当前位置： 首页 > news >正文

Mac 环境 VVenC 编译与编码命令行工具使用教程

news 2025/7/15 4:29:57

VVenC

VVenC 是一个开源的高效视频编码器，专门用于支持 H.266/VVC (Versatile Video Coding) 标准的编码。H.266/VVC 是继 HEVC (H.265) 之后的新一代视频编码标准，主要目的是提供比 HEVC 更高的压缩效率，同时保持或提高视频质量。H.266/VVC 可以在相同的视频质量下减少大约 50% 的比特率，或者在相同的比特率下提供更好的视频质量。

VVenC 特性

  • 高压缩效率：
    VVenC 实现了 H.266/VVC 标准，可以在相同的画质下提供比 H.265/HEVC 更高的压缩效率。对于流媒体、广播电视和视频存储等场景，VVenC 能显著降低带宽需求。

  • 灵活性和扩展性：
    VVenC 支持 H.266/VVC 标准的所有特性，包括可变帧率、可变分辨率等，适用于多种应用场景，如4K、8K视频编码。

  • 开源与跨平台：
    VVenC 是一个开源项目，允许开发者进行自由修改和分发，同时支持多个平台（Linux、Windows、macOS）。

  • 编码速度与质量优化：
    VVenC 提供了不同的编码速度设置，可以根据需要在速度和质量之间进行权衡，满足实时编码或高质量编码的需求。

VVenC 编译

  1. 源码下载：https://github.com/fraunhoferhhi/vvenc
  2. 编译：比较简单，按照 README中用 CMake 编译即可。
mkdir build
cd build
cmake ..
make
  1. 编译后文件：
    • bin/release-static：vvencFFapp、vvencapp、vvencinterfacetest、vvenclibtest
    • include/vvenc：vvenc.h、vvencCfg.h、vvencDecl.h
    • lib/release-static：libvvenc.a

