CN101031065A

CN101031065A - Method, apparatus and system for switching pictures in video service

Info

Publication number: CN101031065A
Application number: CNA200710097575XA
Authority: CN
Inventors: 朱祥文; 吴宗武; 周旭锋; 莫燕; 瞿新; 刘家锋; 李渊
Original assignee: Huawei Technologies Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2007-04-27
Filing date: 2007-04-27
Publication date: 2007-09-05
Anticipated expiration: 2027-04-27
Also published as: WO2008131644A1; CN100583985C

Abstract

The invention is used for solving the issue existed in currently-used technology that is the video service users can not select image-displaying mode according to their needs. The method thereof comprises: multi-point controller receives a switch request message comprising the amount of sub-images indicated by the image-displaying mode; according to the image-displaying mode the multi-point controller determines the image mixer required by the image-displaying mode, and outputs the images corresponding to the image-displaying mode to the user end sending the switch request message.

Description

A kind of method, Apparatus and system of in video traffic, realizing the picture switching

Technical field

The present invention relates to the multimedia video technique of the communications field, particularly a kind of method, Apparatus and system of in video traffic, realizing the picture switching.

Background technology

Video conferencing system allows meeting participant interactive voice, video and data in real time, be very easy between the meeting participant communication with exchange.In the video conferencing system network configuration as shown in Figure 1, multipoint control unit (MCU, Multipoint Control Unit) connect each meeting-place participate in a conference, be used for reception meeting field signal, processing meeting field signal and will handle after the meeting field signal be issued to each meeting-place of participant; Subscriber terminal equipment is mainly realized the input and output of voice, image and data, and communicates functions such as connection and data transmit-receive; Plural subscriber terminal equipment and MCU have made up a multiple spot, real-time conference communication environment, and the user watches other users' image and listen to other users' sound by terminal equipment.

For satisfying the demand that participant in the video conference can understand other participants' of all or part image information at one time simultaneously, prior art provides the mode that can realize in video conference that many pictures show:

The main handling process of the implementation method that a kind of many pictures show is as follows:

1, will pass through H.221 frame processing module processing from the meeting-place image terminal data of each conference terminal, and form and satisfy the H.221 multiple source high speed serialization view data code stream of standard;

2, described multiple source high speed serialization code image data stream is by source high-speed serial data passage, the footpath switching net exchanges to one or more target high-speed serial data passages, forms one or more target high speed serialization view data code streams of video wall target monitoring terminal;

3, the destination image data code stream is sent to the picture decoding module, forms H.261 form or the H.263 laggard line number word image decoding of image code stream of form, obtain CCIR601 format standard digital picture and CLV Common Intermediate Format CIF standard digital image through separating frame;

4, CCIR601 format standard digital picture is carried out analog image coding, be converted to and be sent to video wall target monitoring terminal behind the analog video signal and show;

5, described CIF standard digital image is sent to the combined picture that the picture composition module combinations becomes certain pattern, form the multiple image data code flow with carrying out framing behind the combined picture coding, by switching net exchange and H.221 the frame processing module send to each terminal after handling, thereby realize that the many pictures of each terminal show.

This technical scheme is primarily aimed at certain concrete TV conference system, need establish the video conferencing wall at the MCU bogie side frame, so application limitation is bigger; Each can only watch identical multiple image with the user in this method in addition, can not independently select picture, can not realize simultaneously mutual conversion between many pictures and the single-image to select interested a certain participant's image information, therefore many menu display functions are single.

Another kind of realize the method for many pictures by the transcoder of video conferencing system, its main handling process is as follows:

1, the user video signal that receives is decoded, sampled, and each vision signal that will obtain after will sampling deposits in the internal memory;

2, from this internal memory, read each vision signal after the sampling by a plurality of picture recombiners and carry out compoundly, form a plurality of identical compound many pictures;

3, encoded by the corresponding codes device in each compound many pictures back of sampling, and the vision signal after the output encoder compression.

