1 July 2014 Just noticeable disparity error-based depth coding for three-dimensional video
Author Affiliations +
Abstract
A just noticeable disparity error (JNDE) measurement to describe the maximum tolerated error of depth maps is proposed. Any error of depth value inside the JNDE range would not cause a noticeable distortion observed by human eyes. The JNDE values are used to preprocess the original depth map in the prediction process during the depth coding and to adjust the prediction residues for further improvement of the coding quality. The proposed scheme can be incorporated in any standardized video coding algorithm based on prediction and transform. The experimental results show that the proposed method can achieve a 34% bit rate saving for depth video coding. Moreover, the perceptual quality of the synthesized view is also improved by the proposed method.
© 2014 Society of Photo-Optical Instrumentation Engineers (SPIE)
Lei Luo, Lei Luo, Xiang Tian, Xiang Tian, Yaowu Chen, Yaowu Chen, } "Just noticeable disparity error-based depth coding for three-dimensional video," Optical Engineering 53(7), 073101 (1 July 2014). https://doi.org/10.1117/1.OE.53.7.073101 . Submission:
JOURNAL ARTICLE
6 PAGES


SHARE
RELATED CONTENT

Human eye interfaced multiple video objects coding
Proceedings of SPIE (June 22 2003)
Depth map coding with distortion estimation of rendered view
Proceedings of SPIE (January 18 2010)
Forward rate control for MPEG recording
Proceedings of SPIE (October 21 1993)
Space-time multiple description video coding
Proceedings of SPIE (January 18 2006)

Back to Top