 |
|
|
|
|||||||||||||||||||||||
|
MPEG Profiles MPEG profiles are a particular implementation or set of required protocols and actions that enables the providing of features and services for particular MPEG applications. These applications range from providing standard television services over a broadcast system to providing video services on a mobile wireless network. The use of profiles allows an MPEG device or service to only use or include the necessary capabilities (such as codec types) that are required to deliver media to the applications. Profiles are created for specific applications and types of media. New profiles are constantly being requested and created. Because the types of applications of applications that use MPEG transmission can dramatically vary, MPEG has tens of profiles and there are different profile types for the MPEG-2 and MPEG-4 systems. MPEG-2 Profiles MPEG-2 profiles include simple profiles (low complexity), main (standard TV), scalable profile (variable capabilities), high profile (HDTV), and 4:2:2 profile (studio quality). Simple Profile (SP) The simple MPEG profile provides video for relatively simple bandwidth limited devices such as mobile telephones and personal digital assistants. The simple profile does not allow the use of bidirectional frames (B-Frames) which keeps the coding complexity low. Main Profile (MP) Main profile is a common set of protocols and processes that are used to provide standard television services. The main profile used in the MPEG system allows for the use of Intra frames (I-Frames), predicted frames (P-Frames) and bidirectional frames (B-Frames). The MPEG |
main profile also allows for the incorporation of background sprites, interlacing and object shapes with transparency. The main profile is commonly used for broadcast television applications. Scalable Profile Scalable profile is the set of MPEG protocols and processes that can provide video with varying bandwidth and performance needs. The scalable profile defines a base layer along with layers that are used to enhance the quality or performance of a video signal. The scalable profile enables the use of mobile devices that have varying capabilities of reception and display resolution. High Profile (HP) High profile is the set of MPEG protocols and processes that are used for high definition television (HDTV). Although the high profile was originally developed for HDTV service, the main profile levels were expanded to enable the transmission of HDTV signals using the main profile. 4:2:2 Profile 4:2:2 profile is a set of protocols and process that are used to provide high quality color video for studio production and distribution services. The 4:2:2 contains more color elements than the standard 4:2:0 YUV color format. The 4:2:2 profile allows for the use of I-frames, P-frames |
||||||||||||||||||||||
18 Nov 2006 Definitions FREE at www.IPTVDictionary.com
|
and B-frames. At standard resolution (720x576), its maximum data rate is 50 Mbps and at high definition (1920x1080), the maximum data rate is 300 Mbps.
Figure 1.32 shows the different types of profiles and how they are used in the MPEG-2 systems. This table shows that the key MPEG-2 profile types include simple, main, scalable, high and 4:2:2 studio profiles. Simple profiles are used for low complexity devices (such as portable media players). The main profile is used for standard broadcast television applications. Scalable profiles offer the ability to provide varying levels of service different types of devices (such as wireless devices). The high profile is used for high definition television. The 4:2:2 profile is used for studio production and distribution. MPEG-4 Profiles MPEG-4 profiles include audio profiles, visual profiles, MPEG-J profiles (application programming interfaces, advanced video coding (higher compression), scene distribution (dimensional control) and graphics profiles (image processing). Audio Profiles Audio profiles are protocols and processes that are used to provide audio transfer and rendering applications and services. The audio profiles focus on specific applications that require tradeoffs such as the tradeoff of having a high audio compression ratio which results in increased audio processing (coding time) delays and higher sensitivity to transmission errors. Additional tradeoffs include speech coding (optimized for human voice) that can offer a high compression ratio as opposed to natural audio coding with relatively low compression ratios that can reproduce any sound. The MPEG-4 system has several audio profiles including speech |
profile, synthesis profile, scalable profile, main profile, high quality audio profile, low delay audio profile, natural audio profile and mobile audio internetworking profile.
The speech profile can be used to produce a range of speech audio functions ranging from high bit rate waveform coding to very low bit rate text to speech audio coding. Text to speech conversion transforms text (ASCII) information into synthetic speech output. This technology is used in voice processing applications that require the production of broad, unrelated and unpredictable vocabularies, e.g., products in a catalog, names and addresses, etc. Synthesis profile defines how the MPEG system can use wave table synthesis and text to speech to create audio. The scalable profile can use different bit rates and audio bandwidths to effectively provide audio that contains music and speech. Main profile contains tools for natural and synthetic audio. High quality audio profile allows for the use of advanced audio coding (AAC) which is more efficient (has a higher data compression ratio) than the MP3 coder. The low delay audio profile uses speech coders and text to speech interfaces to provide low delay real time communication. It is a version of the advanced audio coder that reduces the amount of processing time for coding which can be used for real time two-way communication applications. Natural audio profile contains the capabilities for natural (non-synthetic) audio coding. Mobile audio internetworking profile uses a scalable AAC with low delay to enable high quality audio coding processes for time sensitive applications (such as video conferencing). Visual Profiles Visual profiles are protocols and processes that are used to provide video transfer and rendering services. The MPEG-4 system video profiles have capabilities that include the transfer of arbitrarily shaped objects (as opposed to rectangular shaped objects), scalable (variable |
|
|
|
| Figure 1.32., MPEG-2 Profile Types | |
20 Nov 2006 Definitions FREE at www.IPTVDictionary.com
|
quality) images, layered (images that can be enhanced) and the transfer and remote (local creation) of animated images.