编码工具使用

  1. vvencFFapp 使用：全特征的专家级编码器
admin@bogon release-static % ./vvencFFapp
vvencFFapp: VVenC, the Fraunhofer H.266/VVC Encoder, version 1.12.0 [Mac OS X][clang 15.0.0][64 bit][SIMD=NEON]#======== General Options ================-h,   --help [0]                   show default help--fullhelp [0]               show full help-v,   --Verbosity [verbose]        verbosity level (0: silent, 1: error, 2: warning, 3: info, 4: notice, 5: verbose,6: debug)--stats [1]                  enable or disable printing of statistics (fps, bitrate, estimation of encodingtime)--version [0]                show version #======== Input Options ================-i,   --InputFile []               Original YUV input file name or '-' for reading from stdin-s,   --Size [0x0]                 Input resolution (WidthxHeight)--InputBitDepth [8]          Bit-depth of input file-f,   --FramesToBeEncoded [0]      Number of frames to be encoded (default=all)-fr,  --FrameRate [0]              Temporal rate (framerate numerator) e.g. 25,30, 30000, 50,60, 60000--FrameScale [1]             Temporal scale (framerate denominator) e.g. 1, 1001--fps [0/1]                  Framerate as int or fraction (num/denom) --TicksPerSecond [27000000]  Ticks Per Second for dts generation, (1..27000000, -1: ticks per frame=1)--LeadFrames [0]             Number of leading frames to be read before starting the encoding, use whensplitting the video into overlapping segments--TrailFrames [0]            Number of trailing frames to be read after frames to be encoded, use when splittingthe video into overlapping segments-fs,  --FrameSkip [0]              number of frames to skip at start of input YUV [off]--segment [off]              when encoding multiple separate segments, specify segment position to enable segment concatenation (first, mid, last) [off]first: first segment           mid  : all segments between first and last segmentlast : last segment#======== Output options ================-b,   --BitstreamFile []           Bitstream output file name-o,   --ReconFile []               Reconstructed YUV output file name--OutputBitDepth [0]         Bit-depth of output file#======== Threading, performance ================-t,   --Threads [0]                number of threads (multithreading; -1: resolution < 720p: 4, < 5K 2880p: 8, >= 5K2880p: 12 threads)--MTProfile [off]            set automatic multi-threading setting (-1: auto, 0: off, 1,2,3: on, enables tiles,IFP and WPP automatically depending on the number of threads)--preset [medium]            select preset for specific encoding setting (faster, fast, medium, slow, slower,medium_lowDecEnergy)--Tiles [-1x-1]              Set number of tile columns and rows#======== Slice decision options ================-ip,  --IntraPeriod [0]            Intra period in frames (0: use intra period in seconds (refreshsec), else:n*gopsize)-rs,  --RefreshSec [1]             Intra period/refresh in seconds-dr,  --DecodingRefreshType [cra]  intra refresh type (idr, cra, cra_cre: CRA, constrained RASL picture encoding, none, rpsei: Recovery Point SEI,idr_no_radl: IDR, without leading pictures, use for DASH)-g,   --GOPSize [32]               GOP size of temporal structure (16,32)--PicReordering [1]          Allow reordering of pictures (0:off, 1:on), should be disabled for low delayrequirements--POC0IDR [0]                start encoding with POC 0 IDR#======== Rate control, Perceptual Quantization ================--NumPasses [-1]             number of rate control passes (1,2)--Passes [-1]                number of rate control passes (1,2)--Pass [-1]                  rate control pass for two-pass rate control (-1,1,2)--LookAhead [-1]             Enable pre-analysis pass with picture look-ahead (-1,0,1)--RCStatsFile []             rate control statistics file--TargetBitrate [0]          Rate control: target bitrate [bits/second], use e.g. 1.5M, 1.5Mbps, 1500k,1500kbps, 1500000bps, 1500000--MaxBitrate [0]             Rate control: approximate maximum instantaneous bitrate [bits/second] (0: no ratecap; least constraint)-qpa, --PerceptQPA [0]             Enable perceptually motivated QP adaptation, XPSNR based (0:off, 1:on)--STA [-1]                   Enable slice type adaptation at GOPSize>8 (-1: auto, 0: off, 1: adapt slice type,2: adapt NAL unit type)--MinIntraDistance [-1]      With STA: set a minimum coded frame distance to the previous intra frame (-1:GOPSize)#======== Quantization parameters ================-q,   --QP [32]                    Qp value (0-63)#======== Profile, Level, Tier ================--Profile [auto]             profile (auto, main_10, main_10_still_picture)--Level [auto]               level limit (auto, 1.0, 2.0,2.1, 3.0,3.1, 4.0,4.1, 5.0,5.1,5.2, 6.0,6.1,6.2,6.3,15.5)--Tier [main]                tier for interpretation of level (main, high)#======== VUI and SEI options ================--Sdr [off]                  set SDR mode + BT.709, BT.2020, BT.470 color space. use: off, sdr|sdr_709,sdr_2020, sdr_470bg--Hdr [off]                  set HDR mode + BT.709 or BT.2020 color space (+SEI messages for hlg) If maxcll ormasteringdisplay is set, HDR10/PQ is enabled. use: off, pq|hdr10,pq_2020|hdr10_2020, hlg, hlg_2020
  • 编码命令：./vvencFFapp -i Cactus_1920x1080_50.yuv --Size 1920x1080 -fr 15 --BitstreamFile out.h266 
admin@bogon release-static % ./vvencFFapp -i Cactus_1920x1080_50.yuv --Size 1920x1080 -fr 15  --BitstreamFile out.h266 
vvencFFapp: VVenC, the Fraunhofer H.266/VVC Encoder, version 1.12.0 [Mac OS X][clang 15.0.0][64 bit][SIMD=NEON]
vvencFFapp [info]: started @ Mon Dec 30 15:52:52 2024
vvenc [verbose]: Input File                             : Cactus_1920x1080_50.yuv
vvenc [verbose]: Bitstream File                         : out.h266
vvenc [verbose]: Real Format                            : 1920x1080  yuv420p  15 Hz  SDR  500 frames
vvenc [verbose]: Frames                                 : encode 500 frames 
