Multi-view video consists of multiple video sequences captured simultaneously by multiple closely spaced cameras and enables users to freely change their viewpoints by playing different video sequences. Therefore, the transmission bitrate of multi-view video requires more bandwidth than conventional multimedia. In order to reduce the bandwidth requirement, User Dependent Multi-view Video Transmission (UDMVT), which is based on Multi-view Video Coding (MVC), has been proposed for single users. In UDMVT, the same frames are encoded into different versions for each user, which increases the transmission bitrate for multiple users because of the redundant transmission of overlapping frames. In order to address this problem, we proposed User dependent Multi-view video Streaming for Multi-users (UMSM). UMSM possesses two characteristics: overlapping frames required by multiple users are transmitted only once by multicast while un-overlapping frames required by each user are transmitted by unicast and offset of video requests between multiple users is aligned so as to maximize the area of overlapping frames. UMSM achieves a low transmission bitrate of multi-view video for multiple users by means of combining these characteristics. To investigate the effect of user's view switching on traffic reduction, we design three types of view-switching models, namely, scanning model, random model, and watching model. Simulation results obtained using benchmark test sequences provided by MERL reveal that UMSM decreases the transmission bitrate by 80.6% on average for five users, as compared to MVC, when users gaze a viewpoint for a while after switching views in order to search an attractive viewpoint.

Multi-view video consists of multiple video sequences captured simultaneously from different angles by multiple closely spaced cameras. It enables the users to freely change their viewpoints by playing different video sequences. Transmission of multi-view video requires more bandwidth than conventional multimedia. To reduce the bandwidth, UDMVT (User Dependent Multi-view Video Transmission) based on MVC (Multi-view Video Coding) has been proposed for single user. In UDMVT, for multiple users the same frames are encoded into different versions for each user, which increases the redundant transmission. For this problem, this paper proposes UMSM (User dependent Multi-view video Streaming for Multi-users). UMSM possesses two characteristics. The first characteristic is that the overlapped frames that are required by multiple users are transmitted only once using the multicast to avoid unnecessary duplication of transmission. The second characteristic is that a time lag of the video request by multiple users is adjusted to coincide with the next request. Simulation results using benchmark test sequences provided by MERL show that UMSM decreases the transmission bit-rate 55.3% on average for 5 users watching the same multi-view video as compared with UDMVT.

Multi-view video consists of multiple video sequences captured simultaneously by multiple closely spaced cameras from different angles. It allows the users to change their viewpoints by playing different video sequence. However, the transmission of multi-view video is much larger than conventional media and increases with increasing number of views, which brings much more increment in the bandwidth requirement. In our previous works, we have proposed a live-encoding scheme call UDMVT [1][2] to reduce the transmission bitrate for multi-view video transmission. In UDMVT, the periodic feedback containing the position and switching speed of user is used to predict a triangle area. Frames in this triangle area are possible to be displayed in next period of time. Only these specific frames are transmitted instead of all the frames to reduce the transmission bitrate. Even though, UDMVT hasn't effectively solve the “out of triangle” problem in which the user may switch to the outside of the triangle during the playback in the triangle area. In this paper, we propose a layered scheme called Layered User dependent Multi-view video Streaming (LUMS) to solve the “out of triangle” problem for UDMVT. In LUMS each triangle is layered into several layers according to the minimum and maximum switching speed. The corresponding layers are transmitted according to the current switching speed each time received from feedback. If the “out of triangle” occurs, the higher layers will be retransmitted. All the frames in high layers are encoded into B-frame predicted from the basic and lower layers, so the retransmission bitrate is low. Performance evaluation proves that LUMS reduce transmission bitrate by 44.9% compared to UDMVT.

Multi-view video consists of multiple views which are taken by multiple cameras from different positions and angles. It allows the users to freely change their viewpoints. Typical applications of multi-view video include Free Viewpoint TV, remote medical surgery, wireless multimedia sensor networks and so on. However, the traffic of multi-view video is also several times larger than traditional multimedia, which brings much more increment in the bandwidth requirement. To overcome this problem, Simulcast and MVC (Multi-view Video Coding) can be used to decrease the traffic for multi-view video. However, although Simulcast removes redundant information within each view, it still contains a large amount of inter-view redundant information. MVC further removes the correlation between the views, but all the views should be encoded and transmitted. So the bit-rates are still too high in these schemes. In this paper, in order to reduce the bit-rate for multi-view video transmission, we analyze the user's motion which is classified into random access model and successive motion model. We find that in the successive motion model only part of frames are possible to be displayed in a period of time. According to the analytical result, we propose a user dependent system to decrease the bit-rate of multi-view video transmission. The proposed system only encodes and transmits those frames that are possible to be displayed according to the user's motion. Further, in order to support the proposed system, we also propose a prediction structure as a substitute of the prediction structure in MVC. Performance evaluation proves that the proposed system reduces the average bit-rate for the transmission of multi-view video and is to be more effective when the number of views is large.

The traffic of multi-view video is several times larger than traditional multimedia, which brings much more increment in the bandwidth requirement. To overcome this problem, Simulcast and MVC (Multi-view Video Coding) can be used to decrease the traffic for multi-view video. However, although Simulcast removes redundant information within each view, it still contains a large amount of inter-view redundant information. MVC further removes the correlation between the views, but all the views should be encoded and transmitted. So the bit-rates are still too high in these schemes. In this paper, in order to reduce the bit-rate for multi-view video transmission, we analyze the user's motion which is classified into random access model and successive motion model. We find that in the successive motion model only part of frames are possible to be displayed in a period of time. According to the analytical result, we propose a user dependent scheme to decrease the bit-rate of multi-view video transmission. The proposed scheme only encodes and transmits those frames that are possible to be displayed according to the user's motion. Further, in order to support the proposed scheme, we also propose a prediction structure as a substitute of the prediction structure in MVC. Performance evaluation proves that this proposed scheme reduces the average bit-rate for the transmission of multi-view video.