codec/h264/h264dec/sps.go: commented fields of syntax structures

This commit is contained in:
Saxon 2019-07-31 20:25:38 +09:30
parent 1d6c501bb8
commit 7a4ee894d4
1 changed files with 342 additions and 83 deletions

View File

@ -63,49 +63,156 @@ var (
// SPS describes a sequence parameter set as defined by section 7.3.2.1.1 in // SPS describes a sequence parameter set as defined by section 7.3.2.1.1 in
// the Specifications. // the Specifications.
// For semantics see section 7.4.2.1. Comments for fields are excerpts from
// section 7.4.2.1.
type SPS struct { type SPS struct {
Profile uint8 // pofile_idx and level_idc indicate the profile and level to which the
Constraint0 bool // coded video sequence conforms.
Constraint1 bool Profile, LevelIDC uint8
Constraint2 bool
Constraint3 bool // The constraint_setx_flag flags specify the constraints defined in A.2 for
Constraint4 bool // which this stream conforms.
Constraint5 bool Constraint0 bool
LevelIDC uint8 Constraint1 bool
SPSID uint64 Constraint2 bool
ChromaFormatIDC uint64 Constraint3 bool
SeparateColorPlaneFlag bool Constraint4 bool
BitDepthLumaMinus8 uint64 Constraint5 bool
BitDepthChromaMinus8 uint64
// seq_parameter_set_id identifies this sequence parameter set, and can then
// be reference by the picture parameter set. The seq_parameter_set_id is
// in the range of 0 to 30 inclusive.
SPSID uint64
// chroma_format_idc specifies the chroma sampling relative to the luma
// sampling as specified in caluse 6.2. Range of chroma_format_idc is in
// from 0 to 3 inclusive.
ChromaFormatIDC uint64
// separate_color_plane_flag if true specifies that the three components of
// the 4:4:4 chroma formta are coded separately.
SeparateColorPlaneFlag bool
// bit_depth_luma_minus8 specifies the luma array sample bit depth and the
// luma quantisation parameter range offset QpBdOffset_y (eq 7-3 and 7-4).
BitDepthLumaMinus8 uint64
// bit_depth_luma_minus8 specifies the chroma array sample bit depth and the
// chroma quantisation parameter range offset QpBdOffset_c (eq 7-3 and 7-4).
BitDepthChromaMinus8 uint64
// qpprime_y_zero_transform_bypass_flag equal to 1 specifies that, when QP Y
// is equal to 0, a transform bypass operation for the transform coefficient
// decoding process and picture construction process prior to deblocking
// filter process as specified in clause 8.5 shall be applied.
QPPrimeYZeroTransformBypassFlag bool QPPrimeYZeroTransformBypassFlag bool
SeqScalingMatrixPresentFlag bool
SeqScalingListPresentFlag []bool // seq_scaling_matrix_present_flag equal to 1 specifies that
ScalingList4x4 [][]uint64 // seq_scaling_list_present_flag[ i ] are present. When 0 they are not present
UseDefaultScalingMatrix4x4Flag []bool // and the sequence-level scaling lists specified by Flat_4x4_16 and
ScalingList8x8 [][]uint64 // Flat_8x8_16 shall be inferred.
UseDefaultScalingMatrix8x8Flag []bool SeqScalingMatrixPresentFlag bool
Log2MaxFrameNumMinus4 uint64
PicOrderCountType uint64 // seq_scaling_lit_present_flag[i] specifics whether the syntax structure for
Log2MaxPicOrderCntLSBMin4 uint64 // scaling list i is present. If 1 then present, otherwise not, and scaling
DeltaPicOrderAlwaysZeroFlag bool // list for i is inferred as per rule set A in table 7-2.
OffsetForNonRefPic int64 SeqScalingListPresentFlag []bool
OffsetForTopToBottomField int64
NumRefFramesInPicOrderCntCycle uint64 // The 4x4 sequence scaling lists for each i.
