CPC H04N 19/593 (2014.11) | 15 Claims |
1. A video decoding method, by a decoding apparatus, the method comprising:
acquiring long term reference picture (LTRP) information from a bitstream;
configuring a reference picture set based on the LTRP information;
performing an inter prediction on a current block in a current picture to generate predicted samples based on an LTRP in the reference picture set; and
generating reconstructed samples based on the predicted samples,
wherein the LTRP information comprises a first LTRP information included in a sequence parameter set (SPS) of the bitstream and a second LTRP information included in a slice header of the bitstream;
wherein the LTRP is one of candidate LTRPs specified by the first LTRP information,
wherein the reference picture set comprises an LTRP set including at least one LTRP,
wherein the first LTRP information includes a long term reference pictures present flag,
wherein when a value of the long term reference pictures present flag is equal to 1, the first LTRP information includes an LTRP number syntax element representing a number of the candidate LTRPs specified in the SPS and picture order count (POC) least significant bit (LSB) information representing an LSB of a POC of each of the candidate LTRPs,
wherein the first LTRP information includes used by current picture long term SPS flag information representing whether each of the candidate LTRPs specified in the SPS is not used for a reference picture by the current picture, and
wherein the second LTRP information includes LTRP index information representing an LTRP to be included in the LTRP set among the candidate LTRPs specified in the SPS.
|