DocumentCode :
1783194
Title :
Symmetric bidirectional expansion algorithm to remove artifacts for view synthesis based DIBR
Author :
Hui Ding ; Zhaohui Li ; Rong Hu
Author_Institution :
Sch. of Inf. & Eng., Commun. Univ. of China, Beijing, China
fYear :
2014
fDate :
28-29 Sept. 2014
Firstpage :
1
Lastpage :
4
Abstract :
Virtual view synthesis is the key technology of the current Three-dimensional Video applications and future Free Viewpoint Television. And virtual viewpoint synthesis based on depth image by using two adjacent reference two-dimensional viewpoint color images and corresponding depth images to render the intermediate view is the most potential algorithm for virtual view rendering. First, a modified algorithm proposed which is based on DIBR aimed at solving visibility and resampling problems occurred after the process of three-dimensional warping in DIBR. Second, based on this algorithm, a symmetric bidirectional expansion algorithm was developed which could effectively remove the artifacts of the virtual view image and the presented visual perception was in good performance. Last, the corresponding image fusion algorithm was put forward to improve the PSNR performance. Theoretical analysis and a series of experiments show that the proposed method can generate eventual virtual viewpoint image with high subjective and objective quality.
Keywords :
image colour analysis; image fusion; rendering (computer graphics); video signal processing; PSNR; adjacent reference two-dimensional viewpoint; color images; depth image; free viewpoint television; image fusion; symmetric bidirectional expansion algorithm; three-dimensional video application; three-dimensional warping; view synthesis-based DIBR; virtual view rendering; virtual viewpoint synthesis; visual perception; Algorithm design and analysis; Color; Image fusion; PSNR; Rendering (computer graphics); TV; Three-dimensional displays; DIBR; Free Viewpoint; depth image; expansion algorithm; multi-view;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multisensor Fusion and Information Integration for Intelligent Systems (MFI), 2014 International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4799-6731-5
Type :
conf
DOI :
10.1109/MFI.2014.6997739
Filename :
6997739
Link To Document :
بازگشت