Error concealment via Kalman filter for heavily corrupted videos in H.264/AVC

Error concealment at the decoder side is an economical approach to ensuring an acceptable and stable video quality in case of packet erasure or loss, and thus it has attracted lots of research interest. Current techniques mainly employ the spatial or temporal correlation to predict the motion vectors (MVs) of the missing blocks, and interpolation, extrapolation or boundary matching schemes are usually effective. However, for heavily corrupted sequences, e.g., with macroblock loss rate beyond 50%, most methods might perform less satisfactorily. Inspired by the tracking efficiency of Kalman filter (KF), in the present work, we adopted it to predict the missing MVs, and the unpredicted ones (minority) were recovered complementarily using the bilinear motion field interpolation (MFI) method. Since the KF prediction is independent of the loss rate, the present framework is especially robust for heavily corrupted videos. Experimental results on typical sequences reveal that the proposed algorithm outperforms the boundary matching algorithm embedded in the H.264/AVC reference code, the MFI algorithm in the literature, and some other existing techniques by up to about 5.68 dB.