SIGGRAPH '96 Proceedings of the 23rd annual conference on Computer graphics and interactive techniques
Multi-Scale Blur Estimation and Edge Type Classification for Scene Analysis
International Journal of Computer Vision
A compact algorithm for rectification of stereo pairs
Machine Vision and Applications
Light field mapping: efficient representation and hardware rendering of surface light fields
Proceedings of the 29th annual conference on Computer graphics and interactive techniques
Multiple View Geometry in Computer Vision
Multiple View Geometry in Computer Vision
Radon Transform Orientation Estimation for Rotation Invariant Texture Analysis
IEEE Transactions on Pattern Analysis and Machine Intelligence
2D/3D freeview video generation for 3DTV system
Image Communication
New Coding Tools for Illumination and Focus Mismatch Compensation in Multiview Video Coding
IEEE Transactions on Circuits and Systems for Video Technology
Hi-index | 0.00 |
Ray-space interpolation is one of the key technologies to generate virtual viewpoint images, typically in epipolar plane images (EPI), to realize free viewpoint television (FTV). In this paper, a novel Radon transform based ray-space interpolation algorithm (RTI) is proposed to generate virtual viewpoint images in the EPI. In the proposed RTI, feature points of each EPI are first extracted to form the corresponding feature epipolar plane image (FEPI). Radon transform is then applied to each FEPI to detect the candidate interpolation direction set. Then corresponding pixels in neighboring real view rows for each pixel to be interpolated are found by some block matching based interpolation method, in which the smoothness property of the disparity field and the correlation among neighboring EPIs are explored. Possible occlusion regions are processed by the proposed one-sided interpolation. Finally, to solve the problem that the cameras are not equally spaced, a novel ray-space based spacing correction algorithm is proposed to correct the EPI such that pixels corresponding to the same scene point will form a straight line. Experimental results suggest that the proposed RTI and the spacing correction algorithm can achieve good performance. Moreover, compare with traditional methods, our proposed algorithm can generate good interpolation result even if apriori depth range is not known.