25 February 2010 Is a no-reference necessary and sufficient metric for video frame and stereo view interpolation possible?
Author Affiliations +
Abstract
This paper explores a novel metric which can check the consistency and correctness of a disparity map, and hence validate an interpolated view (or video frame for motion compensated frame interpolation) from the estimated correspondences between two or more input views. The proposed reprojection error metric (REM) is shown to be sufficient for the regions where the observed 3D scene has no occlusions. The metric is completely automatic requiring no human input. We also explain how the metric can be extended to be useful for 3D scenes (or videos) with occlusions. However, the proposed metric does not satisfy necessary conditions. We discuss the issues which arise during the design of a necessary metric, and argue that necessary metrics which work in finite time cannot be designed for checking the validity of a method which performs disparity estimation.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Vikas Ramachandra, Vikas Ramachandra, Truong Q. Nguyen, Truong Q. Nguyen, } "Is a no-reference necessary and sufficient metric for video frame and stereo view interpolation possible?", Proc. SPIE 7524, Stereoscopic Displays and Applications XXI, 752413 (25 February 2010); doi: 10.1117/12.839942; https://doi.org/10.1117/12.839942
PROCEEDINGS
6 PAGES


SHARE
RELATED CONTENT

A semi automatic 2D to 3D video conversion with adaptive...
Proceedings of SPIE (November 03 2014)
Efficient view synthesis from uncalibrated stereo
Proceedings of SPIE (January 29 2006)
Hybrid depth cueing for 2D-to-3D conversion system
Proceedings of SPIE (February 17 2009)
Coherent spatial and temporal occlusion generation
Proceedings of SPIE (February 17 2009)

Back to Top