Simple visual profile is a low complexity video coding system that can be used for portable video devices such as multimedia cellular telephones. Because it is a simple profile, it does not include the ability to provide interlaced video. Simple scalable visual profile can carry over multiple sequences (temporal scalability) or over multiple areas (spatial scalability) to allow operation at different bit rates. The fine granularity scalability profile has the ability to layer multiple levels of image quality. It starts with a base layer and ads enhancement layers to improve the resolution of the images. It can be used in systems to offer different quality levels when bandwidth is selectable or adjustable. Advanced simple profile (ASP) can use B-frames and can use ¼ pel compensation to increase the compression ratio (it is approximately 1/3rd more efficient than simple visual profile). Core visual profile has the ability to define objects that have different (arbitrary) shapes. It is useful for applications that have interactivity between media objects. The core scalable profile includes improved scalability options for arbitrarily shaped objects. Main visual profile includes the ability to process interlaced video objects. The main visual profile can be used for broadcast and interactive DVD applications. N-bit visual profile defines how to code video objects using pixel depths that range from 4 bits to 12 bits. Advanced core profile includes the ability to process arbitrarily shaped video and still images. Advance coding efficiency profile (ACE) adds tools that can increase the compression ratio for both rectangular and arbitrarily shaped video objects. Simple studio profile is used to maintain the quality of video for studio distribution and editing functions. The studio simple only uses I frames. The core studio profile ads P-frames to the simple studio profile allowing a higher compression ratio at the expense of added complexity. The simple facial animation visual profile uses a mathematical model to animate facial images. This allows for the sending of a very small amount of data (e.g. lips move a small amount) to create 3 dimensional images. The simple face and body animation profile uses a mathematical model to animate both facial and body images. Advanced real time simple (ARTS) uses a low delay coding process along with robust error coding to provide real time two-way communication that cannot tolerate delays and which are subject to high error rates. The scalable texture visual profile allows for the mapping of textures onto images. Basic animated 2D texture visual profile has the capability to insert textures onto 2D images. Advanced scalable texture profile can perform scalable shape coding. This includes wavelet tiling. |
The hybrid visual profile includes the capability to decode synthetic and natural objects.
Advanced Video Coding Profiles Advanced video coding profiles are protocols and processes that are used to provide advanced video transfer and rendering capabilities to underlying MPEG services. The AVC profiles use the H.264/AVC coder that provides a compression ratio that is approximately double the MPEG-2 video coder. The AVC profiles range from low complexity baseline profile (BP) to an ultra high professional quality 4:4:4 profile. The baseline profile (BP) is used for low complexity device such as portable media players. Extended profile (XP) is designed for streaming video over networks that may have high packet loss (such as the Internet). The main profile was designed for broadcast (e.g. Cable TV) and storage (e.g. DVD) applications. High profile (HiP) can provide high definition television for broadcast and stored media distribution systems. High 10 profile (Hi10P) provides for increased quality allow for up to 10 bits per sample of decoded picture elements. High 4:2:2 profile (Hi422P) is used for professional applications that require higher levels of chroma (color) elements. High 4:4:4 profile (Hi422P) adds additional chroma (color) elements and allows up to 12 bits per image element. MPEG-J Profiles MPEG-J profiles are protocols and processes that are used to provide application programming interface (API) capabilities to underlying MPEG services. The MPEG-4 system has several MPEG-J profiles including personal profile and main profile. Personal MPEG-J profile is used for low complexity portable devices such as gaming devices, mobile telephones and portable media players. Main MPEG-J profile includes all the profiles from personal MPEG-J profiles plus APIs that can be used to select and configuring decoders along with interfaces to access service information. Scene Graph Profiles Scene graph profiles are the sets of protocols and processes that are used to define a composite (mixture) of media objects and how they relate to each other in an area. Scene graph profiles can be used to define the hierarchical relationship between videos, graphic images and audio components in a 2 dimensional or 3 dimensional
environment. |
21 Nov 2006 Subscribe FREE at www.IPTVMagazine.com
|
Some of the scene graphs profiles are based on the virtual reality modeling language (VRML) protocol. VRML is a text based language that is used to allow the creation of three-dimensional viewpoints, primarily for use with Web browsing.