vvenc [verbose]: Internal format                        : 1920x1080  15 Hz  SDR
vvenc [verbose]: Threads                                : 0  (parallel frames: 0)
vvenc [verbose]: Rate control                           : QP 32
vvenc [verbose]: Perceptual optimization                : Disabled
vvenc [verbose]: Intra period (keyframe)                : 16
vvenc [verbose]: Decoding refresh type                  : CRA
vvenc [verbose]: Sequence PSNR output                   : Linear average only
vvenc [verbose]: Hexadecimal PSNR output                : Disabled
vvenc [verbose]: Sequence MSE output                    : Disabled
vvenc [verbose]: Frame MSE output                       : Disabled
vvenc [verbose]: Cabac-zero-word-padding                : Enabled
vvenc [verbose]: Frame index                            : all frames
vvenc [verbose]: Profile                                : main_10
vvenc [verbose]: Level                                  : 3.1
vvenc [verbose]: CU size                                : 128
vvenc [verbose]: Max TB size                            : 64
vvenc [verbose]: Min CB size                            : 4
vvenc [verbose]: Motion search range                    : 384
vvenc [verbose]: QP                                     : 32
vvenc [verbose]: Max dQP signaling subdiv               : 0
vvenc [verbose]: Cb QP Offset (dual tree)               : 0 (0)
vvenc [verbose]: Cr QP Offset (dual tree)               : 0 (0)
vvenc [verbose]: GOP size                               : 32
vvenc [verbose]: PicReordering                          : 1
vvenc [verbose]: Input bit depth                        : (Y:8, C:8)
vvenc [verbose]: MSB-extended bit depth                 : (Y:8, C:8)
vvenc [verbose]: Internal bit depth                     : (Y:10, C:10)
vvenc [verbose]: cu_chroma_qp_offset_subdiv             : -1
vvenc [verbose]: log2_sao_offset_scale_luma             : 0
vvenc [verbose]: log2_sao_offset_scale_chroma           : 0
vvenc [verbose]: Cost function                          : Lossy coding
vvenc [verbose]: Film grain analysis                    : 0vvenc [verbose]: CODING TOOL CFG: CTU128 QTMin44BTT21 IBD:1 SAO:1 ALF:1 (NonLinLuma:0 NonLinChr:0) CCALF:1 Tiles:-1x-1 Slices:1 WPP:0 WPP-Substreams:1 TMVP:1 DQ:1 SDH:0 CST:1 BDOF:1 DMVR:1 MTSImplicit:1 SBT:0 JCbCr:1 CabacInitPresent:1 AMVR:5 SMVD:3 LMCS:2 (Signal:SDR Opt:0) CIIP:0 MIP:1 AFFINE:4 (PROF:1, Type:1) MMVD:3 DisFracMMVD:1 SbTMVP:1 GPM:3 LFNST:1 MTS:0 ISP:3 TS:2 TSLog2MaxSize:4 useChromaTS:0 BDPCM:2 IBC:2 BCW:0 
vvenc [verbose]: ENC. ALG. CFG: QPA:0 HAD:1 RDQ:1 RDQTS:1 ASR:1 MinSearchWindow:96 EDO:2 MCTF:2 BIM:1 
vvenc [verbose]: PRE-ANALYSIS CFG: STA:1 LeadFrames:0 TrailFrames:0 
vvenc [verbose]: FAST TOOL CFG: ECU:0 FEN:3 FDM:1 FastSearch:4 (SCC:2) LCTUFast:1 FastMrg:2 PBIntraFast:1 AMaxBT:1 FastQtBtEnc:1 ContentBasedFastQtbt:1 FastMIP:3 FastIntraTools:1 IntraEstDecBit:2 FastLocalDualTree:1 IntegerET:0 FastSubPel:1 ReduceFilterME:2 QtbttExtraFast:3 FastTTSplit:5 IBCFastMethod:3 FIM:0 ALFSpeed:0 QuantThr:4 SelectiveRDQO:0 
vvenc [verbose]: RATE CONTROL CFG: RateControl:0 QP:32 LookAhead:0 FirstPassMode:0 
vvenc [verbose]: PARALLEL PROCESSING CFG: NumThreads:0 MaxParallelFrames:0 IFP:0 (IFPLines:0) WppBitEqual:0 WF:0 POC    15 TId: 0 (IDR_W_RADL, I-SLICE, QP 29, TF 2)     680504 bits [Y 36.7620 dB    U 39.1978 dB    V 42.0054 dB] [ET    14 ] [L0 ] [L1 ]
POC     7 TId: 2 (      RADL, B-SLICE, QP 34, TF 0)     115280 bits [Y 35.5292 dB    U 38.5115 dB    V 40.7113 dB] [ET     4 ] [L0 15 ] [L1 15 ]
POC     3 TId: 3 (      RADL, B-SLICE, QP 37)            41024 bits [Y 35.0316 dB    U 38.4638 dB    V 40.4562 dB] [ET     3 ] [L0 7 15 ] [L1 7 15 ]
POC     1 TId: 4 (      RADL, B-SLICE, QP 40)            28888 bits [Y 34.4404 dB    U 38.4675 dB    V 40.3796 dB] [ET     1 ] [L0 3 7 ] [L1 3 7 ]
POC     0 TId: 5 (      RADL, B-SLICE, QP 41)            14752 bits [Y 34.0539 dB    U 38.3535 dB    V 40.1910 dB] [ET     0 ] [L0 1 ] [L1 1 ]
POC     2 TId: 5 (      RADL, B-SLICE, QP 41)             4488 bits [Y 34.6672 dB    U 38.4853 dB    V 40.4677 dB] [ET     0 ] [L0 1 ] [L1 3 ]
POC     5 TId: 4 (      RADL, B-SLICE, QP 40)             9184 bits [Y 35.0381 dB    U 38.4972 dB    V 40.5848 dB] [ET     1 ] [L0 3 7 ] [L1 7 15 ]
POC     4 TId: 5 (      RADL, B-SLICE, QP 41)             3744 bits [Y 34.8863 dB    U 38.4668 dB    V 40.5362 dB] [ET     0 ] [L0 3 ] [L1 5 ]
POC     6 TId: 5 (      RADL, B-SLICE, QP 41)             2480 bits [Y 35.1109 dB    U 38.4947 dB    V 40.6200 dB] [ET     0 ] [L0 5 ] [L1 7 ]
POC    11 TId: 3 (      RADL, B-SLICE, QP 38)            32888 bits [Y 35.3560 dB    U 38.7223 dB    V 41.0028 dB] [ET     2 ] [L0 7 15 ] [L1 15 7 ]
POC     9 TId: 4 (      RADL, B-SLICE, QP 40)            11888 bits [Y 35.1461 dB    U 38.6167 dB    V 40.8196 dB] [ET     1 ] [L0 7 11 ] [L1 11 15 ]
POC     8 TId: 5 (      RADL, B-SLICE, QP 41)             3008 bits [Y 35.2190 dB    U 38.5908 dB    V 40.7640 dB] [ET     0 ] [L0 7 ] [L1 9 ]
POC    10 TId: 5 (      RADL, B-SLICE, QP 41)             4112 bits [Y 35.0811 dB    U 38.6898 dB    V 40.8922 dB] [ET     0 ] [L0 9 ] [L1 11 ]
POC    13 TId: 4 (      RADL, B-SLICE, QP 40)            14984 bits [Y 35.4017 dB    U 38.8514 dB    V 41.2206 dB] [ET     1 ] [L0 11 7 ] [L1 15 11 ]
POC    12 TId: 5 (      RADL, B-SLICE, QP 41)             4768 bits [Y 35.2061 dB    U 38.7486 dB    V 41.0721 dB] [ET     0 ] [L0 11 ] [L1 13 ]
vvenc [info]: stats:   3.0% frame=  15/500 fps=   0.4 avg_fps=   0.4 bitrate=   971.99 kbps avg_bitrate=   971.99 kbps elapsed= 00h:00m:40s left= 00h:21m:31s 
.
.
.
vvenc [info]: stats summary: frame= 500/500 avg_fps= 0.4 avg_bitrate= 848.51 kbps
vvenc [info]: stats summary: frame I:  31, kbps: 9666.70, AvgQP: 28.52
vvenc [info]: stats summary: frame P:   0, kbps:     nan, AvgQP: nan
vvenc [info]: stats summary: frame B: 469, kbps:  265.64, AvgQP: 39.93vvenc [info]:	Total Frames |   Bitrate     Y-PSNR    U-PSNR    V-PSNR    YUV-PSNR   
