Method of decoding intra prediction data

A method of decoding intra prediction data, the method comprising:

• determining whether a selected intra prediction mode is included in one of:
• a first set including most probable mode prediction candidates,
• a second set including one or more prediction modes that are not most probable mode prediction candidates, or
• a third set including one or more prediction modes that are not most probable mode prediction candidates and not included in the second set;
• deriving most probable mode candidates, wherein deriving most probable mode candidates includes determining the most probable mode candidates based on at least one condition indicating whether an intra prediction mode of an available left neighboring prediction unit is equal to an intra predication mode of an available above neighboring prediction unit;
• in a case where the selected intra prediction mode is included in the first set:
• parsing a first index value indicating which of the most probable mode candidates corresponds to the selected intra prediction mode;
• in a case where the selected intra prediction mode is included in the third set:
• parsing a second index value indicating the selected intra prediction mode, wherein the selected intra prediction mode is derived based on the second index value;
• generating a prediction block using the selected intra prediction mode,
• wherein generating the prediction block includes filtering reference samples based on the prediction mode which is a planar prediction mode or an above-left diagonal prediction mode,
• wherein the reference samples are filtered in a case that size of the prediction block is greater than 32×32.