A New Shape Error Concealment Method

Based on the rotation invariant properties of both OurHarris interest point detector and local pseudo-Zernike moments, a rotation robust shape error concealment scheme is proposed. Firstly, the best matching pairs of interest points between two objects are computed by comparing the Euclidean distance of local pseudo-Zernike moments defined on the interest point neighborhood. Then, the global motion parameters are determined and the previous boundary is motion compensated. Finally, the missing boundary pieces are reconstructed using boundary matching and compensation method. Comparisons with traditional scheme are presented and the results demonstrating the performance of the proposed method.