PROJECT TITLE :

Region-Based Prediction for Image Compression in the Cloud - 2018

ABSTRACT:

Thanks to the increasing number of pictures stored within the cloud, external image similarities will be leveraged to efficiently compress images by exploiting inter-images correlations. In this Project, we tend to propose a completely unique image prediction scheme for cloud storage. Unlike current state-of-the-art methods, we tend to use a semi-local approach to use inter-image correlation. The reference image is 1st segmented into multiple planar regions determined from matched local options and super-pixels. The geometric and photometric disparities between the matched regions of the reference image and this image are then compensated. Finally, multiple references are generated from the estimated compensation models and arranged during a pseudo-sequence to differentially encode the input image using classical video coding tools. Experimental results demonstrate that the proposed approach yields important rate-distortion performance improvements compared with this image inter-coding solutions like high potency video coding.