This technical scheme has solved the problem that a plurality of user video signals have different image pixel precision by transcoder, but, each participant user can only watch identical multiple image in this method, can not independently select picture, can not realize simultaneously mutual conversion between many pictures and the single-image to select interested a certain participant's image information, therefore many menu display functions are single.

The number of number that has participated in also providing in video conference is carried out according to video conference in the prior art or the video source that is input to video conferencing system is the method for the composite mode of the many pictures of decision dynamically, when the number of participant in the video conference increases or reduce, the corresponding dynamic adjustment of the composite mode of many pictures, but, each participant user can only watch identical multiple image in this method, can not independently select picture, can not realize simultaneously mutual conversion between many pictures and the single-image to select interested a certain participant's image information, many menu display functions are single.

In summary it can be seen, exist the single problem of many menu display functions in the process that the many pictures of realization show in the existing video conference, each participant user can only watch identical multiple image simultaneously, and can not be according to watching needs to select the picture display mode, can not between many pictures display mode and single-image display mode, switch simultaneously, therefore in video conference, can not satisfy user's the demand of watching.

Summary of the invention

The embodiment of the invention provides a kind of method, Apparatus and system that picture switches of realizing in video traffic, be used to realize that the user who participates in video traffic can independently select the picture display mode to watch image as required.

A kind of method that realizes the picture switching in video traffic comprises step:

Multipoint control unit MCU receives handoff request message, and described request message comprises that the picture display mode shows the sprite number that picture shows;

Described MCU determines the picture blender of needs according to described picture display mode, and to the user terminal output and the corresponding picture image of picture display mode of asking that send handoff request message.

A kind of method that realizes many pictures demonstrations in video traffic comprises step:

Multipoint control unit MCU specifies the picture blender of an acquiescence;

Described acquiescence picture blender carries out many picture compositions according to the control information of many pictures to the video source image information of the user terminal correspondence of participation video traffic;

Multiple image after described MCU will make up is exported to each user terminal.

A kind of picture device for switching that realizes in video traffic comprises:

Control module is used to manage the control information of many pictures, and to the user terminal displaying video menu prompt information that participates in video traffic, and according to the dynamically definite picture blender of the picture display mode of user's selection;

Decoder module is used for the video source code image data stream of each user terminal is decoded;

The picture composition module is used for carrying out many picture compositions according to described many pictures control information by one or more picture blenders;

Coding module is used for the multiple image information after the combination is carried out corresponding encoded, and the multiple image that obtains is exported to user terminal.

A kind of video service system, a plurality of user terminals that comprise multipoint control unit MCU, link to each other with described MCU by network; Described MCU is to the user terminal displaying video menu prompt information that participates in video traffic; User terminal sends handoff request message according to the picture display mode that the user selects to described MCU, and described picture display mode shows the sprite number that picture shows; Described MCU determines the picture blender of needs according to described picture display mode, and to the user terminal output and the corresponding picture image of picture display mode of asking that send handoff request message.

Above-mentioned multipoint control unit MCU exports the picture image corresponding with this picture display mode according to the picture display mode that the user selects to the terminal at this user place, the user who participates in video traffic can watch image with identical picture display mode, also can independently select the picture display mode to watch image as required, realize the user in the flexible switching between many pictures display mode, between many pictures display mode and the single-image display mode, satisfied the demand that the user differentiation is watched conference image.

Description of drawings

Fig. 1 is a meeting video system configuration schematic diagram in the prior art;

Fig. 2 is a multipoint control unit structural representation in the embodiment of the invention;

Fig. 3 is a picture composition modular structure schematic diagram in the embodiment of the invention;

Fig. 4 is many pictures display mode schematic diagram in the embodiment of the invention;

Fig. 5 A be in the embodiment of the invention a plurality of user terminals with the realization schematic diagram of four picture display mode display conference images;

Fig. 5 B be in the embodiment of the invention a plurality of user terminals with the schematic diagram of four picture display mode display conference images;

Fig. 6 A is the realization flow figure that user A selects six picture display mode display conference images in the embodiment of the invention;

Fig. 6 B is when user A selects six picture display modes in the embodiment of the invention, the schematic diagram of user A, B, C, D terminal display conference image;

Fig. 7 A is the realization flow figure that user C selects single-image display mode display conference image in the embodiment of the invention;

Fig. 7 B is when user C selects the single-image display mode in the embodiment of the invention, the schematic diagram of user A, B, C, D terminal display conference image;

Fig. 8 A is the realization flow figure of frame sign sprite in the embodiment of the invention;

Fig. 8 B is the demonstration schematic diagram of frame sign sprite in the embodiment of the invention.

