H.264/mpeg-4 Part paper-Translation

1. Introduction

Broadcast (broadcast) television and home Entertainment (entertainment) have been revolutionised (radical change) by the Advent (Advent) of digital TV anddvd- Video. (---broadcast and home entertainment have been radically changed after digital TV and Dvd-video appeared--) These applications and many more were made possible by the Standardisati On video compression technology. (--these applications and many other things are slowly becoming possible after this video compression technology standard--) The next standards in the MPEG series (--mpeg Next standard--), MPEG4 is enabling (enabled) a new generation of (nominee) internet-based Video Applications (--MPEG4 provides the possibility for a new generation of video applications for the Internet--) Whilst (also) the ITU-T H.263 standard for video compression are now widely used in video conferencing (conference) systems. (--itu-t H .263 video Compression Standard is now widely used in video conferencing systems--)

MPEG4 (Visua (VIDEO) L) and H.263 is standards that is based on video compression ("Video coding") technology from circa . 1995. (--MPEG4 and H.263 are the standard for video compression (video coding) technology starting around 1995. --The groups responsible for these standards (--the standard that organizes these things), the Motion picture experts Group and the Video Coding E Xperts Group (MPEG and VCEG) (--Motion picture Expert Group, and video coding expert Group--) is in the final stages of developing a new standard that promises to s ignificantly outperform MPEG4 and H.263 (--dedicated to developing a more than MPEG4--,), providing better compression of video images together with A range of features supporting high-quality, Low-bitrate streaming video. (--high performance, high quality, low bit stream--) The history of the new Standar　　D, "Advanced Video Coding" (AVC), goes least 7 years.

　　After finalising (finalized) the original (original) H.263 standard for Video telephony (tel) in 1995 (--in 1995 to finalize H.263 original standards for video telephony), the I TU-T Video Coding Experts Group (VCEG) started  work on two further development areas (ITU-T video coding Expert Group began to develop in both in-depth areas:): A " Short-term "effort to add extra features (features, features) to h.263 (resulting in Version 2  of the standard) (--one is to increase H through short-term efforts.) 263 more features (set out in version 2 of the standard)--) and a "long-term" effort to develop a new standards for the low bit rate visual communications. (--another is through Long-term efforts to develop a new version in order to accommodate low bitstream video compression--the long-term effort led to the draft ' H.26L ' Standard, (--long-term team effort leadership drafted h.26l standards--) Offering significantly (significant) better video compression efficiency (efficiency) than previous ITU-T Standards (---provides a significantly better compression efficiency than the ITU-U standard--). In 2001, the ISO Motion Picture Experts Group (MPEG) recognised the potential benefits (potential benefits) of h.26l (--2001 year MPEG Group found H.2 64 potential bit--) and the Joint Video team (JVT) was formed, (the JVM team is raised.) Including experts from MPEG and VCEG (includes experts from these two groups). JVT ' s main task is to develop tHe draft h.26l "model" into a full-international standard. (The main task of the JVM team is to develop the "h.26l standard to International standards") in fact, the outcome will is two identical standards (in fact they have introduced two standards): ISO MPEG4 part ten of MP EG4 and ITU-T H. For these two) the "official" title of the new standard was Advanced Video Coding (AVC) (the official claim that this is the real version); However, it is widely known to its old working title, and the h.26l and by its ITU document number, H. [1]. (yet more widely known is the use of older naming methods 。。。。。 ）

2. H. CODEC

In Common with earlier standards (such as MPEG1, MPEG2 and MPEG4) (--similar to previous versions), the H. Draft standard does not Explicitly (clear) define a CODEC (Encoder/decoder pair). (--h.264 does not explicitly define a codec machine--) rather, the standard defines the syntax of a encoded video bitstream together with the method O F decoding this bitstream. (--and this standard defines only the methods for encoding video as Bitstream and decoding from Bitstream. --) In practice, however, a compliant encoder and decoder is likely to include the functional elements shown in Figure 2- 1 and Figure 2-2. (--however, a qualified codec machine needs to contain the functional elements of Figure 2-1 and 2-2--) whilst the functions shown in these figures is likely to being necessary for compliance (It is also important for a qualified codec machine to demonstrate these functions.) ), there is scope for considerable variation in the structure of the CODEC (codec machine still has considerable variation in space). The basic functional elements (these essential functional elements) (prediction, transform, quantization, entropy encoding) (prediction, transmission, quantization, entropy coding) are little Different from previous standards (similar to previous standards) (MPEG1, MPEG2, MPEG4, h.261, H.263); The important changes in H. occur in the DETAils of each functional element. The main changes in H. I are the implementation details of these modules. ）

The Encoder (Figure 2-1) includes two dataflow paths (encoding consists of two data stream paths), a "forward" path (left-to-right, shown-in blue) (a Forward, ( Left-to-right) and a ' reconstruction ' path (right-to-left, shown-in magenta). (There is also a refactoring path that is the path of a straight-to-go, magenta route) The dataflow path in the Decoder (Figure 2-2) are shown from right to left to illustrate the similarities between Encoder and Decoder. (The data stream of the decoder is represented by To illustrate the common denominator of the codec machine. ）

2.1 Encoder (forward Path) 2.1 Encoder (forward path)

An input frame F n was presented for encoding. The frame is processed in units of a macroblock (corresponding to 16x16 pixels in the original image). Each macroblock are encoded in intra or inter mode. In either case, a prediction macroblock P was formed based on a reconstructed frame. In Intra mode, P was formed from samples on the current frame n that has previously encoded, decoded and reconstructed (UF ' N in the figures; Note that the unfiltered samples is used to form P). In Inter mode, P was formed by motion-compensated prediction from one or more reference frame (s). In the figures, the reference frame is shown as the previous encoded frame F ' n-1; However, the Predicton for each macroblock is formed from one or both past or future frames (in time order) that has a Lready been encoded and reconstructed.
The prediction P is subtracted from the current macroblock to produce a residual or difference macroblock D N. This was transformed (using a block transform) and quantized to give X, a set of quantized transform coefficients. These coefficients is re-ordered and entropy encoded. The entropy-encoded coefficients, together with side information required to decode the macroblock (such as the Macroblock Prediction mode, Quantizer step size, motion vector information describing how the Macroblock is motion-compensated, etc ) Form the compressed bitstream. This is passed to a Network abstraction Layer (NAL) for transmission or storage.

H.264/mpeg-4 Part paper-Translation