18-07-2011, 10:15 AM
[attachment=14533]
Abstract
This paper presents a case study of an adaptive multimedia message box (M3-Box) based on MPEG-7
technology. To enable a bi-directional communication, MPEG-7 related techniques such as the
Extensible Stylesheet Language (XSL) are incorporated into the M3-Box. MPEG-7 descriptions for the
multimedia access, user preferences, and media summarization are instantiated and transmitted along
the media content of the messages. These components allow to set up a message center which integrates
different kinds of structured messages containing media elements such as video, voice and text. They
also provide access to these messages with a variety of different devices. The paper analyses the
scenario, gives a specification of the application requirements, and describes an implementation of the
M3-Box.
1. Introduction
Nowadays the number of device classes which are used for communication is growing very fast. These
devices are most often adapted to the situations where they should be used in. Accordingly we communicate
with each other by using different devices which are often restricted with respect to the bandwidth and their
capabilities. But this enormous growth of the number of mobile multimedia enabled device types gives us the
possibility to be contacted most of the time during the day not depending what we are doing or where we are.
To control this “contactability” an adaptive messaging is required. For traditional communications like
phone, facsimile and e-mail first unified messaging center products exist, providing access to messages while
not restricting the type of information delivered by the capabilities of the device which is used.
In this paper we present the architecture of the Multi Media Message Box (M3-Box), a unified multimedia
message center. While adaptive technologies based on MPEG-7 description schemes are engaged here, the
main focus of this application is the handling of structured multimedia messages. The structuring information
which can be generated by using different communication devices is coded according to the upcoming
MPEG-7 standard [1,2].
The Paper is structured in the following way: in the next section related technologies are discussed. In section
three the scenario of the unified multimedia messaging is proposed in detail. Then we describe the
requirements of a unified multimedia message center. In section five the overall design of the M3-Box is
explained giving details on the generation, storage and consumption of structured messages with different
communication devices and the M3-Box. Finally a conclusion is drawn.
2. Related Work
The MultiMedia Message Box application focuses on the one side on the generation of structured messages
which includes several multimedia elements to enable an intuitive offline communication. On the other side it
deals with the problem to overcome the barrier between the communication of different communication
devices. When describing these structured messages in the MPEG-7 information, the meta information can be
used to adapt the presentation to the output device which is used. Algorithms to control the adaptation of
multimedia data to different devices have already been proposed. In [3] an InfoPyramid is specified to
determine the optimal adaptation of multimedia presentations with respect to the bandwidth which is needed
for transmission and the loss of information. For this purpose possible adaptations and transcodings of the
media element are described regarding its size and importance. In [4] a dynamic adaptation while
transmitting the media element is proposed. This also includes a switching of the media type if required by
the transmission conditions. A consistent representation of the content is achieved by annotating the content
of the media elements. Both solutions describe alternative representations of the media content with respect
to the loss of information and the reduction in size.
Proc. Second Int’l Symposium on Mobile Multimedia Systems & Applications, Delft, November 9-10, 2000, pp. 6-13.
A similar approach is presented as a draft "Annotation of web content for transcoding" to the W3C [5]. This
Resource Description Framework (RDF) based approach allows to describe these hints separate from the
content itself. Similar to the approach taken by the InfoPyramid, MPEG-7 provides the possibility to describe
the media content with respect to possible transcodings which can be applied, and how significant these
representations are regarding the contained information [6][7]. Additionally also hints to simplify the
transcoding can be specified. For instance in the case of sequences which contain slow motion the search
range of the motion prediction in a video transcoder can be limited with no loss of information but significant
faster transcoding.
For transcoding of media on the server side, the client device, the transmission channel, and the user
preferences should be known to adapt the media to these circumstances. An architecture to transmit this
information is proposed by the "Composite Capabilities/Preference Profiles: Requirements and Architecture"
in [8]. In this approach the device capabilities and the user preferences regarding the media formats are
specified and transmitted to the server by two or three tier architectures. While a device description is not
specified within MPEG-7 it seems that describing media types which can be rendered on the device is
sufficient for most cases. So not the client device is described but the media which should be received. This
has furthermore the advantage that with using MPEG-7 the target media type for transcoding can also depend
on the semantical content of the media described within MPEG-7.
As already mentioned composing and handling of structured messages different requirements than dealing
with sequential media files or multimedia presentations. Therefore the requirements on the components of the
communication chain are specified after presenting the scenario of the multimedia messaging in the next
section in more detail.
