Visual Media Coding and Transmission

Context based Visual Media Content Adaptation 531 Table 11.7 Channel BLER for vehicular A environment. Reproduced by Permission of Ó2008 IEEE Eb/No 1/2 CC 1/3 CC 3 dB 0.92 0.78 4 dB 0.78 0.53 6 dB 0.31 0.13 8 dB 0.047 0.013 10 dB 0.0020 0.0010 12 dB 0.0010 0.000 channel state prediction (thus, actual instantaneous channel condition) in the expected distortion calculation. Results and Discussions The accuracy of the distortion model is evaluated by comparing the estimated performance and the actual video performance over a simulated UMTS vehicular A environment. 1/3 rate convolutional coding with a spreading factor of 32 is considered. The Singer and Kettle sequences are encoded at a frame rate of 30 fps. Experiments are carried out for a range of channel conditions, and the performance of the composite sequence of the two input sequences is shown in Figure 11.39. Each experiment is repeated 25 times to simulate the average effect of bursty channel errors on the performance. Initial test results demonstrate that the estimated peak signal-to-noise ratio (PSNR) values closely match the actual PSNR values. This accurate modeling of expected distortion during transmission can be used to allocate maximum protection to segments with higher distortion estimates. It also incorporates an importance Actual Estimated 32 Frame PSNR (dBs) 31 30 29 28 9 17 25 33 41 49 57 65 73 81 1 Frame No Figure 11.39 Estimated performance comparisons: Eb/No ¼ 10 dB, BLER ¼ 0.0003

532 Visual Media Coding and Transmission Figure 11.40 Subjective performance: (a) Eb/No ¼ 10 dB, BLER ¼ 0.0003; (b) Eb/No ¼ 8 dB, BLER 0.0130. Reproduced by Permission of Ó2007 IEEE level for certain regions of the video content, along with the distortion estimates for optimal rate allocations. Figure 11.40 shows the subjective quality of the output composite sequence. 11.6.8.2 Cropping of H.264 Encoded Video for User-centric Content Adaptation As shown in Table 11.2, one of the adaptation requirements for the virtual classroom application is the sequence-level cropping of the visually salient part(s) of video scenes exchanged between collaborators (for example between lecturers/presenters and remote students). Thus, the scope of this subsection is to describe an AE that carries out the sequence-level cropping-type recommendations specified in the adaptation decision to provide ROI-based content adaptation. An example of the resulting video sequence after performing the adaptation operation is illustrated in Figure 11.41. Sequence-level cropping-type content adaptation can be performed at two locations along the content delivery chain: (1) at an external Figure 11.41 (a) Original sequence; (b) ROI adapted sequence

Context based Visual Media Content Adaptation 533 adaptation engine located at an edge of the network; and (2) at the user terminal. If the second option is utilized, the entire content has to be delivered to the user terminal. However, it should be noted that some of the information traversed through the network may be discarded due to bandwidth limitations or other transmission-related reasons without presenting to the user. Therefore, such an adaptation option can be considered a waste of precious bandwidth resources. Figure 11.42 illustrates that up to four times more bandwidth is necessary at times when the adaptation is performed within the user terminal device. Therefore, it is more advantageous if the cropping-type adaptations are performed at a network node and/or gateway. The AE is designed to accept both scalable and non-scalable H.264/AVC-coded digital video streams. If the input video is an IROI scalable [111] and has restricted motion compensation, it is a matter of selecting a substream that describes the ROI. However, in this subsection we only concentrate on non-scalable video adaptation. The following subsections describe the AE needed to realize such adaptation, and present relevant simulation results on the reduction of the computation complexity incurred by the adaptation operation, which is performed through transcoding. Paris News PSNR (dB) 50 PSNR (dB) 50 45 45 40 CIF 40 CIF 35 Cropped (QCIF) 35 Cropped (QCIF) 30 30 25 500 1000 1500 2000 2500 25 500 1000 1500 2000 20 Bit Rate (kbps) 20 0 0 Bit Rate (kbps) Coastguard Flower 50 50 45 45 PSNR (dB) 40 40 PSNR (dB) 35 35 30 CIF 30 CIF Cropped (QCIF) Cropped (QCIF) 25 25 1000 2000 3000 4000 5000 6000 2000 4000 6000 8000 20 Bit Rate (kbps) 20 0 0 Bit Rate (kbps) Figure 11.42 Rate distortion comparison of the original resolution video and the adapted video

534 Visual Media Coding and Transmission The Content Adaptation Architecture The AE is an integral part of the content adaptation scenario shown in Figure 11.21. The service provider is informed of the user terminal capabilities, such as display and/or decoder capabilities and buffer size, during the initial negotiation phase. Unique to the adaptation scenario under consideration is the provision of a user-interaction service, through which it is assumed that the user can specify a selected ROI. This information can be sent in a feedback message to the service provider during the playback. Upon receiving the ROI feedback message, the service provider consults an ADE, which determines the nature of adaptation needed after processing a number of context descriptors. In addition to the user-driven ROI feedback signaling, the ROI information is used when it is necessary to adapt the content to meet usage environment constraints. For example, if there is a need to reduce the aspect ratio to fit the video into a smaller terminal display then a similar adaptation operation can be performed, taking into consideration the ROI information, which is extracted automatically. Similarly, if it is necessary to reduce the bit rate to address channel bandwidth constraints, the ADE may decide to utilize cropping rather than spatial scaling or quality scaling so that the ROI information can be presented at a better visual quality. These decisions are made after processing a number of context descriptors, which in turn describe the user-defined ROI and other constraints, such as QoS, terminal capabilities, access network capabilities, usage environment, DRM, and so on. In this subsection, an AE that carries out sequence-level cropping-type recommendations specified in the adaptation decision is described. This AE achieves interactive ROI-based user-centric video adaptation for H.264/AVC-coded video streams through transcoding, and maximally utilizes values of the syntax elements from the input bitstreams to minimize the computational complexity. Adaptation Engine (AE) The heart of the AE is essentially a transcoder, which performs sequence-level cropping of video frames of an H.264/AVC-encoded high-resolution video sequence, and produces an H.264/AVC-compatible output bitstream. The architecture of this transcoder is illustrated in Figure 11.43. The decoder decodes the incoming bitstream, and subsequently the reconstructed frame and parsed syntax elements are passed to the encoder through the Frame Adaptor (FA). The function of the FA is: . To crop the decoded frames as specified in the adaptation decision received from the ADE. . To select and organize relevant syntax elements of the MBs that fall within the specific cropped region. In order to simplify the AE, it is assumed that the cropping is performed along the MB boundaries, and it is the ADE’s responsibility to consider this constraint when user requests are processed. Under the above assumption, there is a one-to-one correspondence in terms of MB Figure 11.43 The AE. Reproduced by Permission of Ó2007 IEEE