The MPEG-4 system scene graph profiles include the basic 2D profile, simple 2D scene graph profile, complete 2D scene graph profile, core 2D profile, advanced 2D profile, main 2D profile, X3D profile, complete scene graph profile, audio scene graph profile and 3D audio scene graph profile. The basic 2D profile is used to define simple scenes on a 2 dimensional area. It is used for audio only or video only applications. The simple 2D scene graph profile can place media objects into a scene on a 2 dimensional area. It is a low complexity profile and it and does not define interaction with the media objects. Complete 2D scene graph profile allows for alpha blending and interactivity with the media objects. Alpha blending is the combining of a translucent foreground color with a background color to produce a new blended color. The core 2D profile can use both audio and visual media objects. It allows for local animation and interaction. Local animation is a process that changes parameters or features of an image or object over time that is processed at the location where the animation is playing (e.g. within a television or multimedia computer). The advanced 2D profile contains the capabilities of the basic 2D and core 2D profiles along with adding scripting capability and local interaction. The main 2D profile ads FlexeTime model which allows input sensors and additional tools for interactive applications. It is designed to interoperate with synchronized multimedia integration language (SMIL). SMIL is a protocol that is used to control the user interface with multimedia sessions. SMIL is used to setup and control the operation of media files along with the placement and operation windows on the users display. |
The X3D profile is a small footprint (limited memory and processing requirements) 3 dimensional media processing profile. It is designed to interoperate with X3D specification created by the Web3D consortium. Extensible 3D (X3D) is a storage, retreival and rendering (playback) industry standard for real time graphics media objects which can be adjusted in relative position and possibly interact with each other.
Complete scene graph profile is a combined set of 2D and 3D scene graph profiles from the binary format for scenes (BIFS) toolbox. Complete scene graph profile can be used for vitual gaming that have 3 dimensional worlds. Audio scene graph proifle is used for applications that only require audio media. The 3D audio scene graph profile describes how to position sound in a 3 dimensional environment. It allows for interaction of sounds with objects within the scene. Graphics Profiles Graphics profiles are sets of protocols and processes that are used to define and control graphics elements that are used in scenes. Some of the graphics profiles used in MPEG-4 systems include simple 2D graphics profile, simple 2D + text profile, core 2D profile, advanced 2D profile, complete 2D graphics profile, comlete graphics profile, 3D audio graphics profile, and X3D core profile. The simple 2D graphics profile is used for placing media objects in a scene that has only 2 dimensions. The simple 2D + text profile adds the ability to include text on the screen and to allow the text to move and be transformed (e.g. become transparent) with other media objects. Core 2D profile is used for picture in picture (PIP), video warping, animated advertisements and logo insertion. The advanced 2D profile adds graphic user interface (GUI) capabilities along with more complex graphics control for animation. |
|
|
|
| Figure 1.32., MPEG-2 Profile Types | |
22 Nov 2006 Definitions FREE at www.IPTVDictionary.com
|
Complete 2D graphics profile is a full set of 2D graphics control capabilities including the use of multiple shaped graphic objects. The complete graphics profile allows for the use of elevation grids, extrusions and lighitng effects to create virtual worlds.
3D audio graphics profile is used to define the acoustical properties of a scene. It includes features such as acoustics absorption, acoustic diffusion, acoustic transparency and tele-presence. The X3D core profile includes 3 dimenstional media object processing capabilities for advanced gaming and virtual environments. It is compatible with X3D specification. Figure 1.33 shows the different types of profiles used in the MPEG-4 systems. This table shows that the key MPEG-4 profile types include audio, visual, advanced video, MPEG-J, scenes and graphics profiles. The audio profiles range from very low bit rate to high quality audio that can be used in cinemas. The visual profiles range from low resolution profiles that can be used in portable devices to high quality studio profiles. Advanced video coding profiles take advantage of the high compression AVC coder. MPEG-J profiles allow the device to have direct programming interfaces to the MPEG media and devices. Scenes profiles define the relationships between media objects. Graphics profiles define media objects and how they are positioned on displays and how they can be changed in scenes. MPEG Levels MPEG levels are the amount of capability that a MPEG profile can offer. The use of levels allows products to define their quantitative capabilities such as memory size, resolution and maximum bit rates. For example, MPEG levels can range from low detail (low resolution) to very high capability (high resolution). Low level MPEG signals have a resolution format of up to 360 x 288 (SIF). Main level MPEG signals provide for resolution of up to |
720x576 (standard definition television). High level can support resolution of up to 1920x1152 (high definition). One of the more common profile combinations is the main profile at the main level (MP@ML). This combination provides television signals for standard definition (SD).
Figure 1.34 shows the different levels used in the MPEG-2 system. This table shows that low level has low resolution capability with a maximum bit rate of 4 Mbps. The main level has standard definition (SD) capability with a maximum bit rate of 15 Mbps. The high 1440 profile is a high definition format that has a maximum bit rate of 60 Mbps. The high level profile has a high definition format with 1152 lines that have 1920 pixels per line which can have a bit rate of up to 80 Mbps. Conformance Points Conformance points are a combination of profiles and levels in a system (such as an MPEG system) where different products can interoperate (by conforming to that level and profile). An example of conformance points is if ability of a mobile video server to support creation and playback of a simple visual profile at level 0, any mobile phone that has these conformance points should be able to play a video with these profiles and levels. |
|
|
|
Series Source: Introduction to MPEG $19.99 printed $16.99 eBook |
|
|
|
||
| Figure 1.34., MPEG-2 Levels | ||
23 Nov 2006 Subscribe FREE at www.IPTVMagazine.com
Copyright © 2007 Althos Publishing. All rights reserved.