OffsetForRefFrameList []int ScalingList4x4 [][]uint64
MaxNumRefFrames uint64
GapsInFrameNumValueAllowed bool // Flag to indicate for a 4x4 scaling list, if we use the default.
PicWidthInMBSMinus1 uint64 UseDefaultScalingMatrix4x4Flag []bool
PicHeightInMapUnitsMinus1 uint64
FrameMBSOnlyFlag bool // The 8x8 sequence scaling lists for each i.
MBAdaptiveFrameFieldFlag bool ScalingList8x8 [][]uint64
Direct8x8InferenceFlag bool
FrameCroppingFlag bool // Flag to indicate for a 8x8 scaling list, if we use the default.
FrameCropLeftOffset uint64 UseDefaultScalingMatrix8x8Flag []bool
FrameCropRightOffset uint64
FrameCropTopOffset uint64 // log2_max_frame_num_minus4 allows for derivation of MaxFrameNum using eq 7-10.
FrameCropBottomOffset uint64 Log2MaxFrameNumMinus4 uint64
VUIParametersPresentFlag bool
VUIParameters *VUIParameters // pic_order_cnt_type specifiess the method to decode picture order count.
PicOrderCountType uint64
// log2_max_pic_order_cnt_lsb_minus4 allows for the dreivation of
// MaxPicOrderCntLsb using eq 7-11.
Log2MaxPicOrderCntLSBMin4 uint64
// delta_pic_order_always_zero_flag if true indicates delta_pic_order_cnt[0]
// and delta_pic_order_cnt[1].
DeltaPicOrderAlwaysZeroFlag bool
// offset_for_non_ref_pic is used to calculate the picture order count of a
// non-reference picture as specified in clause 8.2.1.
OffsetForNonRefPic int64
// offset_for_top_to_bottom_field is used to calculate the picture order count
// of a bottom field as specified in clause 8.2.1.
OffsetForTopToBottomField int64
// num_ref_frames_in_pic_order_cnt_cycle is used in the decoding process for
// picture order count as specified in clause 8.2.1.
NumRefFramesInPicOrderCntCycle uint64
// offset_for_ref_frame[ i ] is an element of a list of
// num_ref_frames_in_pic_order_cnt_cycle values used in the decoding process
// for picture order count as specified in clause 8.2.1.
OffsetForRefFrameList []int
// max_num_ref_frames specifies the max number of short-term and long-term
// reference frames, complementary reference field pairs, and non-paired
// reference fields that may be used by the decoding process for inter prediction.
MaxNumRefFrames uint64
// gaps_in_frame_num_value_allowed_flag specifies the allowed values of
// frame_num as specified in clause 7.4.3 and the decoding process in case of
// an inferred gap between values of frame_num as specified in clause 8.2.5.2.
GapsInFrameNumValueAllowed bool
// pic_width_in_mbs_minus1 plus 1 specifies the width of each decode picutre
// in units of macroblocks. See eq 7-13.
PicWidthInMBSMinus1 uint64
// pic_height_in_map_units_minus1 plus 1 specifies the height in slice group
// map units of a decoded frame or field. See eq 7-16.
PicHeightInMapUnitsMinus1 uint64
// frame_mbs_only_flag if 0 coded pictures of the coded video sequence may be
// coded fields or coded frames. If 1 every coded picture of the coded video
// sequence is a coded frame containing only frame macroblocks.
FrameMBSOnlyFlag bool
// mb_adaptive_frame_field_flag if 0 specifies no switching between
// frame and field macroblocks within a picture. If 1 specifies the possible
// use of switching between frame and field macroblocks within frames.
MBAdaptiveFrameFieldFlag bool
// direct_8x8_inference_flag specifies the method used in the derivation
// process for luma motion vectors for B_Skip, B_Direct_16x16 and B_Direct_8x8
// as specified in clause 8.4.1.2.
Direct8x8InferenceFlag bool
// frame_cropping_flag if 1 then frame cropping offset parameters are next in
// the sequence parameter set. If 0 they are not.