Context based Visual Media Content Adaptation 535 Figure 11.44 One to one relationship between MBs in the original and the adapted frames. Reproduced by Permission of Ó2008 IEEE content. That is to say, for the ith MB (in raster order) in the adapted frame (MBi,adapted), there is a corresponding MB j in the decoded frame that has exactly the same luminance and chrominance composition, as illustrated (MBj,input) in Figure 11.44. Before encoding MBi,adapted, the encoder checks whether the coding mode of MBi,input and the associated syntax elements are valid for MBi,adapted. Possible tests evaluate the validity of motion vectors, intra-prediction mode, SKIP mode motion vector predictors, and so on. In this chapter, the discussions are limited to the MB SKIP mode [110] in predictive-coded pictures. Other coding modes, including block-level SKIP mode, are determined using the rate distortion optimization for performance evaluation purposes. SKIP-mode MBs are reconstructed using motion compensation at the decoder. Necessary motion vector is estimated at the decoder without any supporting information from the encoder. The estimated motion vector is known as the SKIP-mode motion vector predictor (MVP). Motion vectors of the surrounding MBs are involved in the estimation process [110]. MVP of a given MB may remain identical when the MB is decoded and re-encoded during transcoding, depending on the availability of surrounding MBs and their motion vectors. If the encoder and decoder come up with the same MVP, MB SKIP mode is assumed to be the best coding mode to re-encode the MB. Based on this assumption, the algorithm described in Figure 11.45 is executed. Inputs to this process are the adapted frame and syntax elements of corresponding MBs, which are gathered while the original sequence is being parsed. For each MB, a SKIP- mode MVP is estimated at the encoder functional units (MVPSKIP,evaluated) if the corresponding MB in the original bitstream (MBinput) is coded in MB SKIP mode. If MVPSKIP,evaluated is identical to that calculated by the decoder when the original bitstream is decoded (MVPSKIP, input), MB SKIP mode is considered to be the optimum coding mode for encoding the MB in the adapted frame. Otherwise, the RD optimization algorithm is invoked to find a suitable coding mode.

536 Visual Media Coding and Transmission Figure 11.45 The algorithm to determine the coding mode of an MB in the cropped frame. Reproduced by Permission of Ó2009 IEEE Experimentation Setup In order to evaluate the credibility of the above algorithm, a transcoder is used, which is based on the Joint Scalable Video Model (JSVM) encoder and decoder version 7.13. The input bitstream is H.264/AVC extended profile compatible and the output bitstream is H.264/AVC baseline profile compatible. Experimental results are presented for CIF test video sequences available in the public domain and experimental conditions are described in detail in Table 11.8. Once the ROI window is selected, it is assumed to stay unchanged throughout the length of the

Context based Visual Media Content Adaptation 537 Table 11.8 Experimental conditions for the selected video test sequences. Reproduced by Permission of Ó2011 IEEE Test Format Number of Origin of the Format of the MB Count within sequence frames ROI (in Pixels) ROI the ROI Paris (A) CIF 1060 (0, 0) QCIF 104 940 Paris (B) CIF 1060 (176, 16) QCIF 104 940 News (A) CIF (0, 64) QCIF News (B) CIF 290 (176, 64) QCIF 28 710 Coastguard CIF 290 (48, 64) QCIF 28 710 Flower CIF 290 (0, 64) QCIF 28 710 240 23 760 sequence for simplicity. PSNR is used for objective quality evaluation, and is computed only over the selected ROI. Results and Discussions Figure 11.46 compares the adapted quarter common intermediate format (QCIF) versions based on the cropping method, as described in Table 11.8, with the original CIF formatted and scaled QCIF versions. When the frame is scaled to fit to the display size, visually important details such as facial expressions become less visible, and thus the quality of the visual experience becomes minimal. However, with the adaptation mechanism under consideration, such details are preserved, as demonstrated in Figure 11.46. This is because the resolution scaling operation is not performed over the ROIs at all. Therefore, this technique is envisaged to produce better visual experiences for users of this AE technology. The first set of analytical experiments is carried out to investigate the utilization of MB SKIP mode for coding MBs within the selected ROIs. SKIP mode usage statistics within the cropped ROIs are presented in Table 11.9. In general, at lower bit rates (i.e. when the quantization parameter (QP) is higher) SKIP mode is used more often. When the QP is high, the distortion cost becomes more significant than the bit-rate cost. Therefore, the RD cost of SKIP mode (MB SKIP cost), which is independent of the QP, becomes greater than that of the case in which the residues are quantized with lower QPs. However, when the QP becomes higher, the distortion cost of other modes increases. As a result, the bit rate becomes the dominant factor in determining the coding mode. Since the bit budget of a SKIP-mode MB is very low, RD optimization favors this mode. Therefore, when the QP is higher, a higher number of MBs are coded in the MB SKIP mode. Statistics presented in Table 11.9 confirm the validity of the above argument. Table 11.9 also shows that a significant percentage of MBs are coded in SKIP mode. Therefore, there is a considerable potential for reducing the coding complexity significantly just by utilizing the MB SKIP-mode information derived from the input bitstream wherever possible. When a given MB within the ROI is considered before and after cropping, its surrounding areas may not be identical. Therefore, there is a substantial possibility that the decoder and encoder functional blocks of the transcoder come up with different SKIP-mode MVPs for a given MB. If this is the case, it is impossible to rely completely on the information available in the input bitstream to decide the transcoding mode without considering the RD cost. Table 11.10 summarizes the probability that MVPSKIP,input and MVPSKIP,evaluated are different. The results clearly show that the probability is higher when motion and texture in the scene are

538 Visual Media Coding and Transmission Figure 11.46 ROI selections described in Table 11.8

Context based Visual Media Content Adaptation 539 Table 11.9 MB SKIP mode usage statistics within the cropped areas in the original H.264/AVC coded bitstreams. Reproduced by Permission of Ó2012 IEEE QP Usage of MB SKIP mode for coding the ROI in the original bits streams for each video test sequence (%) Paris (A) Paris (B) News (A) News (B) Coastguard Flower 10 10.92 11.13 38.28 35.48 0.00 17.65 15 42.16 50.72 45.61 46.10 0.06 29.77 20 61.86 65.55 59.43 53.08 5.28 36.39 30 76.04 77.97 71.69 68.61 23.55 50.76 40 87.88 87.89 82.68 80.65 57.82 64.85 complicated. However, the probability stays relatively smaller for stationary sequences and sequences with regular motion (e.g. the Flower sequence). The results also indicate that this probability becomes smaller for low-bit-rate encoding. In summary, there is a considerable probability that MB SKIP-mode MVPs remain identical when decoding and re-encoding during content adaptation through transcoding. Figure 11.47 compares the rate distortion characteristics when the JSVM RD optimization is used to estimate the coding modes of all MBs and when the technique described in this subsection is used. Input sequences are generated by encoding the raw video sequences at five different quantization parameter settings. Each of the resulting bitstreams is then transcoded using different quantization parameter settings to obtain the RD performance for transcoding each input bitstream. The objective quality is measured by comparing the original video sequence with the one generated by decoding the transcoded bitstreams. The objective quality results illustrated in Figure 11.47 clearly show that there is not any noticeable RD penalty associated with the described technique. Table 11.11 also compares the average per frame coding time. These coding times are indicative, since the JSVM software is not optimized, and are obtained by measuring the time taken when transcoding longer versions of the video sequences. Each of these longer versions is at least 3000 frames long, and obtained by cascading the original video. Considering the above results, it can be concluded that up to 34% coding complexity reduction can be achieved with the transcoding technique discussed here. In summary, IROI adaptation of H.264/AVC-coded video applicable to a user-centric video adaptation scheme crops a user-specified attention area (i.e. ROI), enabling the user to actively Table 11.10 Probability of estimating different SKIP mode MVPs for a given MB with the decoder and encoder functional units of the transcoder. Reproduced by Permission of Ó2013 IEEE QP The probability that MVPSKIP,input and MVPSKIP,evaluated are different Paris (A) Paris (B) News (A) News (B) Coastguard Flower 10 0.014 0.018 0.006 0.002 0.722 0.461 15 0.007 0.006 0.026 0.025 0.641 0.387 20 0.006 0.007 0.033 0.045 0.446 0.328 30 0.011 0.015 0.048 0.098 0.548 0.115 40 0.027 0.023 0.036 0.052 0.105