vvenc [info]:	      500    a     848.5082   35.6287   39.0437   41.5993   36.6601vvencFFapp [info]: finished @ Mon Dec 30 16:12:41 2024
vvencFFapp [info]: Total Time:     1174.888 sec. [user]     1188.995 sec. [elapsed]
  • 加速编码命令（增加了线程、preset、MTPfrofile）：./vvencFFapp -i Cactus_1920x1080_50.yuv --Size 1920x1080 -fr 15 --Threads 4 --preset faster --MTProfile -1 --BitstreamFile out.h266
  • 编码数据汇总：
    在这里插入图片描述
  1. vvencapp 使用：基础编码器
admin@bogon release-static % ./vvencFFapp
vvencFFapp: VVenC, the Fraunhofer H.266/VVC Encoder, version 1.12.0 [Mac OS X][clang 15.0.0][64 bit][SIMD=NEON]#======== General Options ================-h,   --help [0]                   show default help--fullhelp [0]               show full help-v,   --Verbosity [verbose]        verbosity level (0: silent, 1: error, 2: warning, 3: info, 4: notice, 5: verbose,6: debug)--stats [1]                  enable or disable printing of statistics (fps, bitrate, estimation of encodingtime)--version [0]                show version #======== Input Options ================-i,   --InputFile []               Original YUV input file name or '-' for reading from stdin-s,   --Size [0x0]                 Input resolution (WidthxHeight)--InputBitDepth [8]          Bit-depth of input file-f,   --FramesToBeEncoded [0]      Number of frames to be encoded (default=all)-fr,  --FrameRate [0]              Temporal rate (framerate numerator) e.g. 25,30, 30000, 50,60, 60000--FrameScale [1]             Temporal scale (framerate denominator) e.g. 1, 1001--fps [0/1]                  Framerate as int or fraction (num/denom) --TicksPerSecond [27000000]  Ticks Per Second for dts generation, (1..27000000, -1: ticks per frame=1)--LeadFrames [0]             Number of leading frames to be read before starting the encoding, use whensplitting the video into overlapping segments--TrailFrames [0]            Number of trailing frames to be read after frames to be encoded, use when splittingthe video into overlapping segments-fs,  --FrameSkip [0]              number of frames to skip at start of input YUV [off]--segment [off]              when encoding multiple separate segments, specify segment position to enable segment concatenation (first, mid, last) [off]first: first segment           mid  : all segments between first and last segmentlast : last segment#======== Output options ================-b,   --BitstreamFile []           Bitstream output file name-o,   --ReconFile []               Reconstructed YUV output file name--OutputBitDepth [0]         Bit-depth of output file#======== Threading, performance ================-t,   --Threads [0]                number of threads (multithreading; -1: resolution < 720p: 4, < 5K 2880p: 8, >= 5K2880p: 12 threads)--MTProfile [off]            set automatic multi-threading setting (-1: auto, 0: off, 1,2,3: on, enables tiles,IFP and WPP automatically depending on the number of threads)--preset [medium]            select preset for specific encoding setting (faster, fast, medium, slow, slower,medium_lowDecEnergy)--Tiles [-1x-1]              Set number of tile columns and rows#======== Slice decision options ================-ip,  --IntraPeriod [0]            Intra period in frames (0: use intra period in seconds (refreshsec), else:n*gopsize)-rs,  --RefreshSec [1]             Intra period/refresh in seconds-dr,  --DecodingRefreshType [cra]  intra refresh type (idr, cra, cra_cre: CRA, constrained RASL picture encoding, none, rpsei: Recovery Point SEI,idr_no_radl: IDR, without leading pictures, use for DASH)-g,   --GOPSize [32]               GOP size of temporal structure (16,32)--PicReordering [1]          Allow reordering of pictures (0:off, 1:on), should be disabled for low delayrequirements--POC0IDR [0]                start encoding with POC 0 IDR#======== Rate control, Perceptual Quantization ================--NumPasses [-1]             number of rate control passes (1,2)--Passes [-1]                number of rate control passes (1,2)--Pass [-1]                  rate control pass for two-pass rate control (-1,1,2)--LookAhead [-1]             Enable pre-analysis pass with picture look-ahead (-1,0,1)--RCStatsFile []             rate control statistics file--TargetBitrate [0]          Rate control: target bitrate [bits/second], use e.g. 1.5M, 1.5Mbps, 1500k,1500kbps, 1500000bps, 1500000--MaxBitrate [0]             Rate control: approximate maximum instantaneous bitrate [bits/second] (0: no ratecap; least constraint)-qpa, --PerceptQPA [0]             Enable perceptually motivated QP adaptation, XPSNR based (0:off, 1:on)--STA [-1]                   Enable slice type adaptation at GOPSize>8 (-1: auto, 0: off, 1: adapt slice type,2: adapt NAL unit type)--MinIntraDistance [-1]      With STA: set a minimum coded frame distance to the previous intra frame (-1:GOPSize)#======== Quantization parameters ================-q,   --QP [32]                    Qp value (0-63)#======== Profile, Level, Tier ================--Profile [auto]             profile (auto, main_10, main_10_still_picture)--Level [auto]               level limit (auto, 1.0, 2.0,2.1, 3.0,3.1, 4.0,4.1, 5.0,5.1,5.2, 6.0,6.1,6.2,6.3,15.5)--Tier [main]                tier for interpretation of level (main, high)#======== VUI and SEI options ================--Sdr [off]                  set SDR mode + BT.709, BT.2020, BT.470 color space. use: off, sdr|sdr_709,sdr_2020, sdr_470bg--Hdr [off]                  set HDR mode + BT.709 or BT.2020 color space (+SEI messages for hlg) If maxcll ormasteringdisplay is set, HDR10/PQ is enabled. use: off, pq|hdr10,pq_2020|hdr10_2020, hlg, hlg_2020
admin@bogon release-static % ./vvencapp  
