For the purposes of the present document, the terms and definitions given in TR 21.905 and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in TR 21.905. IM session: Telepresence:

For the purposes of the present document, the abbreviations given in TR 21.905 and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in TR 21.905. AS BFCP CLUE DTLS IMS MCC MRF MRFC MRFP SCTP SDP SIP TP TP UE UE

As an architectural framework for provision of IP multimedia services, IMS is capable of delivering various service functionalities and easing to integrate with new kinds of application, such as telepresence. Compared to traditional video conference, telepresence is a communication system with multiple cameras, microphones and screens that has the characteristics of gaze direction, eye contact, spatial audio and scaling images to true size, which in all to achieve the immersive "being there" experience for participants. IMS is using the IETF defined session control mechanism with the inherited capability to negotiate multiple multimedia streams in one single session, which could be applied as a basis for supporting telepresence in IMS that always has the necessity of producing and rendering various media streams with high qualities among the involved parties, even in the point to point case. Based on the existing procedures as specified in TS 24.229 and TS 24.147, this specification introduces updates and enhancements for IMS by incorporating CLUE with SIP, SDP and BFCP to facilitate controlling multiple spatially related media streams in an IM session supporting telepresence. To provide a "being there" experience for conversational audio and video session between remote locations, a variety of information needs to be coordinated, such as:

• audio and video spatial information;
• information to enable eye contact, gaze awareness, body language and natural image size; and
• information to coordinate the environments;

A spatial relationship is representative of the arrangement in space of two or more objects in the same capture scene, in contrast to retain in time or other relationships. It involves mainly both video and audio sources in telepresence conferencing system. For video, a spatial description of source video images sent in video streams, which includes the order of images in the actual captured scene and may be in two or three dimensions, enables a reproduction of the original scene at the receiver side. For audio, a spatial description of source audio including the point of capture, line of capture and capture sensitivity pattern (i.e. omni, shotgun, cardioid, hyper-cardioid) enable a reproduction at the receiver side in a spatially correct manner. Spatial matching may also be needed between audio and video streams coming from a specific party. When advertising video and audio media captures in an IM session supporting telepresence, a TP UE as well as a TP enabled conference focus sends spatial information, e.g. physical dimensions of the capture area for each video capture, and associated audio captures spatial information. This allows the receiving party to coordinate the capture scenarios and perform a proper rendering. Consider a TP UE of a typical triple-screen/camera system as example, of which each camera can provide one video capture for each 1/3 section of the room. Each capture has spatial information to indicate the scope of view, where a capture showing a zoomed out view of the whole room has the spatial information indicating a full global view.

The media information is enhanced in an IM session supporting telepresence by introducing the source components of the media capture, e.g. the original media captures like a camera or a composed media captures indicating a mix of audio streams, a composed or switched media capture indicating the dynamic or most appropriate subset of a "whole". The enhanced media information enables a sender to describe the sources and a receiver to choose which it wants to see. Taking the above example in subclause 4.2, the TP UE can further provide a single capture representing the active speaker based on voice activity detection, and a single capture representing the active speaker with the other two captures composed as a picture in picture. The media information can be used to distinguish the media captures from each other. Further media information may also be needed, such as simultaneity constraints. For example, a room camera have two options which are zoomed-in view and zoomed-out view, but there is no way to get them simultaneously.

Meeting description includes view information, language information, person information and type, as described below which enable the receivers to choose and render different captures.

• View information: indicates a physical or logical region as captured;
• Language information: used in case of multi-lingual and/or accessible conferences;
• Person information: provides specific information about people participating within a multi-media conference; and
• Person type: indicates the type of people participating within a multi-media conference with respect to the meeting agenda.

In addition, there may be some descriptive information which contains a relative priority between different captures, embedded textual information, or additional complementary information.

Presentation indicates resource sharing from one or more specific devices, including slides, video, data and etc. Presentations may have unfixed sources, which varies in placement and can be seen by all the involved parties.

The information detailed above may be used to obtain a better experience during an IM session between involved parties with different capabilities, such as different number of devices, different picture aspect ratios, or different number of media streams for sending and receiving. The usage of the information depends on the application scenarios. The TP UE described in the example of subclause 4.2 can provide at least 6 video captures, the message for negotiation therefore needs to contain enough parameters to describe the characteristics of each capture in order to allow the receiving party to clearly differentiate the captures and provide a proper rendering, e.g. spatial view, media composition, person information and type, and etc. Protocols adopted for IM session supporting telepresence, which exchanges above information among involved parties, enables interoperability by handling multiple streams in standardized way.