540 Visual Media Coding and Transmission 50 Paris (A) News (A) 45 PSNR (dB) 40 PSNR (dB) 50 35 45 30 500 1000 1500 40 500 1000 1500 25 Bit Rate (kbps) 35 Bit Rate (kbps) 30 0 Coastguard 25 Flower PSNR (dB) 49 PSNR (dB) 0 44 39 1000 2000 3000 49 500 1000 1500 34 44 Bit Rate (kbps) 29 39 24 34 29 0 24 0 Bit Rate (kbps) Input QP = 10; Transcode Method = RDOPT Input QP = 10; Transcode Method = Proposed Input QP = 15; Transcode Method = RDOPT Input QP = 15; Transcode Method = Proposed Input QP = 20; Transcode Method = RDOPT Input QP = 20; Transcode Method = Proposed Input QP = 30; Transcode Method = RDOPT Input QP = 30; Transcode Method = Proposed Input QP = 40; Transcode Method = RDOPT Input QP = 40; Transcode Method = Proposed Figure 11.47 Rate distortion comparison for input bitstreams coded with various QPs when JSVM RD optimization (RDOPT) and the adaptation technique (Adapted) are used, respectively. Reproduced by Permission Ó2010 IEEE select an attention area. The adapted video sequence is also an H.264/AVC-compatible bitstream. Subjective quality test results demonstrate that this scheme is capable of producing the targeted ROIs, which in turn are envisaged as providing better user experience by preserving the details of the attention area. 11.6.8.3 Adapting Scalability Extension of H.264/AVC Compatible Video Even though the applicability of transcoding cannot be played down as an option for some of the operations, such as cropping, summarization, and error-resilience, the SVC approach is still

Table 11.11 Reduction in coding complexity with the adaptation technique. Reproduced by Permission of Ó2014 IEEE Quantizer Test sequence Paris (A) News (A) Coastguard Flower Average Improvement Average Improvement Average Improvement Average Improvement coding time of coding coding time of coding coding time of coding coding time of coding per frame (s) time (%) per frame (s) time (%) per frame (s) time (%) per frame (s) time (%) RD Opti- RD Opti- RD Opti- RD Opti- Adapted mization Adapted mization Adapted mization Adapted mization 10 0.447 0.469 4.63 0.428 0.503 15.07 0.53 0.53 -0.65 0.500 0.517 3.23 0.51 0.51 0.00 0.454 0.483 6.03 15 0.374 0.450 16.98 0.400 0.476 15.94 0.50 0.50 0.00 0.442 0.483 8.62 0.46 0.47 2.22 0.371 0.454 18.35 20 0.335 0.441 23.98 0.359 0.459 21.81 0.43 0.48 10.07 0.333 0.429 22.33 30 0.297 0.435 31.67 0.331 0.459 27.82 40 0.279 0.427 34.66 0.300 0.438 31.50

542 Visual Media Coding and Transmission BSD Adaptation Adapted Decision BSD Description Adaptation Demultiplexer Input DI Multiplexer Adapted DI Packet Filter H.264/SVC Adapted Video H.264/SVC Video Figure 11.48 The scalability extension of H.264/AVC compatible (H.264/SVC) video AE feasible for most of the adaptation operations required in the virtual classroom application. The advantage of the SVC-based adaptation is that it is less computationally intensive than the transcoding-based adaptations. The architecture of the SVC-based AE for the virtual classroom application is illustrated in Figure 11.48. The demultiplexer separates the BSD and the SVC extension of H.264/AVC-encoded video from the input DI. The description adapter reformats the BSD according to the description provided in the adaptation decision. While performing the BSD adaptation, the description adapter drives the packet filter to discard or truncate the corresponding H.264/AVC SVC video data packets in order to perform the video adaptation. This adapted bitstream is then combined with the adapted BSD to form the output DI at the multiplexer. In this subsection, the feasibility of using scalability options in SVC extension of H.264/ AVC to achieve some of the required adaptation operations in the virtual classroom scenario is evaluated and discussed using publicly-available video sequences [130]. In these evaluations, the JSVM 7.13 and a H.264/AVC transcoder based on the same JSVM version are used as the software platform. Figure 11.49 compares the objective quality of bit-rate adaptation using transcoding and fine-grain fidelity scalability for the Crowd Run video test sequence. In the tests, four scalable and one non-scalable source bitstreams are used as the original video sequences. Scalable bitstreams are obtained by varying the number of temporal and spatial scalability levels. Highest spatial and temporal resolutions are 1280 Â 704 pixels (cropped from the original 1280 Â 720 resolution to meet the frame size constraints in JSVM soft- ware [131]) and 50 Hz (progressive), respectively. Lower resolutions are dyadic subdivisions of the maximum resolution. Adapted bitstreams also have the same temporal and spatial resolutions. For comparison purposes, the rate distortion performance that can be achieved by directly encoding the raw video sequence is also shown in the figure. According to Figure 11.49, it is clear that the fidelity scalability offers the flexibility of adjusting the bit rate over a large range. Since fine granular scalability layers can be truncated, a large number of bit-rate adaptation points can be achieved with this technology, allowing the adaptation platform to react to the dynamics of the available network resources more precisely. These experimental results also suggest that the objective quality can be further improved by selecting the most appropriate number of spatial and temporal scalability layers for a given situation. For example, when there is no demand for resolutions below 640 Â 352 pixels, the

Context based Visual Media Content Adaptation 543 39 37 35 PSNR (dB) 33 Transcoded from AVC 31 Scaled (T=3, S=3, Q=4) Scaled (T=4, S=3, Q=4) 29 Scaled (T=3, S=2, Q=4) Scaled (T=4, S=2, Q=4) AVC Encoded 27 25 20 000 30 000 40 000 50 000 60 000 10 000 Bit rate (kbps) Figure 11.49 Objective quality comparison of bit rate adaptation using fine grain fidelity scalability: T, number of temporal scalability levels; S, number of spatial scalability levels; Q, number of fine grain fidelity scalability levels. Reproduced by Permission of Ó2007 IEEE encoder can omit the 320 Â 176 pixel resolution. Limiting the number of resolution layers to two (i.e. S ¼ 2) can achieve an additional objective quality gain of over 0.5 dB. In order to make such a decision dynamically, though, it is necessary to have information regarding the required number of adaptation levels at a given time. Since ADE tracks the dynamics of the prevailing context, a feedback from the ADE can be used to decide the level of adaptation. The number of temporal scalability levels can be increased only at the expense of delay. However, unlike with spatial scalability levels, increasing the number of temporal levels in turn increases the compression efficiency, as illustrated in Figure 11.49. The reason behind the improved compression efficiency is the use of more hierarchically-predicted bidirectional pictures (B-pictures) to achieve more scalability layers [108]. Consequently, the allowed maximum delay becomes the major factor in selecting the number of temporal scalability layers. Figure 11.50 shows the rate distortion characteristics of adapting the abovementioned source bitstreams to achieve 640 Â 352 pixel and 25 Hz temporal resolution. Source bitstreams used for this adaptation are the same as those used in the adaptations described in Figure 11.49. In this case, when S ¼ 2, the SVC-based adaptation performs better than transcoding. This is because when S ¼ 2, 640 Â 352 pixel resolution is coded as the base layer, which is H.264/AVC compatible. Both Figure 11.49 and Figure 11.50 demonstrate that the scalability options are only available at the cost of coding efficiency. However, the computational complexity associated with the adaptation operation, which can be achieved by a simple packet selection, is negligible compared to the transcoding operations that achieve the same adaptation from a non-scalable bitstream. Therefore, it can be concluded that scalability is a more effective way of assisting spatial and temporal adaptation requirements in the virtual classroom application. Further

