PROJECT TITLE :

Multiple-Level Feature-Based Measure for Retargeted Image Quality - 2018

ABSTRACT:

Objective image retargeting quality assessment aims to use computational models to predict the retargeted image quality consistent with subjective perception. During this Project, we have a tendency to propose a multiple-level feature (MLF)-primarily based quality live to predict the perceptual quality of retargeted images. We 1st give an in-depth analysis on the low-level aspect ratio similarity feature, and then propose a mid-level edge cluster similarity feature, to raised address the shape/structure related distortion. Furthermore, a high-level face block similarity feature is intended to house sensitive region deformation. The multiple-level options are complementary as they quantify different aspects of quality degradation in the retargeted image, and also the MLF measure learned by regression is employed to predict the perceptual quality of retargeted pictures. In depth experimental results performed on 2 public benchmark databases demonstrate that the proposed MLF measure achieves higher quality prediction accuracy than the prevailing relevant state-of-the-art quality measures.