JP-2022537664-A5 -

Dates

Publication Date

20230518

Application Date

20200602

Description

[0272] Example 72. A method for encoding video data, the method comprising: identifying at least a portion of a picture; determining that a weighted prediction is selected that is valid for the portion of the picture; identifying a reference picture for the portion of the picture; generating a zero-value picture order count offset indicating the reference picture from a list of reference pictures; generating a bitstream, the bitstream comprising the portion of the picture and a zero-value picture order count offset associated with the portion of the picture. The invention described in the original claims of this application is listed below. [C1] A device for decoding video data, wherein the device is Memory and The circuit comprises a processor implemented in the circuit, and the processor is To obtain at least a portion of the picture contained in the bitstream, From the bitstream, it is determined that weighted prediction is enabled for at least the portion of the picture, Based on the decision that weighted prediction is enabled for at least the portion of the picture, identify a zero-value picture order count offset indicating a reference picture from the reference picture list, An apparatus configured to reconstruct at least a portion of a reference picture using at least a portion of the reference picture identified by the zero-value picture sequence count offset. [C2] The aforementioned processor, Obtaining a plurality of portions of the picture from the bitstream, and the plurality of portions including at least the portion of the picture reconstructed using the reference picture identified by the zero-value picture sequence count offset, Identifying a plurality of corresponding picture sequence count offsets for the plurality of parts of the picture, wherein the zero-value picture sequence count offset is a corresponding picture sequence count offset among the plurality of corresponding picture sequence count offsets associated with the reference picture. Reconstructing the picture using the multiple reference pictures identified by the multiple corresponding picture sequence count offsets. The apparatus described in C1, further configured to perform the following. [C3] The apparatus according to C1, wherein the processor is further configured to determine that weighted prediction is enabled by parsing the bitstream to identify one or more weighted prediction flags for the picture. [C4] The apparatus according to C3, wherein the one or more weighted prediction flags for the portion of the picture include a sequence parameter set weighted prediction flag and a picture parameter set weighted prediction flag. [C5] The apparatus according to C4, wherein the sequence parameter set weighting prediction flag and the picture parameter set weighting prediction flag are flags for unidirectionally predicted frames. [C6] The apparatus according to C4, wherein the sequence parameter set weighting prediction flag and the picture parameter set weighting prediction flag are flags for bidirectional prediction frames. [C7] The apparatus according to C4, wherein the picture parameter set weighting prediction flag is constrained by the sequence parameter set weighting prediction flag. [C8] The aforementioned processor, To obtain at least a portion of the second picture contained in the bitstream, From the bitstream, it is determined that the weighted prediction is invalid for the portion of the second picture, Based on the decision that weighted predictions are disabled for the portion of the second picture, it is determined that a second zero-value picture order count offset is not possible, such that it indicates a second reference picture from the reference picture list for the portion of the second picture. The apparatus described in C1, further configured to perform the following. [C9] The aforementioned processor, Parsing the syntax element of the bitstream indicating the picture order count offset for the second reference picture by generating the reconstructed value of the syntax element as the signaled value + 1, based on the decision that the second zero-value picture order count offset is not made possible and the weighted prediction is disabled for the second picture, Reconstructing the second picture using the reconstructed values of the syntax elements The apparatus described in C8, further configured to perform the following. [C10] The apparatus according to C8, wherein, when weighted prediction is disabled, the value of the reference picture syntax element specifies the absolute difference between the picture order count value of the second picture and the previous reference picture entry in the reference picture list for the second reference picture as the signaled value + 1. [C11] The reference picture relates to a first set of weights, and the picture relates to a second set of weights different from the first set of weights. Herein, the reference picture has a different size from the picture, Here,