544 Visual Media Coding and Transmission PSNR (dB) 45 43 41 Tanscoded from AVC 39 Scaled (T=4, S=3, Q=4) 37 Scaled (T=4, S=2, Q=4) 35 Scaled (T=3, S=3, Q=4) 33 Scaled (T=3, S=2, Q=4) 31 29 2000 4000 6000 8000 10 000 12 000 14 000 16 000 27 Bit rate (kbps) 25 0 Figure 11.50 Objective quality comparison of adaptation to 640 Â 352 spatial and 25 Hz using temporal, spatial, and fine grain fidelity scalability investigations are needed in order to enable the IROI scalability in SVC extension of H.264/ AVC to achieve attention area-based content adaptation for user-centric media processing. 11.6.9 Interfaces between Modules of the Content Adaptation Platform This section provides insights into the interfaces required between the modules of the adaptation platform and the sequence of events that take place while performing DRM-based adaptation for a particular user or group of users. Figure 11.51 represents the functional Figure 11.51 Functional architecture of the platform for context aware and DRM enabled multimedia content adaptation. Reproduced by Permission of Ó2008 IIMC Ltd and ICT Mobile Summit

Context based Visual Media Content Adaptation 545 Context Adaptation Message Semantic Meaning Providers Decision REQUEST ADE invokes the contextual information to be Engine STATUS extracted by the CxPs (CxP) (ADE) METADATA The contextual information is sent by the respective CxPs REQUEST STATUS METADATA UPDATED UPDATED ADE sends an acknowledgement message saying that the contextual information has been received correctly Figure 11.52 Message exchange between CxP and ADE during service negotiation architecture of this platform, in which the interfaces between modules are illustrated. For this distributed environment, the exchange of messages is addressed using SOAP, a simple and extensible Web Service protocol. 11.6.9.1 ADE–CxP Interface To obtain low-level context information, the ADE can either query CxPs or listen for events sent by CxPs, depending on the service status. During service negotiation, the ADE queries the CxPs as shown Figure 11.52. The received contextual information is formatted in standard MPEG-21 DIA UED descriptors, and registered in the ontology model. After the service is launched, however, the CxPs work in a ‘‘push’’ model, notifying the ADE when new context is available via the context update message illustrated in Figure 11.53. In this way, the ADE is enabled to react to any significant changes in context and adjust the adaptation parameters accordingly, in order to maximize user satisfaction under new usage environment conditions. 11.6.9.2 AE–ADE Interface While designing the interface between the AES and ADE, factors such as ability to have multiple AESs operating within the system and their ability to join, leave, and migrate seamlessly are considered. In order to provide the aforementioned flexibility, a dedicated system initialization phase, which is initiated by the AES, is introduced. The sequence of Context Adaptation Message Semantic Meaning Providers Decision Engine CONTEXT The CxP sends the respective new (CxP) EVENT contextual information to the ADE (ADE) UPDATE ADE sends an acknowledgement message CONTEXT EVENT UPDATED saying that the contextual information has UPDATE been received correctly UPDATED Figure 11.53 Message exchange between CxP and ADE after the service is launched

546 Visual Media Coding and Transmission Figure 11.54 Message exchange between AE and ADE during service negotiation messages exchanged between the ADE and AE during system initialization is illustrated in Figure 11.54. The parameters required at this stage are mostly related to the AES’s capabilities. Therefore, the AES informs adaptation capabilities and necessary metadata related to those capabilities, for example maximum and minimum bit rates, maximum spatial resolution, and so on, along with the registration request message. In order to conclude the registration on the ADE database, the AES should also inform the ADE of its IP address and the service identifier. Once the registration phase is completed, the AES is ready to perform adaptation operations when the ADE requests them. After making an adaptation decision, the ADE invokes the service initialization operation shown in Figure 11.55. During the service initialization, the ADE informs AESs to invoke the selected adaptation operations on the original DI and forward the adapted DI to the user. This request also contains the related adaptation parameters, including the source digital item (DI), desired adaptation operations, and associated metadata. Figure 11.55 Message exchange between AE and ADE during service initialization

Context based Visual Media Content Adaptation 547 Message Semantic Meaning AUTHORIZATION This message is sent by the ADE to request REQUEST information about the adaptation authorization Adaptation Adaptation associated with a content and User Decision Authorizer Engine (AA) (ADE) AUTHORIZATION AUTHORIZATION The AA provides the authorization information – REQUEST AUTHORIZATION RESPONSE a list of permitted adaptation operations and RESPONSE RECEIVED associated constraints RECEIVED An acknowledgement message is sent from the ADE Figure 11.56 Message exchange between AA and ADE during service negotiation 11.6.9.3 ADE–AA Interface The ADE’s content adaptation decision is preceded by an authorization request, which identifies the User that consumes the adapted content and the multimedia resource that is going to be adapted, by its DI identifier. The sequence of messages exchanged between the AA and ADE to obtain permitted adaptation operations is illustrated in Figure 11.56. Once the AA has received the authorization request from the ADE, it responds with all the adaptation-related information contained in the license associated with the referred multimedia resource and User. This information includes the permitted adaptation operations, as well as the adaptation constraints associated with those operations. Both the permitted adaptation operations and related constraints are expressed in a format compatible with MPEG-21 DIA. 11.6.9.4 An Example Use Case This subsection details the sequence of messages exchange between modules of the content adaptation platform, based on an example use case. In the selected use case, a student wishes to attend a virtual classroom session using their PDA over a 3G mobile network. The minimum bandwidth required to receive the best-quality multimedia virtual classroom material is 1 Mbps and the associated video is of VGA (640  480 pixels) resolution. The sequence of messages transferred between modules to address content adaptation needs in the aforementioned use case is summarized in the sequence chart shown in Figure 11.57. During the system initialization phase, newly-commissioned AESs inform the ADE of their capabilities and required parameters. An example of the registration request message is shown in Table 11.12.