vvencapp: VVenC, the Fraunhofer H.266/VVC Encoder, version 1.12.0 [Mac OS X][clang 15.0.0][64 bit][SIMD=NEON]#======== General Options ================-h,   --help [0]                 show default help--fullhelp [0]             show full help-v,   --verbosity [info]         verbosity level (0: silent, 1: error, 2: warning, 3: info, 4: notice, 5: verbose, 6:debug)--stats [1]                enable or disable printing of statistics (fps, bitrate, estimation of encoding time)--version [0]              show version #======== Input Options ================-i,   --input []                 original YUV input file name or '-' for reading from stdin-s,   --size [1920x1080]         specify input resolution (WidthxHeight)-c,   --format [yuv420]          set input format (yuv420, yuv420_10, yuv420_10_packed, yuv400 (gray), yuv400_10(gray10)-r,   --framerate [60]           temporal rate (framerate numerator) e.g. 25,30, 30000, 50,60, 60000 --framescale [1]           temporal scale (framerate denominator) e.g. 1, 1001 --fps [60/1]               framerate as int or fraction (num/denom) --tickspersec [27000000]   ticks per second for dts generation, (1..27000000, -1: ticks per frame=1)-f,   --frames [0]               max. frames to encode [all]-fs,  --frameskip [0]            number of frames to skip at start of input YUV [off]--segment [off]            when encoding multiple separate segments, specify segment position to enable segment concatenation (first, mid, last) [off]first: first segment           mid  : all segments between first and last segmentlast : last segment#======== Output Options ================-o,   --output []                bitstream output file name#======== Encoder Options ================--preset [medium]          preset for detailed parameter configuration (faster, fast, medium, slow, slower,medium_lowDecEnergy)-b,   --bitrate [0]              bitrate for rate control (0: constant-QP encoding without rate control; otherwisebits/second; use e.g. 1.5M, 1.5Mbps, 1500k, 1500kbps, 1500000bps, 1500000)-m,   --maxrate [0]              approximate maximum instantaneous bitrate for constrained VBR in rate control (0:no rate cap; use e.g. 3.5M, 3.5Mbps, 3500k, 3500kbps, 3500000bps, 3500000), usesuffix 'x' to specify as a multiple of target bitrate-p,   --passes [-1]              number of encoding passes with rate control (1: single-pass, -1, 2: two-pass RC)--pass [-1]                rate control pass for two-pass rate control (-1: both, 1: first, 2: second pass)--rcstatsfile []           rate control statistics file name-q,   --qp [-1]                  quantization parameter, QP (0, 1, .. 63)--qpa [on]                 enable perceptually motivated QP adaptation based on XPSNR model (0: off, 1: on)-t,   --threads [-1]             number of threads (multithreading; -1: resolution < 720p: 4, < 5K 2880p: 8, >= 5K2880p: 12 threads)--mtprofile [auto]         set automatic multi-threading setting (-1: auto, 0: off, 1,2,3: on, enables tiles,IFP and WPP automatically depending on the number of threads)--ifp [auto]               inter-frame parallelization(IFP) (-1: auto, 0: off, 1: on, with sync. offset of twoCTU lines)-rt,  --refreshtype [cra]        intra refresh type (idr, cra, cra_cre: CRA, constrained RASL picture encoding, none, rpsei: Recovery Point SEI,idr_no_radl: IDR, without leading pictures, use for DASH)-rs,  --refreshsec [1]           intra period/refresh in seconds-ip,  --intraperiod [0]          intra period in frames (0: specify intra period in seconds instead, see -refreshsec)--tiles [-1x-1]            number of tile columns and rows#======== Profile, Level, Tier ================--profile [auto]           profile (auto, main_10, main_10_still_picture)--level [auto]             level limit (auto, 1.0, 2.0,2.1, 3.0,3.1, 4.0,4.1, 5.0,5.1,5.2, 6.0,6.1,6.2,6.3,15.5)--tier [main]              tier for interpretation of level (main, high)#======== HDR/SDR and Color Options ================--sdr [off]                set SDR mode + BT.709, BT.2020, BT.470 color space. use: off, sdr|sdr_709, sdr_2020,sdr_470bg--hdr [off]                set HDR mode + BT.709 or BT.2020 color space (+ SEI messages for hlg) use: off,pq|hdr10, pq_2020|hdr10_2020, hlg, hlg_2020
  • 编码命令：./vvencapp -i Cactus_1920x1080_50.yuv --Size 1920x1080 -r 15 --threads 4 --preset faster --output out.h266 
admin@bogon release-static % ./vvencapp -i Cactus_1920x1080_50.yuv --Size 1920x1080 -r 15 --threads 4 --preset faster  --output out.h266  
vvencapp: VVenC, the Fraunhofer H.266/VVC Encoder, version 1.12.0 [Mac OS X][clang 15.0.0][64 bit][SIMD=NEON]
vvenc [info]: Input File                             : Cactus_1920x1080_50.yuv
vvenc [info]: Bitstream File                         : out.h266
vvenc [info]: Real Format                            : 1920x1080  yuv420p  15 Hz  SDR  500 frames
vvenc [info]: Frames                                 : encode 500 frames 
vvenc [info]: Internal format                        : 1920x1080  15 Hz  SDR
vvenc [info]: Threads                                : 4  (parallel frames: 4)
vvenc [info]: Rate control                           : QP 32
vvenc [info]: Perceptual optimization                : Enabled
vvenc [info]: Intra period (keyframe)                : 16
vvenc [info]: Decoding refresh type                  : CRAvvenc [info]: stats: 100.0% frame= 500/500 fps=  14.0 avg_fps=  14.0 bitrate=   921.63 kbps avg_bitrate=   921.63 kbps elapsed= 00h:00m:36s left= 00h:00m:00s    
vvenc [info]: stats summary: frame= 500/500 avg_fps= 14.0 avg_bitrate= 921.63 kbps
vvenc [info]: stats summary: frame I:  31, kbps: 10668.23, AvgQP: 27.45
vvenc [info]: stats summary: frame P:   0, kbps:      nan, AvgQP: nan
vvenc [info]: stats summary: frame B: 469, kbps:   277.40, AvgQP: 39.00vvenc [info]:	Total Frames |   Bitrate     Y-PSNR    U-PSNR    V-PSNR    YUV-PSNR   
vvenc [info]:	      500    a     921.6324   34.7893   38.8460   41.1811   35.8816
vvencapp [info]: Total Time: 35.827 sec. Fps(avg): 13.956 encoded Frames 500
  • 编码数据汇总：
    在这里插入图片描述
  1. vvencinterfacetest：测试 vvenc 接口，无异常则返回空信息
admin@bogon release-static % ./vvencinterfacetest
  1. vvenclibtest：测试 vvenc 库，无异常则返回“all of 174 tests succeeded% ”
admin@bogon release-static % ./vvenclibtest all of 174 tests succeeded%