Embodiment

For solve the problem that many menu display functions are single and the user can not change in the prior art between many pictures and single-image display mode, present embodiment is that example is elaborated to embodiment of the invention technical scheme with the video conferencing system.

Video conferencing system in the present embodiment as shown in Figure 1, the a plurality of user terminals that comprise multipoint control unit MCU, are connected with MCU by network, wherein, multipoint control unit MCU structure as shown in Figure 2, this MCU comprises control module 20, decoder module 21, picture composition module 22 and coding module 23; Wherein:

Control module 20, be used to manage the control information of many pictures, this information is by consulting to obtain with the user terminal that participates in video conference, and to the user terminal displaying video menu prompt information that participates in video conference, and the picture display mode dynamic assignment picture blender of selecting according to the user; Described many pictures control information comprises input video source information, picture mixed information and many pictures output information; Input video source information comprises the identification information of each decoder and each input video source image pixel precision etc.; The picture mixed information comprises the attribute information of number, multiple image pixel accuracy and each picture blender of all the picture blenders that need support of this meeting-place, wherein, the multiple image pixel accuracy is defined as the maximum image pixel accuracy in each participant terminal usually, also can be provided with according to system's needs, the attribute information of picture blender comprises the position of each sprite in pattern that many pictures show, this many pictures display mode and the identification information of the corresponding encoder of each sprite etc.; Many pictures output information comprises each encoder identification information, corresponding blender sign and output image pixel accuracy etc.

Described control module 20 is specified the picture blender of an acquiescence when the video conference initialization, the attribute of this picture blender can carry out dynamically changing according to the user terminal quantity that participates in video conference in the process in meeting, but can not delete; Carry out in the process in video conference, determine corresponding picture blender, can increase corresponding picture and mix blender, also can dynamically delete these picture blenders according to the picture display mode that the user selects.

Decoder module 21 is made up of a plurality of decoders, is used for the video source image code stream of the user terminal that participates in video conference is decoded, and sends the video source image information that decoding obtains to picture composition module 22 by data/address bus; The decoding algorithm that each image code stream of user terminal output is corresponding to be distributed a decoder, each decoder to use the image code stream encryption algorithm with input to be complementary (as H.263, MPEG-4 etc.) the input picture code stream is decoded, restore image information; Each decoder is all supported multiple decoding algorithm, carries out the initialization matching algorithm when distributing to the input picture code stream.

Picture composition module 22, be used for by one or more picture blenders a plurality of video source image informations of decoder output being carried out many picture compositions, and the multiple image information after will making up sends coding module 23 to by data/address bus according to the control information of many pictures; As shown in Figure 3, described picture composition module 22 further comprises: view data input unit 220 is used to receive the video source image information that described decoder module 21 is exported; A plurality of first samplers 221, be used for pixel accuracy and this video source image image pixel precision of correspondence position in many pictures of the corresponding video source image of view data input unit 220 are compared, when both are inconsistent, the video source image information is sampled, and the video source image information after will sampling outputs to picture blender 222; Otherwise, the video source image information is directly outputed to picture blender 222; Sampling comprises and rises sampling and down-sampled, as being that the image drop sampling of 704 * 576 (4CIF) is the image of 352 * 288 (CIF) with pixel accuracy, perhaps be that the image liter of 176 * 144 (QCIF) is sampled as the image that pixel accuracy is 352 * 288 (CIF) with pixel accuracy; Picture blender 222 be used for each video source image information of first sampler, 221 outputs is carried out many picture compositions, and the multiple image information after will making up outputs to second sampler 223; A plurality of second samplers 223, be used for the pixel accuracy of the multiple image of picture blender 222 output and the output image pixel accuracy of many pictures output information are compared, when both are inconsistent, multiple image information carried out corresponding sampling and will sample after the multiple image information that obtains send image data-outputting unit 224 to by data/address bus; Otherwise, multiple image information is directly outputed to image data-outputting unit 224; Image data-outputting unit 224 is used for image information is exported to coding module 23.

