MPEG (MOVING PICTURE EXPERTS Group) After successfully defined MPEG-1 and MPEG-2, the new MPEG-4 standard began in July 1993 and officially announced version 1 in early 1999 and early 2000, respectively. Version 2. By October 2001, MPEG-4 has defined 19 Visual Profiles, where newly defined simple studio classes and core studio classes make MPEG-4 pairs. The MPEG-2 category retains some compatible, and its yield rate can be up to 2Gbps. With the continuous expansion of the MPEG-4 standard, it can support multimedia communications with less than 64kbps, but also support broadcast-level video applications. MPEG-4 standards will be widely used in digital TV, dynamic images, web (WWW), real-time multimedia monitoring, content storage and retrieval multimedia systems, video streams on the Internet, video streams, and virtual conferences based on facial expressions. Interactive multimedia applications on DVD, based on computer network visual cooperative laboratory scene applications, student TVs, etc., which will push the ultimate integration of three major networks of telecommunications, computers, and broadcast TV, which has become the mainstream of compression standards in the future.
Main features and functions of MPEG-4 standards
MPEG-4 video coding standards support most of MPEG-1, MPEC-2, providing different video standard source formats, code rates, valid coding of rectangular images under frame frequencies, and also supports content-based image encoding. The following figure shows a schematic diagram of the relationship between the code rate and the corresponding function set of MPEG-4.
The bottom layer of this feature is the core of VLBV (VERY LOW BIT RATE VIDEO). It provides algorithms and tools to provide algorithms and tools in the 5-six4kbps range, support lower spatial resolution (less than 352 x 288 pixels) and lower frame rate (less than 15 Hz). The special features of the VLBV core support include: a valid code of the rectangular image sequence, a search and random access of a multimedia database.
MPEG-4 HBV (High Bit Rate Video) also supports the above functions, and its code rate ranges between 64 kbps-10MBbps, which uses the same or similar algorithm with the VLBL core, but it supports higher space and time resolution. Allow transmission and storage of high quality video signals suitable for the studio, which can be standard signals of ITU-R REC.601, typical applications for digital television broadcasting and interactive retrieval. MPEG-4 finally supported code rates will be higher than MPEG-2. MPEG-4 proposes content-based access concepts that allow users to interact with the scene. It encodes the contents of the moving image, the specific encoding object is the audio and video in the image, called the AV object (AVO: Audio Video Object). AV objects can form AV scenes (AVOS: Audio Video Object In A Scene). Therefore, the basic content of the MPEG-4 standard is to encode, organize, store, and transmit AV objects in efficiently.
MPFG-4 standards support 8 new or improved features, can be divided into the following 3:
1. Content-based interactivity
Content-based multimedia data access tool; content-based code stream manipulation and editing; natural and synthetic data mixed coding; enhanced time domains random access.
2, high compression ratio
Improve coding efficiency; encoding multiple concurrent data streams.
3. Flexible and diverse access
Errors in the blonde environment; content-based scalability is based on content. MPEG-4 standard
1, DMIF
DMIF is the overall framework of multimedia, which mainly solves the operational issues of multimedia applications in the broadcast environment and disk applications. With DMIF, MPEG-4 can establish channels with special quality service (QoS: Quality of Service) and bandwidth for each basic stream.
2, data plane
The data planes in MPEG-4 can be divided into two parts: transmission relationship section and media relational part. In order to appear in the same scene in the same field, MPEG-4 references the concept of an object description (OD) and flow graph (SMT). OD transmission information flow diagram related to special AV objects. The desktop connects each stream with a CAT (Channel Association Tag), and CAT can achieve the smooth transmission of the stream.
3, buffer management and real-time identification
MPEG-4 defines a system decoding mode (SDM), which describes an ideal decoding device for processing a schematic symmetric semantics, which requires special buffers and real-time modes. By effective management, limited buffer space can be better utilized.
4, audio coding
The superiority of MPEG-4 is that it not only supports natural sounds, but also supports synthetic sounds. The audio portion of the MPEG-4 combines audio synthetic encoding and coding of natural sounds and supports audio object feature.
5, video coding
Similar to audio coding, MPBG-4 also supports encoding of natural and synthetic visual objects. The synthetic visual object includes 2D, 3D animation and human face expression animation.
6, scene description
MPEG-4 provides a range of tools for forming a set of objects in the scene. Some necessary synthetic information constitute a scenario description.
MPEG-4 standard video coding technology
In order to support the content-based interactivity, it is supported to independently, decoded, decoded, and MPEG-4 video inspection models introduced the concept of video objects (VIDEO OBJECT PLANE). Assuming that each frame of the input video sequence is split into a plurality of arbitrary shapes (video object surfaces), each region may overwrite the specific image or video content of the scene in the scene. The VOP for encoding can be any shape, and the shape and position can be changed as frame. Continuous VOP sequences of the same physical objects belonging to the scene are called video objects. The shape, motion, and texture information of the VOP sequence of the same video object is encoded, or encoded as an isolated video object layer (Vol: Video Object Layer).
