Call For Paper Volume:4 Issue:10 Oct'2017 |

Methodology For Multi-Metric Fusion Of 3D Stereoscopic Compressed Image

Publication Date : 28/04/2015

Author(s) :


Volume/Issue :
Volume 2
Issue 4
(04 - 2015)

Abstract :

The objective approaches of 3D image and video quality assessment play an important role in a variety of image processing applications. A new methodology for objective image quality assessment (IQA) with multi-metric fusion(MMF) is employed. There is no single method that can give the best performance for all the images. So we are going for multi-metric fusion method.To achieve MMF score we use regression approach. The new MMF score is set to be the non-linear combination of scores from multiple methods with suitable weights obtained by the training process. The distorted images are classified according to the distortion types and perform regression approach which is called “context-dependent MMF”. The additional new challenges faced by 3D image quality assessment are asymmetric stereo compression, depth perception and virtual view synthesis than its 2D counterparts. These challenges are not faced by most commonly used 2D metrics (e.g., PSNR and SSIM).This is verified by the low correlation between the objective and subjective measures (MOSs). Besides 2D-metrics, the binocular integration behaviors—the binocular combination and binocular frequency integration are used as the measures for stereoscopic 3D image with MMF. The proposed MMF method using support vector regression outperforms a large number of existing IQA methods and can be used for both Color-Plus-Depth 3D images well

No. of Downloads :



Web Design MymensinghPremium WordPress ThemesWeb Development

Methodology For Multi-Metric Fusion Of 3D Stereoscopic Compressed Image

April 27, 2015