一些疑问

  1. 用 vvencapp 和vvenFFcapp 工具编码出来都没有 P 帧，分析命令行参数说明也没发现 P 帧的设置，用 elecard 工具分析，码流中 B 帧只有单向参考，并没有双向参考，细节和原理还需要进一步分析源码。
    在这里插入图片描述

相关文章：

Mac 环境 VVenC 编译与编码命令行工具使用教程

VVenC VVenC 是一个开源的高效视频编码器&#xff0c;专门用于支持 H.266/VVC (Versatile Video Coding) 标准的编码。H.266/VVC 是继 HEVC (H.265) 之后的新一代视频编码标准&#xff0c;主要目的是提供比 HEVC 更高的压缩效率&#xff0c;同时保持或提高视频质量。H.266/VVC…...

编程日记 2025/1/1 11:09:55

如何在 Ubuntu 22.04 上部署 Nginx 并优化以应对高流量网站教程

简介 本教程将教你如何优化 Nginx&#xff0c;使其能够高效地处理高流量网站。 Nginx 是一个强大且高性能的 Web 服务器&#xff0c;以其高效处理大量并发连接的能力而闻名&#xff0c;这使得它成为高流量网站的流行选择。 正确优化 Nginx 可以显著提高服务器的性能&#xff0…...

编程日记 2025/1/1 11:06:52

springcloud各个组件介绍

Spring Cloud 是一系列框架的集合&#xff0c;它基于 Spring Boot 提供了在分布式系统&#xff08;如配置管理、服务发现、断路器、智能路由、微代理、控制总线、一次性令牌、全局锁、领导选举、分布式会话和集群状态&#xff09;中快速构建一些常见模式的工具。下面是对 Sprin…...

编程日记 2025/1/1 11:05:50

HTML5实现好看的喜庆圣诞节网站源码

HTML5实现好看的喜庆圣诞节网站源码 前言一、设计来源1.1 主界面1.2 圣诞介绍界面1.3 圣诞象征界面1.4 圣诞活动界面1.5 圣诞热度界面1.6 圣诞纪念界面1.7 联系我们界面 二、效果和源码2.1 动态效果2.2 源代码 源码下载结束语 HTML5实现好看的喜庆圣诞节网站源码&#xff0c;圣…...

编程日记 2025/1/1 11:03:47

《学习之道》

《学习之道》主要讲述了以下内容&#xff1a; 学习的原理 大脑的两种认知模式&#xff1a;介绍了专注模式和发散模式。专注模式适合集中精力解决具体问题、进行深度理解和记忆推理&#xff0c;但长时间使用易疲惫和陷入思维定式&#xff1b;发散模式则让大脑在更广泛的认知网…...

编程日记 2025/1/1 11:02:45

【Unity3D】ECS入门学习（十一）ComponentSystem、JobComponentSystem

ComponentSystem&#xff1a;仅支持主线程执行&#xff0c;不支持多线程&#xff0c;并且无法使用SystemBase介绍的扩展方法。 using Unity.Entities; using Unity.Transforms; using Unity.Mathematics;/// <summary> /// 仅主线程执行&#xff0c;不支持多线程 /// &l…...

