NAUL（国外英文资料）.doc

NAUL（国外英文资料） 1. The introduction The main goal of h.264: High video compression ratio Good network affinity The solution: VCL video coding layer video coding layer NAL network abstraction layer network extraction layer VCL: the definition of the grammar level of the core algorithm engine, block, macro block, and slice NAL: magnitude of grammatical level (such as sequence parameters and image parameter sets), at the same time support the following functions: independent decoding, the initial code only assurance, SEI and coding data transfer flow format VCL design goal: to be able to decode efficiently as independently as possible NAL design goal: package the data in a different format according to different networks, and match the bitstrings generated by VCL to a wide variety of networks and multiple environments. NALU head structure: NALU type (5bit), importance indicator (2bit), and no bit (1bit). Type of NALU: 1 ~ 12 is used by h.264. 24 ~ 31 is used by applications other than h.264. Importance indicator: indicates the importance of the NAL unit for rebuilding. The larger the value, the more important it is. Forbidden bits: the network finds that the NAL unit has a bit error that can be set to 1 so that the receiver loses the unit. 2. NAL grammar semantics NAL layer syntax: The basic unit of data is the syntactic element in the coder output. The syntax represents the structure of the syntactic elements. Semantic elaboration of the concrete meaning of the syntactic elements. The group has a head, and the decoder can easily detect the boundaries of the NAL, and take the NAL to decode. But in order to save the code stream, h.264 doesnt have a syntactic element that represents the starting position of the head of NAL. If the encoded data is stored in the medium, because of the NAL is, in turn, are closely linked, decoder cant distinguish each NAL in data flow starting position and end position. Solution: add a start code before each NAL: 0X000001 In certain types of m