FrameCroppingFlag bool
// frame_crop_left_offset, frame_crop_right_offset, frame_crop_top_offset,
// frame_crop_bottom_offset specify the samples of the pictures in the coded
// video sequence that are output from the decoding process, in terms of a
// rectangular region specified in frame coordinates for output.
FrameCropLeftOffset uint64
FrameCropRightOffset uint64
FrameCropTopOffset uint64
FrameCropBottomOffset uint64
// vui_parameters_present_flag if 1 the vui_parameters() syntax structure is
// present, otherwise it is not.
VUIParametersPresentFlag bool
// The vui_parameters() syntax structure specified in appendix E.
VUIParameters *VUIParameters
} }
// NewSPS parses a sequence parameter set raw byte sequence from br following // NewSPS parses a sequence parameter set raw byte sequence from br following
@ -225,41 +332,158 @@ func NewSPS(rbsp []byte, showPacket bool) (*SPS, error) {
// SPS describes a sequence parameter set as defined by section E.1.1 in the // SPS describes a sequence parameter set as defined by section E.1.1 in the
// Specifications. // Specifications.
// Semantics for fields are define in section E.2.1. Comments on fields are
// excerpts from the this section.
type VUIParameters struct { type VUIParameters struct {
AspectRatioInfoPresentFlag bool // aspect_ratio_info_present_flag if 1 then aspect_ratio_idc is present,
AspectRatioIDC uint8 // otherwsise is not.
SARWidth uint32 AspectRatioInfoPresentFlag bool
SARHeight uint32
OverscanInfoPresentFlag bool // aspect_ratio_idc specifies the value of sample aspect ratio of the luma samples.
OverscanAppropriateFlag bool AspectRatioIDC uint8
VideoSignalTypePresentFlag bool
VideoFormat uint8 // sar_width indicates the horizontal size of the sample aspect ratio (in
VideoFullRangeFlag bool // arbitrary units).
ColorDescriptionPresentFlag bool SARWidth uint32
ColorPrimaries uint8
TransferCharacteristics uint8 // sar_height indicates the vertical size of the sample aspect ratio (in the
MatrixCoefficients uint8 // same arbitrary units as sar_width).
ChromaLocInfoPresentFlag bool SARHeight uint32
ChromaSampleLocTypeTopField uint64
ChromaSampleLocTypeBottomField uint64 // overscan_info_present_flag if 1 then overscan_appropriate_flag is present,
TimingInfoPresentFlag bool // otherwise if 0, then the display method for the video signal is unspecified.
NumUnitsInTick uint32 OverscanInfoPresentFlag bool
TimeScale uint32
FixedFrameRateFlag bool // overscan_appropriate_flag if 1 then the cropped decoded pictures output
NALHRDParametersPresentFlag bool // are suitable for display using overscan, othersise if 0, then the cropped
NALHRDParameters *HRDParameters // decoded pictures output should not be displayed using overscan.
VCLHRDParametersPresentFlag bool OverscanAppropriateFlag bool
VCLHRDParameters *HRDParameters
LowDelayHRDFlag bool // video_signal_type_present_flag equal to 1 specifies that video_format,
PicStructPresentFlag bool // video_full_range_flag and colour_description_present_flag are present,
BitstreamRestrictionFlag bool // otherwise if 0, then they are not present.
VideoSignalTypePresentFlag bool
// video_format indicates the representation of the pictures as specified in
// Table E-2, before being coded in accordance with this Recommendation |
// International Standard.
VideoFormat uint8
// video_full_range_flag indicates the black level and range of the luma and
// chroma signals as derived from E_Y, E_PB, and E_PR or E_R, E_G,
// and E_B real-valued component signals.
VideoFullRangeFlag bool
// colour_description_present_flag if 1 specifies that colour_primaries,
// transfer_characteristics and matrix_coefficients are present, otherwise if
// 0 then they are not present.
ColorDescriptionPresentFlag bool
// colour_primaries indicates the chromaticity coordinates of the source
// primaries as specified in Table E-3 in terms of the CIE 1931 definition of
// x and y as specified by ISO 11664-1.