编程日记 2025/1/1 11:00:43

力扣刷题：栈和队列OJ篇（上）

大家好&#xff0c;这里是小编的博客频道 小编的博客&#xff1a;就爱学编程 很高兴在CSDN这个大家庭与大家相识&#xff0c;希望能在这里与大家共同进步&#xff0c;共同收获更好的自己&#xff01;&#xff01;&#xff01; 目录 1.用队列实现栈&#xff08;1&#xff09;题目…...

编程日记 2025/1/1 10:55:37

XGPT用户帮助手册

文章目录 2024 更新日志2024.12.272024.12.29 摘要 本文详细介绍了XGPT软件的功能及发展历程。XGPT是一款融合了当前最先进人工智能技术的多模态智能软件&#xff0c;专为国内用户优化设计。除了强大的智能问答功能外&#xff0c;XGPT还结合日常办公和科学研究的需求&#xff0…...

编程日记 2025/1/1 10:54:37

Oracle 数据库 dmp文件从高版本导入低版本的问题处理

当前有个需求是将oracle 19c上的数据备份恢复到oracle 11g上使用。我们通过exp命令远程进行备份&#xff0c;然后通过imp进行恢复时出现IMP-00010: not a valid export file, header failed verification报错。 这是数据库版本问题&#xff0c;在使用exp命令导出的时候使用的客…...

编程日记 2025/1/1 10:51:33

ShardingSphere-Proxy分表场景测试案例

快速入门文章参考&#xff1a;《ShardingSphereProxy:快速入门》 基于K8S部署文章参考&#xff1a;《基于K8s部署ShardingSphere-Proxy》 基于golang的测试用例参考&#xff1a;《ShardingSphere-Proxy 连接实战&#xff1a;从 Golang 原生 SQL 到 GORM 的应用》 背景 我们…...

编程日记 2025/1/1 10:49:30

学技术学英文：Tomcat的线程模型调优

导读&#xff1a; tomcat 线程调优关键需要理解下面这几个参数&#xff1a; 1. maxConnections 描述&#xff1a;指定服务器能够同时接受和处理的最大连接数。也就是说&#xff0c;服务器在任何时候都能处理的最大并发连接数。作用&#xff1a;限制服务器在任何给定时间点能…...

编程日记 2025/1/1 10:47:29

创建flutter项目遇到无法连接源的问题

Flutter 环境信息 Flutter版本: 3.19.4 (channel stable) Framework: revision 68bfaea224 (2024-03-20) Engine: revision a5c24f538d Dart: 3.3.2 DevTools: 2.31.1 项目基本信息 项目路径: D:\F\luichun 域名: www.luichun.com.cn 支持平台: android, web, windows 项目创…...

编程日记 2025/1/1 10:42:23

MAC系统QT图标踩坑记录

MAC系统QT图标踩坑记录 1. 准备图标1.1 方法一&#xff1a;下载准备好的图标1.2 方法二&#xff1a;自己生成图标1.2.1 准备一个png文件1.2.2 用sips生成不同大小的图片1.2.3 用iconutil生成图标文件 2. 配置图标2.1. 把图标改命成自己想要的名字&#xff0c;如icon.icns&#…...

编程日记 2025/1/1 10:38:19

TF-IDF（Term Frequency-Inverse Document Frequency）详解：原理和python实现（中英双语）

中文版 TF-IDF算法详解&#xff1a;理解与应用 TF-IDF&#xff08;Term Frequency-Inverse Document Frequency&#xff09;是信息检索与文本挖掘中常用的算法&#xff0c;广泛应用于搜索引擎、推荐系统以及各种文本分析领域。TF-IDF的核心思想是通过计算一个词在文档中的重要…...

编程日记 2025/1/1 10:37:17

【竞技宝】CS2：HLTV2024职业选手排名TOP15-xantares

北京时间2024年12月30日&#xff0c;HLTV年度选手排名正在持续公布中&#xff0c;今日凌晨正式公布了今年的TOP15选手为EternalFire战队的xantares选手。 选手简介 xantares是一名来自于土耳其的CS职业选手&#xff0c;出生于1995年&#xff0c;今年已经29岁。早在2012年&…...

编程日记 2025/1/1 10:35:15

Spring-kafka快速Demo示例

使用Spring-Kafka快速发送/接受Kafka消息示例代码&#xff0c;项目结构是最基础的SpringBoot结构&#xff0c;提前安装好Kafka&#xff0c;确保Kafka已经正确启动 pom.xml&#xff0c;根据个人情况更换springboot、java版本等 <?xml version"1.0" encoding&qu…...

编程日记 2025/1/1 10:32:11

客户案例：基于慧集通集成平台，打通屠宰管理系统与用友U8C 系统的全攻略

一、引言 本原型客户成立于2014年&#xff0c;是一家集饲草种植、肉牛养殖、精深加工、冷链物流、餐饮服务于一体的大型农牧综合体。公司下设三个子公司分别涵盖农业、畜牧业、肉制品加工业与餐饮物流服务业。公司严格按照一二三产业融合发展要求&#xff0c;以肉牛产业化为支…...

编程日记 2025/1/1 10:31:10

模型 九屏幕分析法

系列文章 分享 模型&#xff0c;了解更多&#x1f449; 模型_思维模型目录。九屏幕法&#xff1a;全方位分析问题的系统工具。 1 九屏幕分析法的应用 1.1 新产品研发的市场分析 一家科技公司计划开发一款新型智能手机&#xff0c;为了全面评估市场潜力和风险&#xff0c;他们…...