Abstract
This paper presents a case study of an adaptive multimedia message box (M3-Box) based on MPEG-7
technology. To enable a bi-directional communication, MPEG-7 related techniques such as the
Extensible Stylesheet Language (XSL) are incorporated into the M3-Box. MPEG-7 descriptions for the
multimedia access, user preferences, and media summarization are instantiated and transmitted along
the media content of the messages. These components allow to set up a message center which integrates
different kinds of structured messages containing media elements such as video, voice and text. They
also provide access to these messages with a variety of different devices. The paper analyses the
scenario, gives a specification of the application requirements, and describes an implementation of the
M3-Box.
1. Introduction
Nowadays the number of device classes which are used for communication is growing very fast. These
devices are most often adapted to the situations where they should be used in. Accordingly we communicate
with each other by using different devices which are often restricted with respect to the bandwidth and their
capabilities. But this enormous growth of the number of mobile multimedia enabled device types gives us the
possibility to be contacted most of the time during the day not depending what we are doing or where we are.
To control this “contactability” an adaptive messaging is required. For traditional communications like
phone, facsimile and e-mail first unified messaging center products exist, providing access to messages while
not restricting the type of information delivered by the capabilities of the device which is used.
In this paper we present the architecture of the Multi Media Message Box (M3-Box), a unified multimedia
message center. While adaptive technologies based on MPEG-7 description schemes are engaged here, the
main focus of this application is the handling of structured multimedia messages. The structuring information
which can be generated by using different communication devices is coded according to the upcoming
MPEG-7 standard [1,2].
The Paper is structured in the following way: in the next section related technologies are discussed. In section
three the scenario of the unified multimedia messaging is proposed in detail. Then we describe the
requirements of a unified multimedia message center. In section five the overall design of the M3-Box is
explained giving details on the generation, storage and consumption of structured messages with different
communication devices and the M3-Box. Finally a conclusion is drawn.
2. Related Work
The MultiMedia Message Box application focuses on the one side on the generation of structured messages
which includes several multimedia elements to enable an intuitive offline communication. On the other side it
deals with the problem to overcome the barrier between the communication of different communication
devices. When describing these structured messages in the MPEG-7 information, the meta information can be
used to adapt the presentation to the output device which is used. Algorithms to control the adaptation of
multimedia data to different devices have already been proposed. In [3] an InfoPyramid is specified to
determine the optimal adaptation of multimedia presentations with respect to the bandwidth which is needed
for transmission and the loss of information. For this purpose possible adaptations and transcodings of the
media element are described regarding its size and importance. In [4] a dynamic adaptation while
transmitting the media element is proposed. This also includes a switching of the media type if required by
the transmission conditions. A consistent representation of the content is achieved by annotating the content
of the media elements. Both solutions describe alternative representations of the media content with respect
to the loss of information and the reduction in size.
Proc. Second Int’l Symposium on Mobile Multimedia Systems & Applications, Delft, November 9-10, 2000, pp. 6-13.
A similar approach is presented as a draft "Annotation of web content for transcoding" to the W3C [5]. This
Resource Description Framework (RDF) based approach allows to describe these hints separate from the
content itself. Similar to the approach taken by the InfoPyramid, MPEG-7 provides the possibility to describe
the media content with respect to possible transcodings which can be applied, and how significant these
representations are regarding the contained information [6][7]. Additionally also hints to simplify the
transcoding can be specified. For instance in the case of sequences which contain slow motion the search
range of the motion prediction in a video transcoder can be limited with no loss of information but significant
faster transcoding.
For transcoding of media on the server side, the client device, the transmission channel, and the user
preferences should be known to adapt the media to these circumstances. An architecture to transmit this
information is proposed by the "Composite Capabilities/Preference Profiles: Requirements and Architecture"
in [8]. In this approach the device capabilities and the user preferences regarding the media formats are
specified and transmitted to the server by two or three tier architectures. While a device description is not
specified within MPEG-7 it seems that describing media types which can be rendered on the device is
sufficient for most cases. So not the client device is described but the media which should be received. This
has furthermore the advantage that with using MPEG-7 the target media type for transcoding can also depend
on the semantical content of the media described within MPEG-7.
As already mentioned composing and handling of structured messages different requirements than dealing
with sequential media files or multimedia presentations. Therefore the requirements on the components of the
communication chain are specified after presenting the scenario of the multimedia messaging in the next
section in more detail.