ColorPrimaries uint8
// transfer_characteristics either indicates the reference opto-electronic
// transfer characteristic function of the source picture, or indicates the
// inverse of the reference electro-optical transfer characteristic function.
TransferCharacteristics uint8
// matrix_coefficients describes the matrix coefficients used in deriving luma
// and chroma signals from the green, blue, and red, or Y, Z, and X primaries,
// as specified in Table E-5.
MatrixCoefficients uint8
// chroma_loc_info_present_flag if 1 specifies that chroma_sample_loc_type_top_field
// and chroma_sample_loc_type_bottom_field are present, otherwise if 0,
// they are not present.
ChromaLocInfoPresentFlag bool
// chroma_sample_loc_type_top_field and chroma_sample_loc_type_bottom_field
// specify the location of chroma samples.
ChromaSampleLocTypeTopField, ChromaSampleLocTypeBottomField uint64
// timing_info_present_flag if 1 specifies that num_units_in_tick, time_scale
// and fixed_frame_rate_flag are present in the bitstream, otherwise if 0,
// they are not present.
TimingInfoPresentFlag bool
// num_units_in_tick is the number of time units of a clock operating at the
// frequency time_scale Hz that corresponds to one increment (called a clock
// tick) of a clock tick counter.
NumUnitsInTick uint32
// time_scale is the number of time units that pass in one second.
TimeScale uint32
// fixed_frame_rate_flag if 1 indicates that the temporal distance
// between the HRD output times of any two consecutive pictures in output
// order is constrained as follows. fixed_frame_rate_flag equal to 0 indicates
// that no such constraints apply to the temporal distance between the HRD
// output times of any two consecutive pictures in output order.
FixedFrameRateFlag bool
// nal_hrd_parameters_present_flag if 1 then NAL HRD parameters (pertaining to
// Type II bitstream conformance) are present, otherwise if 0, then they
// are not present.
NALHRDParametersPresentFlag bool
// The nal_hrd_parameters() syntax structure as specified in section E.1.2.
NALHRDParameters *HRDParameters
// vcl_hrd_parameters_present_flag if 1 specifies that VCL HRD parameters
// (pertaining to all bitstream conformance) are present, otherwise if 0, then
// they are not present.
VCLHRDParametersPresentFlag bool
// The vcl_nal_hrd_parameters() syntax structure as specified in section E.1.2.
VCLHRDParameters *HRDParameters
// low_delay_hrd_flag specifies the HRD operational mode as specified in Annex C.
LowDelayHRDFlag bool
// pic_struct_present_flag if 1 then picture timing SEI messages (clause D.2.3)
// are present that include the pic_struct syntax element, otherwise if 0, then
// not present.
PicStructPresentFlag bool
// bitstream_restriction_flag if 1, then the following coded video sequence
// bitstream restriction parameters are present, otherwise if 0, then they are
// not present.
BitstreamRestrictionFlag bool
// motion_vectors_over_pic_boundaries_flag if 0 then no sample outside the
// picture boundaries and no sample at a fractional sample position for which
// the sample value is derived using one or more samples outside the picture
// boundaries is used for inter prediction of any sample, otherwise if 1,
// indicates that one or more samples outside picture boundaries may be used
// in inter prediction.
MotionVectorsOverPicBoundariesFlag bool MotionVectorsOverPicBoundariesFlag bool
MaxBytesPerPicDenom uint64
MaxBitsPerMBDenom uint64 // max_bytes_per_pic_denom indicates a number of bytes not exceeded by the sum
Log2MaxMVLengthHorizontal uint64 // of the sizes of the VCL NAL units associated with any coded picture in the
Log2MaxMVLengthVertical uint64 // coded video sequence.
MaxNumReorderFrames uint64 MaxBytesPerPicDenom uint64
MaxDecFrameBuffering uint64
// max_bits_per_mb_denom indicates an upper bound for the number of coded bits
// of macroblock_layer() data for any macroblock in any picture of the coded
// video sequence.