Picture blender 222 in the described picture composition module 22 can be one or more, the first picture blender is designated or be set to give tacit consent to the picture blender when conference initiation, the first picture blender attribute can carry out changing in the process in meeting, but can not be deleted; In meeting is carried out, can increase other picture blenders, also can dynamically delete this picture blender, to satisfy the demand of watching of user's differentiation; Control module 20 is according to user's operation information, input video source information in the control information of many pictures, picture mixed information and many pictures output information are refreshed, and first sampler 221, picture blender 222 and second sampler 223 are handled accordingly according to new many pictures control information.

Coding module 23, form by a plurality of encoders, be used for the multiple image information of picture composition module 22 outputs is encoded, the multiple image compressed bit stream that formation and participant decoding terminals algorithm are complementary, and it is outputed to each participant terminal respectively, the multiple image compressed bit stream that the participant end-on is received carries out behind the corresponding decoding showing in this locality multiple image; Each encoder is all supported multiple encryption algorithm, carries out initialization coupling when distributing to the output image code stream, and encoder can corresponding one or more participant terminals; The schematic diagram that the many pictures of participant terminal show as shown in Figure 4, many pictures display mode of each picture blender support can be wherein any, but does not limit to several situations of diagram.

In the present embodiment, MCU specifies the picture blender of an acquiescence and selects the picture display mode according to service needed during the video traffic initialization, thereby further carry out many picture compositions and realize that many pictures show by this acquiescence picture blender, can reduce the usage quantity of picture blender like this, thereby save resource.

In order to be illustrated more clearly in the implementation procedure that many pictures show in the present embodiment, describe below by concrete example.

Example one: user A, B, C and D terminal are connected in the video conference, and wherein, the video attribute of user A, B, C and D terminal is as shown in the table:

User terminal	Image compression algorithm	The image pixel precision
User terminal	Image compression algorithm	The image pixel precision	A	H.263	CIF(352×288)
B	MPEG-4	CIF(352×288)	A	H.263	CIF(352×288)
B	MPEG-4	CIF(352×288)	C	H.263	QCIF(176×144)
D	MPEG-4	QCIF(176×144)	C	H.263	QCIF(176×144)

Control module initialization user A, B, C, D terminal are watched many pictures display mode of meeting, many pictures display mode of the first picture blender is set to four picture display modes, the image that is user A, B, C, D terminal respectively accounts for 1/4 of picture, and the multiple image pixel accuracy is chosen as CIF (352 * 288); Shown in Fig. 5 A, the realization principle that four pictures show is as follows:

After the user inserted the meeting-place, control module was distributed to each user's corresponding decoder and initialization decoding algorithm, and the video source image code stream that each decoder is imported each user is exported after being reduced into corresponding video source image information; H.263 the decoding algorithm that user A terminal corresponding decoder 1 and user C terminal corresponding decoder 3 are used is initialized as, and the decoding algorithm that user B terminal corresponding decoder 2 and user D terminal corresponding decoder 4 are used is initialized as MPEG-4;

Decoder

1,2,3,4 is decoded to the image code stream of user A, B, C, D terminal respectively, and the video source image information that reduction obtains is outputed to view data input unit in the picture composition module respectively.

First sampler

1,2 judges that according to the pixel accuracy and the multiple image pixel accuracy of the video source image of user A, the input of B terminal the decoded picture informational needs of user A, B terminal correspondence is down-sampled, therefore

first sampler

1,2 carries out down-sampledly respectively to the decoded picture information of user A, B terminal correspondence, and the image pixel precision that will obtain after will sampling is that the image of QCIF (176 * 144) outputs to the picture blender; Because the decoded picture pixel accuracy of user C, D terminal correspondence is QCIF (176 * 144), therefore

first sampler

3,4 judges that its image information does not need to sample, and directly decoded picture information is outputed to the picture blender.

