Flexible Macroblock Ordering (FMO) is one of the new error resilience tools introduced in H264/AVC. Several slice grouping methods have been studied for improving error robustness using FMO. In this paper, a simple and fast slice grouping method for inter frames is introduced. Fast mode decision and early Skip Mode decision are applied for the first encoding pass, and only the features that are available at the stage of early Skip Mode decision are used for the classification. The computation time cost can be reduced by about 50% on average compared to traditional methods. The proposed scheme is tested under the proposed Unequal Error Protection scheme at the DVB-H link layer. The results are compared to the standard MPE-FEC EEP scheme using traditional FMO type 'interleaved' at the DVB-H link layer. It is shown that the proposed scheme can provide improved error robustness for high error rate channels in a DVB-H system.
Slice group coding using Flexible Macroblock Ordering (FMO) based on Macroblock (MB) importance in H.264/AVC has been studied for providing higher error robustness in error prone environments. However, the efficiency still needs to be improved and the high computational and time cost it causes remains to be an issue. In this paper, an Adaptive Macroblock-to-slice Allocation map (MBAmap) Updating scheme is proposed based on our previous research work. The extra computational and time cost introduced is less than 10% compared to the conventional methods. The PSNR of the reconstructed video at the decoder side can be improved for up to 0.6dB compared to previous proposed method. A Slice Matching (SM) error concealment method is also introduced, which increases the PSNR for a further 0.2 to 0.4dB when applied with the slice group coding scheme.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.