Using a Combination of Local and Global Measures to Assess DIBR-Synthesized Image Quality PROJECT TITLE : Combining Local and Global Measures for DIBR-Synthesized Image Quality Evaluation ABSTRACT: 3D video applications, such as 3D television and free perspective video, rely heavily on depth-image-based rendering (DIBR) techniques (FVV). An irritating viewing experience is caused by the DIBR-synthesized image's many distortions, which are evident to everyone who sees it. Perception-friendly FVV systems necessitate an image quality evaluator that can be used to DIBR-synthesized images. Because the reference image is frequently unavailable, full-reference (FR) methods cannot be used to directly evaluate the quality of the synthesised image. The DIBR-related distortions can't be accurately measured by typical no-reference methods. NR quality is evaluated based on two types of DIBR-related aberrations, namely geometric distortions and sharpness, in this paper. To begin with, the most visible geometric distortions, such as the occluded regions, are captured by comparing local similarities. Finally, another common geometric distortion (i.e., stretching) is found and evaluated by comparing it to its equal-size nearby region. Second, the global sharpness is measured as the distance between the distorted image and its downsampled equivalent, taking into account the condition of scale invariance. Finally, perceptual quality is measured by linearly combining the scores of two geometric distortions and sharpness together.. The proposed strategy outperforms the current FR and NR measures in terms of performance. While APT is superior in terms of effectiveness, it is significantly more time-consuming to execute than any other strategy. Did you like this research project? To get this research project Guidelines, Training and Code... Click Here facebook twitter google+ linkedin stumble pinterest Fixations on CNN Visualizing Discriminative Image Regions: An Unraveling Approach Filamentous Structures for Content-Aware Image Enhancement