Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/22382
Title: | 深度圖為基礎的立體視覺多視角影像生成 Depth-Image-Based Rendering for Stereo Vision |
Authors: | Meng-Sung Wu 吳孟松 |
Advisor: | 莊永裕(Yung-Yu Chuang) |
Keyword: | 深度圖為基礎的影像生成,立體視覺,影像修補, Depth-image-based rendering (DIBR),Stereo(3D) vision,Image inpainting, |
Publication Year : | 2010 |
Degree: | 碩士 |
Abstract: | 對於3D content的儲存,2D-plus-depth在彈性、檔案大小方面皆有優點。然而現存depth-image-based rendering演算法的不足,使得以2D-Plus-Depth格式儲存的3D contenet所呈現的立體效果甚為不佳。在這篇論文中,藉由加強depth-image-based rendering演算法,特別是補洞的部分。藉由這些改進,我們可以使生成的影像與給定的影像差異加大,使得觀察者能感受到較為強烈的立體感。 2D-plus-depth has advantages of flexibility and file size for storing 3D content. However, existing algorithms for depth-image-based rendering are not sufficient to generating high quality stereo images. In this thesis, we enhance depth-image-based rendering algorithm, especially in hole-filling. By doing so, we can enlarge the difference between the generated images and original images to make observers perceive more stereoscopic. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/22382 |
Fulltext Rights: | 未授權 |
Appears in Collections: | 資訊工程學系 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-99-1.pdf Restricted Access | 6.16 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.