The video source image information of user A, the B that the picture blender is exported

first sampler

1,2,3,4 according to the many pictures display mode in the picture mixed information, C, D terminal correspondence is combined into a four-picture image information, and outputs it to second sampler.

Second sampler

1,2 according to the output image pixel accuracy of each user terminal judge this four-picture image information pointer to user A, B terminal not needs sample, by image data-outputting unit it is exported to

corresponding codes device

1,2 respectively;

Second sampler

3,4 judges that this four-picture image information pointer need carry out down-sampled to user C, D terminal, therefore, 3,4 pairs of four-picture image information of second sampler are down-sampled, and the image pixel precision that will obtain after will sampling is that the image of QCIF (176 * 144) is exported to

corresponding codes device

3,4 respectively by image data-outputting unit.

The four-picture image information of 1,2,3, the 4 pairs of picture composition modules of encoder output is encoded respectively, and the image compression code stream that output and user A, B, C, D decoding terminals algorithm are complementary is to corresponding user terminal.

Show four-picture image respectively in this locality after the image compressed code stream that user A, B, C, D end-on are received is decoded, the schematic diagram of its demonstration is shown in Fig. 5 B.

User A, B, C, D terminal are watched conference image with four identical picture display modes in the above-mentioned example, present embodiment user in meeting is carried out can pass through terminal operation and select other many pictures display modes to watch conference image, and free image switching display mode and do not influence watching of other participant sides, as shown in Figure 6A, the user to switch the specific implementation process of many pictures display mode as follows:

Step 600, MCU carry out timing displaying video menu prompt information in the process in meeting, and how the prompting user carries out associative operation.

Step 601, user select corresponding many picture display mode by terminal operation according to information when other many pictures display modes of needs are watched conference image, send handoff request by user place terminal to MCU.

Step 602, MCU upgrade the control information of many pictures according to handoff request, determine the picture blender that needs.

Step 603, MCU are carried out many picture compositions according to the many pictures control information after upgrading by the picture blender of determining, and satisfy the multiple image of many pictures display mode that the user selects and to the corresponding multiple image of other user terminals outputs to the user terminal output that sends handoff request.

Example two: in the meeting of example one is carried out, user A selects six picture display modes to watch conference image by terminal operation, send the request of image switching display mode to MCU by user A terminal, MCU upgrades the picture mixed information according to this request, and determine to need increasing by the second picture blender, its attribute comprises the position of each sprite in six picture display modes, this six pictures display mode and each sprite corresponding codes device identification information etc.First sampler is selected the decoded video source image information of user A, B, C, D terminal correspondence and it is carried out corresponding sampling processing according to the attribute information of input video source information and picture mixed information and the first picture blender and the second picture blender; The first picture blender and the second picture blender carry out four picture compositions and six picture compositions to each video source image information respectively; Four-picture image information after second sampler makes up the first picture blender is exported to user B terminal corresponding codes device 2 by image data-outputting unit, carry out the four-picture image information after the combination down-sampled, and the four-picture image information after will sampling exports to user C, D terminal corresponding codes device 3,4 by image data-outputting unit, and six picture image information after the combination of the second picture blender are exported to user A terminal corresponding codes device 1 by image data-outputting unit; After encoding, the multiple image information of the corresponding output of 1,2,3,4 pairs of picture composition modules of encoder exports to corresponding user terminal; The multiple image that shows correspondence after the image compressed code stream that user A, B, C, D end-on are received is decoded in this locality respectively, be user A watches meeting on terminal six picture images, user B, C, D watch the four-picture image of meeting on terminal separately, the schematic diagram that the many pictures of each user terminal show is shown in Fig. 6 B.

The user selects many pictures display mode to watch conference image by terminal operation, also can select the single-image display mode to watch conference image as required simultaneously, and shown in Fig. 7 A, the user selects the single-image display mode to watch the specific implementation process of conference image as follows:

Step 700, MCU carry out timing displaying video menu prompt information in the process in meeting, and how the prompting user carries out associative operation.