编程日记 2025/1/1 10:29:09

Qanything 2.0源码解析系列6 PDF解析逻辑

Qanything 2.0源码解析系列6: PDF解析逻辑 type: Post status: Published date: 2024/12/04 summary: 深入剖析Qanything是如何拆解PDF的，核心是pdf转markdown category: 技术分享 原文：www.feifeixu.top 😀 前言： 在前面的文章中探究了图片是怎么进行解析的，这篇文章对…...

编程日记 2025/1/1 10:28:08

MAC系统QT Creator的快捷键

安装好QT Creator后使用了一段时间&#xff0c;真是越用越难受&#xff0c;只想说&#x1f5d1;️。。。 找一圈qt creator的快捷键 0. 快捷键界面 这里的搜索真的是…无语&#xff0c;不考虑是人查找吗&#xff1f;&#xff1f; 1. 代码前后浏览 2. 移动代码 3. 半自动导入…...

编程日记 2025/1/1 10:18:59

SpringBoot-17-MyBatis动态SQL标签之常用标签

文章目录 1 代码1.1 实体User.java1.2 接口UserMapper.java1.3 映射UserMapper.xml1.3.1 标签if1.3.2 标签if和where1.3.3 标签choose和when和otherwise1.4 UserController.java2 常用动态SQL标签2.1 标签set2.1.1 UserMapper.java2.1.2 UserMapper.xml2.1.3 UserController.ja…...

编程新知 2025/7/15 0:35:43

Prompt Tuning、P-Tuning、Prefix Tuning的区别

一、Prompt Tuning、P-Tuning、Prefix Tuning的区别 1. Prompt Tuning（提示调优） 核心思想：固定预训练模型参数，仅学习额外的连续提示向量（通常是嵌入层的一部分）。实现方式：在输入文本前添加可训练的连续向量（软提示），模型只更新这些提示参数。优势：参数量少（仅提…...

编程新知 2025/7/14 8:29:30

从WWDC看苹果产品发展的规律

WWDC 是苹果公司一年一度面向全球开发者的盛会&#xff0c;其主题演讲展现了苹果在产品设计、技术路线、用户体验和生态系统构建上的核心理念与演进脉络。我们借助 ChatGPT Deep Research 工具&#xff0c;对过去十年 WWDC 主题演讲内容进行了系统化分析&#xff0c;形成了这份…...

编程新知 2025/7/15 3:38:17

Cesium1.95中高性能加载1500个点

一、基本方式&#xff1a; 图标使用.png比.svg性能要好 <template><div id"cesiumContainer"></div><div class"toolbar"><button id"resetButton">重新生成点</button><span id"countDisplay&qu…...

编程新知 2025/7/8 4:49:22

聊聊 Pulsar：Producer 源码解析

一、前言 Apache Pulsar 是一个企业级的开源分布式消息传递平台&#xff0c;以其高性能、可扩展性和存储计算分离架构在消息队列和流处理领域独树一帜。在 Pulsar 的核心架构中&#xff0c;Producer&#xff08;生产者&#xff09; 是连接客户端应用与消息队列的第一步。生产者…...

编程新知 2025/7/7 17:18:31

Linux云原生安全：零信任架构与机密计算

Linux云原生安全&#xff1a;零信任架构与机密计算 构建坚不可摧的云原生防御体系 引言&#xff1a;云原生安全的范式革命 随着云原生技术的普及&#xff0c;安全边界正在从传统的网络边界向工作负载内部转移。Gartner预测&#xff0c;到2025年&#xff0c;零信任架构将成为超…...

编程新知 2025/7/5 14:33:07

自然语言处理——循环神经网络

自然语言处理——循环神经网络 循环神经网络应用到基于机器学习的自然语言处理任务序列到类别同步的序列到序列模式异步的序列到序列模式 参数学习和长程依赖问题基于门控的循环神经网络门控循环单元&#xff08;GRU&#xff09;长短期记忆神经网络&#xff08;LSTM&#xff09…...

编程新知 2025/7/13 9:06:44

什么是Ansible Jinja2

理解 Ansible Jinja2 模板 Ansible 是一款功能强大的开源自动化工具&#xff0c;可让您无缝地管理和配置系统。Ansible 的一大亮点是它使用 Jinja2 模板&#xff0c;允许您根据变量数据动态生成文件、配置设置和脚本。本文将向您介绍 Ansible 中的 Jinja2 模板&#xff0c;并通…...

编程新知 2025/7/14 22:01:47

算法岗面试经验分享-大模型篇

文章目录 A 基础语言模型A.1 TransformerA.2 Bert B 大语言模型结构B.1 GPTB.2 LLamaB.3 ChatGLMB.4 Qwen C 大语言模型微调C.1 Fine-tuningC.2 Adapter-tuningC.3 Prefix-tuningC.4 P-tuningC.5 LoRA A 基础语言模型 A.1 Transformer &#xff08;1&#xff09;资源 论文&a…...

编程新知 2025/7/13 22:40:31

保姆级教程：在无网络无显卡的Windows电脑的vscode本地部署deepseek

文章目录 1 前言2 部署流程2.1 准备工作2.2 Ollama2.2.1 使用有网络的电脑下载Ollama2.2.2 安装Ollama&#xff08;有网络的电脑&#xff09;2.2.3 安装Ollama&#xff08;无网络的电脑&#xff09;2.2.4 安装验证2.2.5 修改大模型安装位置2.2.6 下载Deepseek模型 2.3 将deepse…...

编程新知 2025/7/13 23:07:00