548 Visual Media Coding and Transmission Figure 11.57 Sequence chart of messages exchanged between each module. Reproduced by Permission of Ó2008 IIMC Ltd and ICT Mobile Summit In the selected use case, the virtual classroom administrator notifies the ADE when the student joins the virtual classroom session. Before taking the adaptation decision, the ADE needs to detect and extract the context information about the terminal capabilities, network conditions, and the surrounding environment. Therefore, the ADE queries the CxPs for context information during the service negotiation phase. Assuming that the CxP responsible for providing terminal capabilities and environment conditions is the user terminal, and the one Table 11.12 Contents of the AES registration request message Parameters Metadata Time Tue, 29 Jan 2008 15:31:55 Multimedia content identifier MPEG 21 DI IP address 202.145.2.98 Service identifier Service ID of the AES Capabilities Scalable Spatial resolution scaling Scalable Temporal resolution scaling Scalable Bit rate scaling Maximum spatial resolution ¼ 720 Â 560 pixels ROI cropping Minimum spatial resolution ¼ 16 Â 16 pixels Maximum temporal resolution ¼ 50 fps Present subtitles (audio to text Minimum temporal resolution ¼ 0 transmoding) Maximum cropping window size ¼ 720 Â 560 pixels Minimum cropping window size ¼ 16 Â 16 pixels Transmoding languages ¼ English, Spanish, Portuguese Display font sizes ¼ small, medium, large Display position ¼ adaptive, coordinates on the display

Context based Visual Media Content Adaptation 549 Table 11.13 Context information received from the terminal Parameters Metadata Time Tue, 29 Jan 2008 15:32:58 Context provider identifier Terminal ID Terminal capabilities Display size (height  width) ¼ QCIF (176  144 pixels) Terminal capabilities Maximum frame rate ¼ 25 fps Terminal capabilities BatteryTimeRemaining ¼ 15 minutes Terminal capabilities Codecs supported ¼ MPEG 4, H.264/AVC responsible for providing the network condition is the network service provider, the contents of context information messages received are listed in Tables 11.13 and 11.14. Moreover, the ADE requires information on the permitted adaptation operations on virtual classroom materials for the particular user and hence it also queries the AA. The contents of the adaptation authorization message received from the AA in response to the ADE’s query are listed in Table 11.15, and the MPEG-21 DIA-formatted message is shown in Table 11.16. Based on the context information, it is clear that the particular user under consideration is using a terminal device with a small display (i.e. the PDA). As a result, the ADE decides to downscale the video resolution to 160  120 pixels, which is well within the authorized minimum resolution specified by the AA. Meanwhile, the ADE realizes that the remaining battery power level of the terminal (i.e. the PDA) is not adequate for presenting the entire lecture session at the highest possible fidelity. Moreover, the available network bandwidth is much less than the data rate required to deliver the audiovisual material at its best quality. Responding to these constraints, the ADE decides to decrease the temporal resolution and the Table 11.14 Context information received from the network service provider Parameters Metadata Time Tue, 29 Jan 2008 15:32:58 Context provider identifier Network service provider’s ID Network conditions Available bandwidth ¼ 128 kbps Table 11.15 Adaptation authorization response Metadata Parameters Tue, 29 Jan 2008 15:32:58 MPEG 21 DI Time MPEG 21 KeyHolder Multimedia content identifier Type of user identifier Possible Adaptation Operations Minimum Spatial Resolution ¼ 150  100 pixels Minimum Temporal Resolution ¼ 10 fps Spatial resolution scaling Minimum nominal bit rate ¼ 30 kbps Temporal resolution scaling Bit rate scaling

550 Visual Media Coding and Transmission Table 11.16 An extract from an MPEG 21 DIA authorization response. Reproduced by Permission of Ó2007 IEEE <dia:permittedDiaChanges> <dia:ConversionDescription xsi:type \"dia:ConversionUriType\"> <- - Adaptation of TemporalResolution - -> <dia:ConversionActUri uri \"urn:visnet:TemporalResolutionScaling\"/> </dia:ConversionDescription> <dia:ConversionDescription xsi:type \"dia:ConversionUriType\"> <- -Adaptation of the SpatialResolution - -> <dia:ConversionActUri uri \"urn:visnet:SpatialResolutionScaling\"/> </dia:ConversionDescription> <dia:ConversionDescription xsi:type \"dia:ConversionUriType\"> <- - Adaptation of the Bit Rate - -> <dia:ConversionActUri uri \"urn:visnet:BitRateTranscoding\"/> </dia:ConversionDescription> <- - Further ConversionDescription go here - -> </dia:permittedDiaChanges> <- - Following constraints apply whether or not the image is adapted - -> <dia:changeConstraint> <dia:constraint> <dia:AdaptationUnitConstraints> <dia:LimitConstraint> <- - Minimum limits for the TemporalResolution - -> <dia:Argument xsi:type \"dia:SemanticalRefType\" semantics \"urn:mpeg:mpeg21:2003:01-DIA-MediaInformationCS- NS:20\"/> <- - 20 refers to TemporalResolution - -> <dia:Argument xsi:type \"dia:ConstantDataType\"> <dia:Constant xsi:type \"dia:IntegerType\"> <dia:Value>10</dia:Value> </dia:Constant> </dia:Argument> <dia:Operation operator \"urn:mpeg:mpeg21:2003:01-DIA-StackFunctionOperatorCS-NS:13\"/> <- - 13 refers to the operator \">\" - -> </dia:LimitConstraint> <– Minimum limit for the SpatialResolution –> <dia:LimitConstraint> <- - width must be more than 150 - -> <dia:Argument xsi:type \"dia:SemanticalRefType\" semantics \"urn:mpeg:mpeg21:2003:01-DIA-MediaInformationCS-NS:17\"/> <- - 17 refers to the width - -> <dia:Argument xsi:type \"dia:ConstantDataType\"> <dia:Constant xsi:type \"dia:IntegerType\">

Context based Visual Media Content Adaptation 551 Table 11.16 (Continued) <dia:Value>150</dia:Value> </dia:Constant> </dia:Argument> <dia:Operation operator \"urn:mpeg:mpeg21:2003:01-DIA-StackFunctionOperatorCS-NS:13\"/> <- - 13 refers to the operator \">\" - -> </dia:LimitConstraint> <dia:LimitConstraint> <- - height must be more than 100 - -> <dia:Argument xsi:type \"dia:SemanticalRefType\" semantics \"urn:mpeg:mpeg21:2003:01-DIA-MediaInformationCS- NS:18\"/> <- - 18 refers to the height - -> <dia:Argument xsi:type \"dia:ConstantDataType\"> <dia:Constant xsi:type \"dia:IntegerType\"> <dia:Value>100</dia:Value> </dia:Constant> </dia:Argument> <dia:Operation operator \"urn:mpeg:mpeg21:2003:01-DIA-StackFunctionOperatorCS-NS:13\"/> <- - 13 refers to the operator \">\" - -> </dia:LimitConstraint> <- - END (minimum limit for the SpatialResolution) - -> <dia:LimitConstraint> <- - Bit Rate minimum limit - -> <dia:Argument xsi:type \"dia:SemanticalRefType\" semantics \"urn:mpeg:mpeg21:2003:01-DIA-MediaInformationCS- NS:7\"/> <- - 7 refers to Nominal Bit Rate- -> <dia:Argument xsi:type \"dia:ConstantDataType\"> <dia:Constant xsi:type \"dia:IntegerType\"> <dia:Value>30000</dia:Value> </dia:Constant> </dia:Argument> <dia:Operation operator \"urn:mpeg:mpeg21:2003:01-DIA-StackFunctionOperatorCS-NS:13\"/> <- - 13 refers to the operator \">\" - -> </dia:LimitConstraint> <- - Further constraints go here - -> </dia:AdaptationUnitConstraints/> </dia:constraint> </dia:changeConstraint>