MaxBitsPerMBDenom uint64
// log2_max_mv_length_horizontal and log2_max_mv_length_vertical indicate the
// maximum absolute value of a decoded horizontal and vertical motion vector
// component, respectively, in 14 luma sample units, for all pictures in the
// coded video sequence.
Log2MaxMVLengthHorizontal, Log2MaxMVLengthVertical uint64
// max_num_reorder_frames indicates an upper bound for the number of frames
// buffers, in the decoded picture buffer (DPB), that are required for storing
// frames, complementary field pairs, and non-paired fields before output.
MaxNumReorderFrames uint64
// max_dec_frame_buffering specifies the required size of the HRD decoded
// picture buffer (DPB) in units of frame buffers.
MaxDecFrameBuffering uint64
} }
// NewVUIParameters parses video usability information parameters from br // NewVUIParameters parses video usability information parameters from br
@ -357,17 +581,52 @@ func NewVUIParameters(br *bits.BitReader) (*VUIParameters, error) {
// HRDParameters describes hypothetical reference decoder parameters as defined // HRDParameters describes hypothetical reference decoder parameters as defined
// by section E.1.2 in the specifications. // by section E.1.2 in the specifications.
// Field semantics are defined in section E.2.2. Comments on fields are excerpts
// from section E.2.2.
type HRDParameters struct { type HRDParameters struct {
CPBCntMinus1 uint64 // cpb_cnt_minus1 plus 1 specifies the number of alternative CPB specifications
BitRateScale uint8 // in the bitstream.
CPBSizeScale uint8 CPBCntMinus1 uint64
BitRateValueMinus1 []uint64
CPBSizeValueMinus1 []uint64 // bit_rate_scale (together with bit_rate_value_minus1[ SchedSelIdx ])
CBRFlag []bool // specifies the maximum input bit rate of the SchedSelIdx-th CPB.
BitRateScale uint8
// cpb_size_scale (together with cpb_size_value_minus1[ SchedSelIdx ])
// specifies the CPB size of the SchedSelIdx-th CPB.
CPBSizeScale uint8
// bit_rate_value_minus1[ SchedSelIdx ] (together with bit_rate_scale)
//specifies the maximum input bit rate for the SchedSelIdx-th CPB.
BitRateValueMinus1 []uint64
// cpb_size_value_minus1[ SchedSelIdx ] is used together with cpb_size_scale
// to specify the SchedSelIdx-th CPB size.
CPBSizeValueMinus1 []uint64
// cbr_flag[ SchedSelIdx ] equal to 0 specifies that to decode this bitstream
// by the HRD using the SchedSelIdx-th CPB specification, the hypothetical
// stream delivery scheduler (HSS) operates in an intermittent bit rate mode,
// otherwise if 1 specifies that the HSS operates in a constant bit rate mode.
CBRFlag []bool
// initial_cpb_removal_delay_length_minus1 specifies the length in bits of the
// initial_cpb_removal_delay[ SchedSelIdx ] and
// initial_cpb_removal_delay_offset[ SchedSelIdx ] syntax elements of the
// buffering period SEI message.
InitialCPBRemovalDelayLenMinus1 uint8 InitialCPBRemovalDelayLenMinus1 uint8
CPBRemovalDelayLenMinus1 uint8
DPBOutputDelayLenMinus1 uint8 // cpb_removal_delay_length_minus1 specifies the length in bits of the
TimeOffsetLen uint8 // cpb_removal_delay syntax element.
CPBRemovalDelayLenMinus1 uint8
// dpb_output_delay_length_minus1 specifies the length in bits of the
// dpb_output_delay syntax element.
DPBOutputDelayLenMinus1 uint8
// time_offset_length greater than 0 specifies the length in bits of the
// time_offset syntax element.
TimeOffsetLen uint8
} }
// NewHRDParameters parses hypothetical reference decoder parameter from br // NewHRDParameters parses hypothetical reference decoder parameter from br