Raw Byte Sequence Packet (RBSP)
String Of Data Bits (SODB)
POC:Picture Order Count
Sequence parameter set (SPS)
Picture parameter set (PPS)
DPB 解碼圖像緩沖區decoder picture buffer
IDR圖像 立即刷新圖像 一個序列的第一個圖像,IDR圖像都是I幀圖像
PAFF means "Picture Adaptive Field Frame", MBAFF means "MacroBlock Adaptive Field Frame".
* PAFF allows to decide, on a frame basis, whether to encode each frame as a frame or as two fields. When encoded as two fields, it's as if you would have done a separatefields() on the frame.
* MBAFF allows to decide, on a macroblock basis, whether to encode as field or frame. Actually, it works on vertical pairs of macroblocks, so on 16x32 areas. You can easily put two frame macroblocks ( on under the other ), or two field ones ( one with the top lines, the other with the bottom lines ).
MBAFF is the most efficient method in almost every cases ( if not all ), as soon as the encoder does a proper job, of course.
Finally, MBAFF and PAFF may be used together. In that case, you can code a frame either as MBAFF ( thus having the choice field/frame per macroblock ) or as field.
UVLC 統一可變長編碼
SODP:String of data bits(original data) coding結束后的原始比特流
RBSP:Raw byte sequence packe end with 1 and trailing o for alignment 在SODP的結尾加上1和一連串0,用於對齊用。
EBSP:Encapsulate byte sequence payload in order to
avoid start code conflict in NALU,add 0x03 affter two ox00
CABAC 和 CAVLC 只用來對殘差編碼,其余語法元素都是用哥倫布編碼或自然二進制編碼
fidelity range extension:高精度擴展
EPZS,其解釋是Enhanced Predictive Zonal Search,可以翻譯為"改進的預測式區域搜索算法",
CBP:P210,,CBP(CodedBlockPattern)是指子宏塊殘差的編碼方案
Annex B:將編碼后的h264流采用 起始碼 + nal的方式放在介質上保持,即在附錄B中所描述的方法。
rtp :將264流的nal按照標准打包放入rtp包中
NALU 分為 VCL NALU 和 non-VCL NALU(見 200503 版標准 7.4.1 小節 nal_unit_type 的語義解釋),我所說的"若不采用數據分割機制,一個片就是一個NALU,一個NALU就是一個片,否則一個片是由三個NALU組成的,標准表7-1中 nal_unit_type的值為2,3,4的三個NALU屬於一個片"是針對 VCL NALU 而言。而片的概念也只適用於 VCL NALU,因此 SPS、PPS 不是片,也不屬於片。
字節流格式主要用於文件存儲,因此在該格式碼流中 NALU 前面只有一個開始前綴
RTP格式碼流因為需要進行網絡傳輸,因此 NALU 前面還有很多輔助信息
分ABC片主要目的是為了對重要程度不同的數據進行不同程度的保護
Highprofile(HP)
Mainprofile(MP)
UMHexagonS:非對稱十字型多層次六邊形格點運動搜索算法
率失真曲線:
率失真(RD)曲線 反映了不同編碼器的編碼性能好壞。
一般RD曲線都是以碼率(Kbps)做為橫坐標,以PSNR(dB)作為縱坐標做出來一條曲線,曲線上的點一般是采用QP=28,32,36,40這四個QP下的編碼碼率和編碼質量 (QPi, butrate,psnr)。
曲線點越高,表明性能越好。