Centroid adapted frequency selective extrapolation for reconstruction of lost image areas

TL;DR

This paper introduces a centroid adaptation to frequency selective extrapolation, improving reconstruction of arbitrarily shaped lost image areas with an average gain of 1.29 dB.

Contribution

The paper presents a novel centroid adaptation method that considers the shape of lost areas in frequency selective extrapolation for better image reconstruction.

Findings

Average reconstruction gain of 1.29 dB on large test set.

Method effectively handles arbitrarily shaped lost areas.

Enhanced reconstruction quality over existing techniques.

Abstract

Lost image areas with different size and arbitrary shape can occur in many scenarios such as error-prone communication, depth-based image rendering or motion compensated wavelet lifting. The goal of image reconstruction is to restore these lost image areas as close to the original as possible. Frequency selective extrapolation is a block-based method for efficiently reconstructing lost areas in images. So far, the actual shape of the lost area is not considered directly. We propose a centroid adaption to enhance the existing frequency selective extrapolation algorithm that takes the shape of lost areas into account. To enlarge the test set for evaluation we further propose a method to generate arbitrarily shaped lost areas. On our large test set, we obtain an average reconstruction gain of 1.29 dB.