Intermediate view synthesis and interpolation based on disparity calibration and region segmentation

Authors:
Ping An;Zhaoyang Zhang;Ran Ma
Affiliations:
School of Communication and Information Engineering, Shanghai University, Shanghai, China;School of Communication and Information Engineering, Shanghai University, Shanghai, China and Key Laboratory of Advanced Displays and System Application, Ministry of Education, Shanghai University ...;School of Communication and Information Engineering, Shanghai University, Shanghai, China
Venue:
MUSP'06 Proceedings of the 6th WSEAS international conference on Multimedia systems & signal processing
Year:
2006

Citing 6
Cited 0

Digital image processing algorithms

Digital image processing algorithms
Teleconferencing eye contract using a virtual camera

CHI '93 INTERACT '93 and CHI '93 Conference Companion on Human Factors in Computing Systems
Stereo Vision for View Synthesis

CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)
Arbitrary view generation from multiple cameras

ICIP '97 Proceedings of the 1997 International Conference on Image Processing (ICIP '97) 3-Volume Set-Volume 1 - Volume 1
Bayesian winner-take-all reconstruction of intermediate views from stereoscopic images

IEEE Transactions on Image Processing
An object-based system for stereoscopic viewpoint synthesis

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

Since image quality of the synthesized intermediate view depends mainly on precise and dense disparity field and on detection of occlusion in multi-view applications, this paper proposes an intermediate view creation and interpolation algorithm taking advantages of transmitted and decoded disparity fields to avoid complex disparity calculation. This algorithm also combines region segmentation and disparity calibration techniques together to correct those mismatched disparities. What's more, in the intermediate view creation, region segmentation technique is used to judge the contents of regions with occlusion and blur, and fill them depending on the disparity characteristics. Experimental results show that the constructed intermediate view images have favorite view effect. This algorithm can be used in 3D videoconference systems in which the intermediate view images have to be synthesized in real-time.