Method for signaling of mixed NAL unit type in coded video stream
A method, computer program, and computer system is provided for coding video data. Video data including one or more subpictures is received. A network abstraction layer (NAL) unit type associated with each of the one or more subpictures is identified based on checking a flag corresponding to mixed NAL units in the one or more subpictures. The video data is decoded based on the identified NAL unit types.
1 . A method for decoding video data, executable by a processor, comprising:
receiving video data comprising one or more pictures;
determining whether motion vector scaling is disabled across multiple pictures, the multiple pictures comprising the one or more pictures;
based on determining that motion vector scaling is not disabled, determining a motion vector scaling process for a current picture based on at least one of an access unit associated with the current picture or a picture order count (POC) associated with the current picture; and
decoding the video data based on the determined motion vector scaling process and whether an POC cycle syntax, which is a poc_cycle_au syntax, is signaled with the video data and indicating a POC value used for all of the multiple pictures, a first value of the poc_cycle_au syntax indicates that the POC value increases uniformly per access unit (AU), a slice_poc_cycle_au syntax is not explicitly signaled, and a value of an access unit count (AUC) for each AU is calculated by dividing a value of the POC by the poc_cycle_au syntax.
2 . The method of claim 1 , wherein a second value of the poc_cycle_au syntax indicates that the POC value does not increase uniformly per AU, and in a case of the second value, a access_unit_cnt syntax is not signaled with the video data, a slice_access_unit_cnt syntax is signaled in a slice header for each slice or picture, and the value of the AUC is calculated by dividing a value of the POC by a slice_poc_cycle_au syntax.
3 . The method of claim 2 , wherein determining whether motion vector scaling is disabled comprises determining that motion vector scaling is disabled across multiple pictures associated with a same access unit.
4 . The method of claim 3 , wherein based on determining that motion vector scaling is disabled across multiple pictures associated with the same access unit, the method further comprises:
determining that a reference picture associated with the current picture is associated with the same access unit as the current picture; and
based on determining that the reference picture associated with the current picture is associated with the same access unit as the current picture, disabling the motion vector scaling process for temporal motion vector prediction for the current picture.
5 . The method of claim 4 , wherein disabling that the motion vector scaling process for the temporal motion vector prediction for the current picture comprises a motion vector scaling function for the current picture returning a value of one.
6 . A method for encoding video data, executable by a processor, comprising:
receiving video data comprising one or more pictures;
determining whether motion vector scaling is to be disabled across multiple pictures, the multiple pictures comprising the one or more pictures;
based on motion vector scaling to not be disabled, determining a motion vector scaling process for a current picture based on at least one of an access unit associated with the current picture or a picture order count (POC) associated with the current picture; and
encoding the video data based on the determined motion vector scaling process and whether an POC cycle syntax, which is a poc_cycle_au syntax, is to be signaled with the video data and indicating a POC value used for all of the multiple pictures, a first value of the poc_cycle_au syntax indicates that the POC value increases uniformly per access unit (AU), a slice_poc_cycle_au syntax is not explicitly signaled, and a value of an access unit count (AUC) for each AU is calculated by dividing a value of the POC by the poc_cycle_au syntax.
7 . The method of claim 6 , wherein a second value of the poc_cycle_au syntax indicates that the POC value does not increase uniformly per AU, and in a case of the second value, a access_unit_cnt syntax is not signaled with the video data, a slice_access_unit_cnt syntax is signaled in a slice header for each slice or picture, and the value of the AUC is calculated by dividing a value of the POC by a slice_poc_cycle_au syntax.
8 . The method of claim 7 , wherein determining whether motion vector scaling is disabled comprises determining that motion vector scaling is disabled across multiple pictures associated with a same access unit.
9 . The method of claim 8 , wherein based on determining that motion vector scaling is disabled across multiple pictures associated with the same access unit, the method further comprises:
determining that a reference picture associated with the current picture is associated with the same access unit as the current picture; and
based on determining that the reference picture associated with the current picture is associated with the same access unit as the current picture, disabling the motion vector scaling process for temporal motion vector prediction for the current picture.
10 . The method of claim 9 , wherein disabling that the motion vector scaling process for the temporal motion vector prediction for the current picture comprises a motion vector scaling function for the current picture returning a value of one.
11 . A method of processing visual media data, executable by a processor, comprising:
performing a conversion between a visual media file and a bitstream of a visual media data according to a format rule,
wherein the bitstream indicates whether motion vector scaling is disabled across multiple pictures, the multiple pictures comprising the one or more pictures,
wherein the bitstream further indicates, based on motion vector scaling not being disabled, a motion vector scaling process for a current picture based on at least one of an access unit associated with the current picture or a picture order count (POC) associated with the current picture, and
wherein the bitstream further indicates the visual media data based on the motion vector scaling process and whether an POC cycle syntax, which is a poc_cycle_au syntax, is signaled with the video data and indicating a POC value used for all of the multiple pictures, a first value of the poc_cycle_au syntax indicates that the POC value increases uniformly per access unit (AU), a slice_poc_cycle_au syntax is not explicitly signaled, and a value of an access unit count (AUC) for each AU is calculated by dividing a value of the POC by the poc_cycle_au syntax.
12 . The method of claim 11 , wherein a second value of the poc_cycle_au syntax indicates that the POC value does not increase uniformly per AU, and in a case of the second value, a access_unit_cnt syntax is not signaled with the video data, a slice_access_unit_cnt syntax is signaled in a slice header for each slice or picture, and the value of the AUC is calculated by dividing a value of the POC by a slice_poc_cycle_au syntax.
13 . The method of claim 12 , wherein whether motion vector scaling is disabled is based on that motion vector scaling being disabled across multiple pictures associated with a same access unit.
14 . The method of claim 13 , wherein the bitstream indicates, based on the motion vector scaling being disabled across multiple pictures associated with the same access unit, that:
a reference picture associated with the current picture is associated with the same access unit as the current picture,
based on the reference picture associated with the current picture being associated with the same access unit as the current picture, the motion vector scaling process is disabled for temporal motion vector prediction for the current picture.