552 Visual Media Coding and Transmission Table 11.17 Adaptation parameters Parameters Metadata Time Tue, 29 Jan 2008 15:32:58 Multimedia content identifier MPEG 21 DI Display size QCIF (176  144 pixels) Required Adaptations Spatial resolution before adaptation ¼ VGA (640  480) Operations Spatial resolution after adaptation ¼ Quarter QVGA (160  120) Spatial resolution scaling Temporal resolution before adaptation ¼ 25 fps Temporal resolution after adaptation ¼ 12.5 fps Temporal resolution scaling Bit rate before adaptation ¼ 1 Mbps Bit rate after adaptation ¼ 128 kbps Bit rate scaling fidelity of the visual content in order to minimize the processor utilization and required bandwidth. Once the ADE takes the adaptation decision, it contacts appropriate AESs to initiate the adaptation operations during the service initialization phase. A list of adaptation parameters conveyed to the AESs to address the adaptation requirements of the selected use case is shown in Table 11.17. During the in-service phase, the ADE keeps on monitoring the dynamics of the context through the context update information received from the CxPs. If it detects any significant change that affects the user’s satisfaction it reviews the adaptation options. If there is any better set of adaptation operations, the ADE reconfigures the AESs accordingly through another service initiation operation. 11.7 Conclusions This chapter has presented comprehensive state-of-the-art discussions on various technologies and trends that are being studied and used to develop context-aware content adaptation systems. Apart from a few references included and standardization-related efforts described, it has been noted that there is not much work published on adaptation decision implementations. Nevertheless, a thorough analysis of the different elements and factors that drive the adaptation decision process has been made, as well as an investigation of the current trends in how to use those elements, notably the low-level contextual information. There is a clear trend in merging the high-level semantic world with the adaptation decision process. Major research efforts point to the use of ontologies and Semantic Web technologies to accomplish that goal. The aim is to be able to use automatically-generated low-level contextual information to infer higher- level concepts that better describe real-world situations. Subsequently, several aspects related to gathering requirements in terms of contextual information, forms of representing and exchanging it, and identification of entities/mechanisms involved in the processing of contextual information, together with the management of digital rights and provision for authorization to the adaptation of media contents, have been presented with a view to a defined application scenario. A number of resource adaptation algorithms have

Context based Visual Media Content Adaptation 553 been described, which would operate in harmony with the decision and authorization operations so as to maximize users’ experiences. It has been shown that adaptation decision-taking operations can greatly benefit from the use of ontologies for deciding the best available resource adaptation method for the context of the usage environment. In line with the aforementioned discussions, the concepts and architecture of a scalable platform for context-aware content adaptation have also been described in this chapter; these are especially suited for virtual collaboration applications. The discussions have particularly focused on providing a consolidated view on the current research panorama in the area of context-aware adaptation and the identification of directions to follow. In particular, a number of relevant topics have been presented to identify real-world situations that would benefit from the use of context and of adaptation operations; to identify relevant contextual information; to formulate approaches to specify context profiles; to address DRM during adaptation; and to identify functionalities to build generic context-aware content adaptation systems. Further- more, the interfaces between the modules of the content adaptation platform and the sequence of events that take place while performing DRM-based adaptation for a particular user or group of users in specific situations have also been presented. The innovative character of the adaptation platform under consideration has been brought out by addressing different aspects concerning the delivery of networked multimedia content simultaneously. This has been highlighted in the platform by combining the use of ontologies and low-level context to drive the adaptation decision process, verifying and enforcing usage rights within the adaptation operations, incorporating multi-faceted AEs, and being able to deliver on-the-fly, on-demand, different adaptation operations that suit different dynamic requirements. It is envisaged that the definition of such a modular architecture with well- defined and standards-based interfaces will greatly contribute to the interoperability and scalability of future content delivery systems. References [1] B. Schilit, N. Adams, and R. Want, ‘‘Context aware computing applications,’’ Proc. IEEE Workshop on Mobile Computing Systems and Applications, Santa Cruz, USA, pp. 85 90, Dec. 1994. [2] S. Pokraev, P.D. Costa, G.P. Filho, and M. Zuidweg, ‘‘Context aware services: state of the art,’’ Telematica Instituut, Tech. Rep. TI/RS/2003/137, 2003. [3] A.K. Dey, ‘‘Providing architectural support for building context aware applications,’’ PhD Thesis, College of Computing, Georgia Institute of Technology, Atlanta, GA, 2000. [4] D.Rios, P.D. Costa, G. Guizzardi, L.F. Pires, J. Gonc¸alves, P. Filho, and M. van Sinderen, ‘‘Using ontologies for modeling context aware services platforms,’’ Proc. Workshop on Ontologies to Complement Software Architectures, Anaheim, CA, Oct. 2003. [5] ‘‘Web Ontology Language (OWL): overview,’’ W3C Recommendation, Feb. 2004. http://www.w3.org/TR/owl features/. [6] P.T.E. Yeow, ‘‘Context mediation: ontology modeling using Web Ontology Language (OWL),’’ Composite Information Systems Laboratory (CISL), Massachusetts Institute of Technology, Cambridge, MA, Working Paper CISL# 2004 11, Nov. 2004. [7] D. Preuveneers, J. van der Bergh, D. Wagelaar, A. Georges, P. Rigole, T. Clerckx, et al., ‘‘Towards an extensible context ontology for ambient intelligence,’’ Proc. Second European Symposium on Ambient Intelligence, Eindhoven, the Netherlands, Vol. 3295 of LNCS, pp. 148 159, Nov. 2004. [8] J. Z. Sun and J. Sauvola, ‘‘Towards a conceptual model for context aware adaptive services,’’ Proc. Fourth International Conference on Parallel and Distributed Computing, Applications and Technologies, Sichuan, China, pp. 90 94, Aug. 2003.

