Adaptive Wavelet Restoration of Noisy Video Sequences (MP-P5)

Author(s) : Nasir Rajpoot (University of Warwick, UK) Zhen Yao (University of Warwick, UK) Roland Wilson (University of Warwick, UK)

Abstract : In this paper, we report work on a novel algorithm for restoration of noisy video sequences by thresholding in the transform domain. A video sequence is first transformed into an optimal $3$D wavelet domain using basis functions adapted to the contents of the sequence. Assuming that all the major spatiotemporal frequency phenomena present in the sequence would produce high amplitude transform coefficients, a modified form of the BayesShrink thresholding method is used to suppress the noise. In order to reduce the effects of Gibbs phenomenon in the restored sequence, translation dependence is removed by averaging the restored instances of the shifted sequence. The algorithm yields promising results in terms of both objective and subjective quality of the restored sequence.

Menu