The basic structure of the MPEG-4 video encoder includes shape encoding (VOS for any shape), motion compensation, and DCT-based texture encoding (using standard 8 x 8 DCT or depending on the shape of adaptive DCT). The specific encoding method is: first perform a scene analysis and object segmentation of the input original image sequence to divide different VOPs to obtain the shape and location information of each VOP, which can be represented by the alpha plane. The sender only needs to transmit the ALPHA plane, and the receiving end can determine the shape and position of the VOP. The number of bits required for the Alpha plane is more, and compression coding is required. Obviously, as long as the contour of the VOP can be encoded and transmitted, the receiving end can restore the Alpha plane, and the contour information is encoded in the contour encoder. The extracted shape and location information is used to control the movement and texture encoding of the VOP. Sports prediction / compensation methods similar to MPEG-1/2 standards are still employed for motion and texture information. Enter the VOP of the n-1 frame stored in the frame memory in the VOP of the N-1 frame stored in the frame memory, and then quantify the difference between the two frames of VOP. The coding of the movement and texture information of different objects can be used in different ways to increase coding efficiency. After encoding the texture information, the motion information and shape information output from the motion encoder and the shape encoder form a bit flow layer forming the VOP. The VOP sequences of different video objects are encoded, and their respective bit flow layers are formed, and they are transmitted over the channel after they are switched. The order in which the transfer is sequentially shaped information, motion information, and texture information. The decoding process of the receiving end is the inverse operation of the encoding process. MPEG-4 standard in broadcasting and television field application prospects
With the constant complementation and improvement of MPEG 1 standard, it has been able to support broadcast-level video applications. The emergence of MPEG-4 will change the traditional TV production concept, and it also gives a larger autonomy of TV audiences. It not only injects new vitality into the radio and television field, but also brings more opportunities and challenges to the majority of TV workers. . MPEG-4 standards will have the following applications in the field of radio and television:
1, virtual studio
MPEG-4 unique content-based interactivity, can bring breakthroughs to virtual studio technology, and the application range of virtual studio will also be expanded.
◇ Virtual attendance
You can achieve virtual attendance by using MPEG-4 to support natural video images with composite data. Several performers in the same place can talk to each other in a virtual scene, and the guests in the field can participate in the program production in the local studio.
◇ Virtual playback
Using MPEG-4 supports the characteristics of the same scene multi-view coding, it can be applied to stereo video, thereby achieving virtual playback. Virtual playback is primarily applied to the broadcast and comment programs of the ball competition, etc., which can simulate real competitions from all angles.
◇ Video tracking
With the MPEG-4 supporting the characteristics of data access and operation based on AV content, video tracking can be effectively implemented. Various arrows, trajectories, routes, and signs can be drawn directly on video images, making the audience more clearly understand every detail in the game.
◇ Virtual advertising
The MPEG-4 provides content-based code-based syndrome and editing, allowing users to select the specified object in the image and change their characteristics, and use this feature to achieve virtual ads. Insert the virtual billboard made by the studio into the air ground of the stadium or the performance site, or replace the original billboard on the site with a virtual advertising, which can achieve the true effect of the fake after synthesis.
2, interactive TV
Using MPEG-4 interactive and supporting multiple concurrent data stream codes, the interactive TV can be implemented. For example, when playing several football games at the same time, the audience can choose to watch their favorite competitions at home. They can even choose to watch 3-4 games at different resolutions, and enjoy additional automatic switching features, so as not to Leave the goal plane in all games. Such applications also utilize the complexity scalable characteristics supported by MPEG-4, ie, allows the end user to decode each flow only with a portion of the receiving end decoding capability. Of course, the prerequisite is that the program itself can decode and display in different resolution. Additionally, the user can also be allowed to do the following: changing the view, listening point of the scene, for example in the scene; drag the object in the scene to a different location; click on a specific object to trigger a series of events, such as start or terminate Video stream; select the language that you want when you want the language. 3, video editing
MPEG-4 defined Simple Studio Profile provides high quality signals for video editing, which only has 1 frame, with shape encoding, supporting multiple Alpha channels, with bit rate up to 2Gbps. Core Studio PROFILE adds P frames based on simple studio classes, making coding more efficient and more complicated. These two newly extended video classes make MPEG-4 for high quality video editing.
◇ Use the MPEG-4 to support the characteristics of content editing, directly select the sound, video content, and flexibly control and display it, the user can select the decoding quality of the object in the scene, and change certain characteristics Conduct TV programs and editing.
◇ Use the MPEG-4 to support the characteristics of the time domain, which can be randomly accessed, and quickly searched in a sequence, video objects in the sequence.
◇ Using MPEG-4 supports the characteristics of natural and synthetic data mixed coding, various stunt function is applied to natural and synthetic AV objects, enhanced program editing capabilities.
◇ Take the high compression ratio of MPEG-4 and in the error in the error environment, you can transfer the newsletter and coarse program, which can easily review, management, production, and display of the program.
Currently, there have been companies to showcase digital cameras encoded in MPBG-4 while shooting. If the corresponding network component can be used, the real-time transmission of the image is performed by the Internet. With the MPEG-4 compilation, the decoding technology is mature, the online TV station generated by the future will have a strong impact on traditional media. The MPEG-4 standard completed the transformation of object and content-based modern coding from pixel-based traditional coding, complying with the development trend of modern image compression coding, will have far-reaching influence on future radio and television.