554 Visual Media Coding and Transmission [9] G.D. Abowd, M.R. Ebling, H.W. Gellersen, G. Hunt, and H. Lei, ‘‘Context aware pervasive computing,’’ IEEE Wireless Commun. J, Vol. 9, No. 5, pp. 8 9, Oct. 2002. [10] T. Strang, C. Linnhoff Popien, and K. Frank, ‘‘CoOL: a context ontology language to enable contextual interoperability,’’ Proc. 4th IFIP WG 6.1 International Conference on Distributed Applications and Interop erable Systems, Vol. 2893 of LNCS, pp. 236 247, 2003. [11] M. Keidi and A. Kemper, ‘‘Towards context aware adaptable Web services,’’ Proc. 13th World Wide Web conference, NY, pp. 55 65, 2004. [12] T. Chaari, F. Laforest, and A. Celentano, ‘‘Adaptation in context aware pervasive information systems: the SECAS project,’’ Int. J. of Pervasive Computing and Commun., Vol. 2, No. 2, Jun. 2006. [13] D. Preuveneers, Y. Vandewoude, P. Rigole, D. Ayed, and Y. Berbers, ‘‘Research challenges in mobile and context aware service development,’’ Proc. Adjunct Proceedings of the 4th International Conference on Pervasive Computing, Dublin, Ireland, pp. 125 128, May 2006. [14] D. Preuveneers, Y. Vandewoude, P. Rigole, D. Ayed, and Y. Berbers, ‘‘Context aware adaptation for component based pervasive computing system,’’ Proc. 4th International Conference on Pervasive Computing, Dublin, Ireland, pp. 125 128, May 2006. [15] http://www.cs.kuleuven.be/$davy/cogito.php. [16] H.L. Chen, ‘‘An intelligent broker architecture for pervasive context aware systems,’’ PhD Thesis, Department of Computer Science and Electrical Engineering, University of Maryland, College Park, MD, 2004. [17] A.K. Dey, ‘‘The context toolkit: a toolkit for context aware applications,’’ http://www.cs.berkeley.edu/$dey/ context.html. [18] A.K. Dey,‘‘Supporting the construction of context aware applications,’’ http://www.vs.inf.ethz.ch/events/ dag2001/slides/anind.pdf. [19] T. Gu, H.K. Pung, and D.Q. Zhang, ‘‘A service oriented middleware for building context aware services,’’ Elsevier Journal of Network and Computer Applications, Vol. 28, No. 1, pp. 1 18, Jan. 2005. [20] T. Gu, X.H. Wang, H.K. Pung, and D.Q. Zhang, ‘‘An ontology based context model in intelligent environ ments,’’ Proc. Communication Networks and Distributed Systems Modeling and Simulation Conference, San Diego, CA, pp. 270 275, Jan. 2004. [21] ‘‘MIT Project Oxygen: overview,’’ http://oxygen.csail.mit.edu/Overview.html. [22] P. Debaty, P. Goddi, and A. Vorbau, ‘‘Integrating the physical world with the Web to enable context enhanced services,’’ HP, Tech. Rep. HPL 2003 192, 2003. [23] S. Yoshihama, P. Chou, and D. Wong, ‘‘Managing behaviour of intelligent environments,’’ Proc. IEEE International Conference on Pervasive Computing and Communications, Dalas, Fort Worth, TX, Mar. 2003. [24] S. F. Chang and A. Vetro, ‘‘Video adaptation: concepts, technologies and open issues,’’ Proc. IEEE, Vol. 93, No. 1, pp. 148 158, Jan. 2005. [25] S. F. Chang, T. Sikora, and A. Puri, ‘‘Overview of the MPEG 7 standard,’’ IEEE Trans. Circuits Syst. Video Technol., special issue on MPEG 7, Vol. 11, No. 6, pp. 688 695, Jun. 2001. [26] ‘‘Multimedia Content Description Interface Part 5: Multimedia Description Scheme,’’ ISO/IEC Standard, MPEG 7 Part 5: ISO/IEC 15938 5, Oct. 2001. [27] A. Vetro, C. Christopoulos, and H. Sun, ‘‘Video transcoding architectures and techniques: an overview,’’ IEEE Signal Processing Mag., Vol. 20, No. 2, pp. 18 29, Mar. 2003. [28] J. Xin, C. W. Lin, and M. T. Sun, ‘‘Digital video transcoding,’’ Proc. IEEE, Vol. 93, No. 1, pp. 84 97, Jan. 2005. [29] I. Ahmad, X. Wei, Y. Sun, and Y. Q. Zhang, ‘‘Video transcoding: an overview of various techniques and research issues,’’ IEEE Trans. Multimedia, Vol. 7, No. 5, pp. 793 804, Oct. 2005. [30] ‘‘Information Technology: Multimedia Framework (MPEG 21): Part 1: Vision, Technologies and Strategy,’’ ISO/IEC Standard, ISO/IEC TR 21000 1:2004, Nov. 2004. [31] I. Burnett, F. Pereira, R. Walle, and R. Koenen, The MPEG 21 Book, John Wiley and Sons, Ltd., 2006. [32] ‘‘Information Technology Multimedia Framework (MPEG 21) Part 7: Digital Item Adaptation,’’ ISO/IEC Standard, ISO/IEC 21000 7:2004, Oct. 2004. [33] F. Pereira et al., IEEE Signal Processing Mag, special issue on universal multimedia access, Mar. 2003. [34] K. Karpouzis, I. Maglogiannis, E. Papaioannou, D. Vergados, and A. Rouskas, ‘‘Integrating heterogeneous multimedia content for personalized interactive entertainment using MPEG 21,’’ Computer Communications: Special issue on Emerging Middleware for Next Generation Networks, 2006. [35] A. Vetro, ‘‘MPEG 21 digital item adaptation: enabling universal multimedia access,’’ IEEE Multimedia J., Vol. 11, No. 1, pp. 84 87, Jan. Mar. 2004.

Context based Visual Media Content Adaptation 555 [36] P. Carvalho,‘‘Multimedia content adaptation for universal access,’’ MSc Thesis, Faculdade de Engenharia da Universidade do Porto, 2004. [37] R. Mohan, J. Smith, and C.S. Li, ‘‘Adapting multimedia Internet content for universal access,’’ IEEE Trans. Multimedia, Vol. 1, No. 1, pp. 104 114, Mar. 1999. [38] S. F. Chang, ‘‘Optimal video adaptation and skimming using a utility based framework,’’ Proc. Tyrrhenian International Workshop on Digital Communications, Capri Island, Italy, Sep. 2002. [39] H. Hellwagner, ‘‘DANAE,’’ QoS Concertation Meeting (CG 2), Dec. 2004. [40] ‘‘DANAE executive summary,’’ Nov. 2005, http://danae.rd.francetelecom.com. [41] ‘‘ENTHRONE,’’ http://www.enthrone.org/. [42] M.T. Andrade, P.F. Souto, P.M. Carvalho, B. Feiten, and I. Wolf, ‘‘Enhancing the access to networked multimedia content,’’ Proc. 7th International Workshop on Image Analysis for Multimedia Interactive Services, Seoul, South Korea, Apr. 2006. [43] ‘‘Information technology: MPEG 21 multimedia framework,’’ Aug. 2006, http://mpeg 21.itec.uni klu.ac.at/ cocoon/mpeg21/ mpeg21Demo.HTML. [44] F. Pereira et al., Signal Process.: Image Commu. J., special issue on multimedia adaptation, Mar. 2003. [45] ‘‘UAProf, user agent profile,’’ OMA Open Mobile Alliance (OMA), OMA TS UAProf V2 0 20060206 A, 2006. [46] ‘‘Resource description framework (RDF): concepts and abstract syntax,’’ W3C Recommendation, Feb. 2004. [47] ‘‘Composite capability/preference profiles (CC/PP): structure and vocabularies 1.0,’’ Standard, W3C Rec ommendation, Jan. 2003. [48] M.H. Butler,‘‘Current technologies for device independence,’’ Hewlett Packard Laboratories, Bristol, UK, Tech. Rep. HPL 2001 83, Apr. 2001, http://www.hpl.hp.com/techreports/2001/HPL 2001 83.pdf. [49] M.H. Butler,‘‘Implementing content negotiation using CC/PP and WAP UAProf,’’ Hewlett Packard Laborato ries, Bristol, UK, Tech. Rep. HPL 2001 190, Aug. 2001, http://www.hpl.hp.com/techreports/2001/HPL 2001 190.pdf. [50] J. Gilbert and M. Butler,‘‘Using multiple namespaces in CC/PP and UAProf,’’ Hewlett Packard Laboratories, Bristol, UK, Tech. Rep. HPL 2003 31, Jan. 2003, http://www.hpl.hp.com/techreports/2003/HPL 2003 31.pdf. [51] X. Wang, T. DeMartini, B. Wragg, M. Paramasivam, and C. Barlas, ‘‘The MPEG 21 rights expression language and rights data dictionary,’’ IEEE Trans. Multimedia, Vol. 7, No. 3, pp. 408 416, Jun. 2005. [52] E. Rodrıguez, ‘‘Standardization of the protection and governance of multimedia content,’’ PhD Thesis, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, 2006. [53] V. Torres, E. Rodrıguez, S. Llorente, and J. Delgado, ‘‘Use of standards for implementing a multimedia information protection and management system,’’ Proc. 1st International Conference on Automated Produc tion of Cross Media Content for Multi channel Distribution, Florence, Italy, pp. 197 204, Nov. Dec. 2005. [54] A. Vetro and C. Timmerer, ‘‘Digital item adaptation: overview of standardization and research activities,’’ IEEE Trans. Multimedia, Vol. 7, No. 3, Jun 2005. [55] C. Timmerer and H. Hellwagner, ‘‘Interoperable adaptive multimedia communication,’’ IEEE Multimedia Mag., Vol. 12, No. 1, pp. 74 79, Jan. Mar. 2005. [56] L. Rong and I. Burnett, ‘‘Dynamic multimedia adaptation and updating of media streams with MPEG 21,’’ Proc. 1st IEEE Conference on Consumer Communications and Networking, pp. 436 441, Jan. 2004. [57] J. Bormans and K. Hill, ‘‘MPEG 21 Overview v.5,’’ ISO/IEC JTC1/SC29/WG11, Tech. Rep. JTC1/SC29/ WG11/N5231, Oct. 2002, http://www.chiariglione.org/mpeg/standards/mpeg 21/mpeg 21.htm. [58] ‘‘DRM specification (Candidate Version 2.0),’’ OMA, OMA TS DRM DRM V2 0 20050614 C, Sep. 2005, http://www.openmobilealliance.org/release program/docs/DRM/V2 0 20050915 C/OMA TS DRM DRM V2 0 20050915 C.pdf. [59] ‘‘TV Anytime,’’ http://www.tv anytime.org/. [60] ‘‘Windows Media digital rights management,’’ http://www.microsoft.com/windows/windowsmedia/forpros/ drm/default.mspx. [61] ‘‘XrML,’’ http://www.xrml.org/. [62] ‘‘ContentGuard,’’ http://www.contentguard.com/. [63] ‘‘Open Digital Rights Language Initiative,’’ Aug. 2007, http://odrl.net. [64] ‘‘International property & real estate systems,’’ http://www.iprsystems.com/. [65] ‘‘The Digital Property Rights Language,’’ Xerox Corporation, Manual and Tutorial, Nov. 1998, http://xml. coverpages.org/DPRLmanual XML2.html.

