Motion estimation using chrominance for wavelet-based video coding

We propose an improved block-matching-based motion estimation technique, which takes advantage of the chrominance information of the video sequence in order to reduce the number of mismatches. The method is adapted to different algorithms. According to the experimental results, it helps improving the quality of the reconstructed video sequence, both in terms of pSNR and visual quality. We also propose a new algorithm in order to reduce the complexity overhead due to the processing of the chrominance information. We show that this fast algorithm offers the same reconstruction quality as the corresponding optimal algorithm, for a significantly lower computational cost. 1