Step 701, user select corresponding single-image display mode by terminal operation according to information when needs single-image display mode is watched conference image, send handoff request by user place terminal to MCU.

Step 702, MCU upgrade the control information of many pictures according to handoff request.

Step 703, MCU are directly exported corresponding single-image image to the user terminal that sends handoff request, and mix by original picture according to the control information of many pictures and to carry out many picture compositions, to the corresponding multiple image of other user terminals outputs.

Example three: in the meeting of example two is carried out, after user C selects to watch the single-image image of user A by terminal operation, user C terminal sends the request of image switching pattern to MCU, MCU upgrades the picture mixed information according to this request, and only need to determine the first picture blender and the second picture blender, its corresponding attribute information remains unchanged; First sampler is selected the decoded picture information of user A, B, C, D terminal correspondence and is carried out corresponding sampling processing according to the attribute information of input video source information and picture mixed information and the first picture blender and the second picture blender; The first picture blender and the second picture blender carry out four picture compositions and six picture compositions respectively; Four-picture image information after second sampler makes up the first picture blender is exported to user B terminal corresponding codes device 2 by image data-outputting unit, carry out the four-picture image information after the combination down-sampled, and the four-picture image information that will obtain after will sampling exports to user D terminal corresponding codes device 4 by image data-outputting unit, and six picture image information after the combination of the second picture blender are exported to user A terminal corresponding codes device 1 by image data-outputting unit; Image information after first sampler will be sampled after with the down-sampled one-tenth of the decoded picture pixel information of user A terminal correspondence QCIF (176 * 144) outputs to user C terminal corresponding codes device 3 by image data-outputting unit; After encoding, the picture image information of the corresponding output of 1,2,3,4 pairs of picture composition modules of encoder exports to corresponding user terminal; The picture image that shows correspondence after the image compressed code stream that user A, B, C, D end-on are received is decoded in this locality respectively, it is user C watches user A on terminal image information, user A watches the image information of six image modes on terminal, user B and user D watch the image information of four pictures respectively on terminal; The schematic diagram that the many pictures of each user terminal show is shown in Fig. 7 B.

In addition, in the present embodiment certain sprite in many pictures is carried out the frame sign to highlight this picture image, for example the sprite to the participant side D of current speaking carries out the frame sign to highlight its image, shown in Fig. 8 A, realizes that the detailed process that the frame sign highlights is as follows:

Step 800, MCU need to determine the sub-screen information of the user terminal correspondence of frame sign.

Step 801, MCU upgrade the control information of many pictures.

Step 802, MCU record this sub-screen information in the picture blender.

Step 803, picture blender carry out the frame sign to this sub-screen image after finishing many picture compositions.

The picture image that step 804, MCU have the frame sign to user terminal output.

The process of the decoding of image information in this process, many picture compositions and coding and the implementation procedure in the foregoing description in like manner repeat no more; The schematic diagram that many pictures of user A, B, C, D terminal show is shown in Fig. 8 B.

In the implementation method that the many pictures that provide in the present embodiment show, the participant terminal is watched conference image with identical many pictures display mode when session initial, in the carrying out of meeting, the user can select to watch the many pictures display mode or the single-image display mode of conference image as required, satisfied the demand that user's differentiation is watched conference image; The participant user can be switched its implementation simple and flexible in the present embodiment between many pictures display mode and single-image display mode by terminal operation; In addition, in the present embodiment sprite in many pictures is carried out the frame sign and highlight picture, satisfied professional diversified needs.

It is above-mentioned that to be example with the video conference describe the implementation procedure of embodiment of the invention technical scheme, but embodiment of the invention technical scheme is not limited to video conference, also be adapted to other video traffics, as video interactive voice answer-back (IVR, Interactive Voice Response) business, in many ways video communication business of Can Yuing etc., when in other video service systems, realizing said method to MCU functional similarity or identical in the requirement of processor (as MCU) on function and the present embodiment, its process that realization picture shows and picture switches and the implementation procedure in the present embodiment in like manner repeat no more.

Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.

Claims

1, a kind of method that realizes the picture switching in video traffic is characterized in that, comprises step:

2, the method for claim 1 is characterized in that, described MCU judges when user terminal requests is switched the single-image display mode, directly exports corresponding single-image image to the user terminal of this handoff request message of transmission; Described MCU judges when user terminal requests is switched many image modes, is undertaken exporting corresponding multiple image to the user terminal that sends this handoff request message behind many picture compositions by the picture blender of determining.

3, method as claimed in claim 1 or 2, it is characterized in that, when the video traffic initialization, described MCU specifies the picture blender of an acquiescence, carry out many picture compositions by described acquiescence picture blender according to the control information of many pictures, and the multiple image after will making up is exported to each user terminal.

4, method as claimed in claim 3 is characterized in that, described MCU on-the-fly modifies the attribute information of described acquiescence picture blender according to the user terminal quantity that participates in video traffic in video traffic carries out.

5, method as claimed in claim 4 is characterized in that, described attribute information comprises the display position of many pictures display mode, each sprite.

6, method as claimed in claim 3 is characterized in that, described MCU will need the relevant information records of sub-screen image of frame sign in the picture blender, and described picture blender carries out the frame sign to this sub-screen image after finishing many picture compositions.

7, a kind of method that realizes many pictures demonstrations in video traffic is characterized in that, comprises step:

Multipoint control unit MCU specifies the picture blender of an acquiescence;

8, method as claimed in claim 7 is characterized in that, described MCU on-the-fly modifies the attribute information of described acquiescence picture blender according to the user terminal quantity that participates in video traffic in video traffic carries out.

9, method as claimed in claim 8 is characterized in that, described attribute information comprises the display position of many pictures display mode, each sprite.

10, as each described method of claim 7-9, it is characterized in that, described MCU will need the relevant information records of sub-screen image of frame sign in the picture blender, and described picture blender carries out the frame sign to this sub-screen image after finishing many picture compositions.

11, a kind of picture device for switching that realizes in video traffic is characterized in that, comprising:

12, device as claimed in claim 11 is characterized in that, described control module specifies the picture blender of an acquiescence to carry out many picture compositions when the video traffic initialization; Described control module video traffic carry out in the process according to the picture display mode that the user selects determine needs the picture blender.

13, device as claimed in claim 12 is characterized in that, described control module on-the-fly modifies the attribute information of described acquiescence picture blender according to the user terminal quantity that participates in video traffic in video traffic carries out.

14, as each described device of claim 11-13, it is characterized in that, described control module will need the relevant information records of sub-screen image of frame sign in the picture blender, and described picture blender carries out the frame sign to this sub-screen image after finishing many picture compositions.

15, device as claimed in claim 14 is characterized in that, described device is multipoint control unit MCU.

16, a kind of video service system is characterized in that, a plurality of user terminals that comprise multipoint control unit MCU, link to each other with described MCU by network; Described MCU is to the user terminal displaying video menu prompt information that participates in video traffic; User terminal sends handoff request message according to the picture display mode that the user selects to described MCU, and described picture display mode shows the sprite number that picture shows; Described MCU determines the picture blender of needs according to described picture display mode, and to the user terminal output and the corresponding picture image of picture display mode of asking that send handoff request message.

17, system as claimed in claim 16 is characterized in that, described MCU specifies the picture blender of an acquiescence to carry out many picture compositions when the video traffic initialization; Described MCU video traffic carry out in the process according to the picture display mode that the user selects determine needs the picture blender.

18, system as claimed in claim 17 is characterized in that, described MCU carries out in the process at video traffic, on-the-fly modifies the attribute information of described acquiescence picture blender according to the user terminal quantity that participates in video traffic.

19, as each described system of claim 16-18, it is characterized in that described MCU comprises:

20, system as claimed in claim 19, it is characterized in that, described control module will need the relevant information records of sub-screen image of frame sign in the picture blender, and described picture blender carries out the frame sign to this sub-screen image after finishing many picture compositions.