can still be relatively satisfied with the depth of the image . Finally, we compared the performance of two algorithms . 2D video into 3D video is an emerging field , it is a difficult research topic . It is based on computer vision and human visual system , while all image and video processing method . The proposed improvements depth map generation algorithm in the field , and achieved satisfactory results through the experiments prove the feasibility and effectiveness of the algorithm .
|
Related Dissertations
- Based on the two-dimensional to three-dimensional motion information video conversion technology research,TP391.41
- Reconstruction of Super-resolution Depth Image Based on Compressive Sensing and Development of Camera Array,TP391.41
- The Study on the Coding of Intra-Frame Synchronized Stereo Video,TN919.81
- Study on Binocular Stereo Imaging Based on Computer Stereo Vision,TP391.41
- Research on Some Key Technologies Concerning Multi-View Stereo Video Based on H.264,TN919.81
- Research on IBR Technology Oriented to the Multi-view System,TP391.41
- Theoretical Research on Depth Map Creation of Auto-stereo Image,TP391.41
- Key Techniques of Data Processing in 3D Shape Measurement System,TP274
- Fruit and vegetable harvesting robot stereo vision technology research,TP391.41
- The Key Techniques for Converting 2D to 3D Video,TP391.41
- Study of the Key Technique on Content_based Video Retrieval,TP391.3
- Sports video retrieval technology research,TP391.3
- Based on the content of video efficient retrieval technology research,TP391.3
- Study on Content-Based Video Analysis and Retrieval,TP391.4
- Research on Technology of Motion Editing and Motion Synthesis,TP391.4
- Research on Robot Binocular Stereo Vision,TP242
- Feature Expression of Digital Dancing Art and Technique of Video Retrieval,TP399
- Technology of Content Based Skin Detection in Videos,TP391.41
- Compression Coding of Color Video Stream of Multi-angle of View Based on Five-dimensional Matrix(5DM),TN919.81
- Study on Object Segmentation in Stereo Images and Stereo Videos,TP391.41
CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net Mobile
|