556 Visual Media Coding and Transmission [66] ‘‘Information technology: multimedia framework (MPEG 21): part 5: rights expression language,’’ ISO/IEC standard, ISO/IEC 21000 5:2004, Mar. 2004. [67] ‘‘Information technology: multimedia framework (MPEG 21): part 6: rights data dictionary,’’ ISO/IEC standard, ISO/IEC 21000 6:2004, May 2004. [68] ‘‘Information technology: multimedia framework (MPEG 21): part 7: digital item adaptation, AMENDMENT 1: DIA conversions and permissions,’’ ISO/IEC standard, ISO/IEC 21000 7:2004/FPDAM 1:2005(E), Jan. 2005. [69] ‘‘ADMITS: adaptation in distributed multimedia IT systems,’’ May 2004, http://admits itec.uni klu.ac.at. [70] ‘‘DAIDALOS,’’ Dec. 2005, http://www.ist daidalos.org. [71] ‘‘aceMedia,’’ Dec. 2006, http://www.acemedia.org. [72] ‘‘AXMEDIS,’’ 2007, http://www.axmedis.org. [73] ‘‘Projecte Integrat,’’ Dec. 2005, http://www.i2cat.cat/i2cat/servlet/I2CAT.MainServlet?seccio 21 33. [74] C. Timmerer, T. DeMartini, and H. Hellwagner, ‘‘The MPEG 21 multimedia framework: conversions and permissions,’’ Department of Information Technology (ITEC), Klagenfurt University, Klagenfurt, Tech. Rep. TR/ITEC/06/1.03, Mar. 2006. [75] ‘‘SOAP version 1.2 part 1: messaging framework (second edition),’’ W3C Recommendation, Apr. 2007, http:// www.w3.org/TR/soap12 part1/. [76] S. Dogan, A.H. Sadka, and M.T. Andrade,‘‘Development and performance evaluation of a shared library of visual format conversion tools,’’ VISNET I FP6 NoE deliverable D17, Nov. 2005. [77] Y. Wang, J. G. Kim, S. F. Chang, and H. M. Kim, ‘‘Utility based video adaptation for universal multimedia access (UMA) and content based utility function prediction for real time video transcoding,’’ IEEE Trans. Multimedia, Vol. 9, No. 2, pp. 213 220, Feb. 2007. [78] M. Kropfberger, K. Leopold, and H. Hellwagner, ‘‘Quality variations of different priority based temporal video adaptation algorithms,’’ Proc. IEEE 6th Workshop on Multimedia Signal Processing, pp. 183 186, Oct. 2004. [79] G.M. Muntean, P. Perry, and L. Murphy, ‘‘Quality oriented adaptation scheme for video on demand,’’ Electronics Letters, Vol. 39, No. 23, pp. 1689 1690, 2003. [80] H. Shu and L.P. Chau, ‘‘Frame skipping transcoding with motion change consideration,’’ Proc. IEEE International Symposium on Circuits and Systems, Vancouver, Canada, Vol. 3, pp. 773 776, May 2004. [81] J.N. Hwang, T.D. Wu, and C.W. Lin, ‘‘Dynamic frame skipping in video transcoding,’’ Proc. IEEE Workshop on Multimedia Signal Processing, Redondo Beach, CA, pp. 616 621, Dec. 1998. [82] A. Doulamis and G. Tziritas, ‘‘Content based video adaptation in low/variable bandwidth communication networks using adaptable neural network structures,’’ Proc. International Joint Conference on Neural Networks, Vancouver, Canada, Jul. 2006. [83] Y. Wang, S.F.C. Alexander, and C. Loui, ‘‘Subjective preference of spatio temporal rate in video adaptation using multi dimensional scalable coding,’’ Proc. IEEE International Conference on Multimedia Expo, Taipei, Taiwan, 2004. [84] W.H. Cheng, C.W. Wang, and J.L. Wu, ‘‘Video adaptation for small display based on content recomposition,’’ IEEE Trans. Circuits Syst. Video Technol., Vol. 17, No. 1, pp. 43 58, 2007. [85] M. Bettini, R. Cucchiara, A.D. Bimbo, and A. Prati, ‘‘Content based video adaptation with user’s preferences,’’ Proc. IEEE International Conference on Multimedia and Expo, Taipei, Taiwan, Vol. 3, pp. 1695 1698, Jun. 2004. [86] M. Xu, J. Li, Y. Hu, L.T. Chia, B.S. Lee, D. Rajan, and J. Cai, ‘‘An event driven sports video adaptation for the MPEG 21 DIA framework,’’ Proc. 2006 IEEE International Conference on Multimedia and Expo, Ontario, Canada, pp. 1245 1248, Jul. 2006. [87] T. Warabino, S. Ota, D. Morikawa, M. Ohashi, H. Nakamura, H. Iwashita, and F. Watanabe, ‘‘Video transcoding proxy for 3G wireless mobile Internet access,’’ IEEE Commun. Mag., Vol. 38, No. 10, pp. 66 71, Oct. 2000. [88] M.A. Bonuccelli, F. Lonetti, and F. Martelli, ‘‘Temporal transcoding for mobile video communication,’’ Proc. 2nd Annual International Conference on Mobile and Ubiquitous Systems: Networking and Services, San Diego, CA, pp. 502 506, Jul. 2005. [89] S. Dogan, S. Eminsoy, A.H. Sadka, and A.M. Kondoz, ‘‘Personalized multimedia services for real time video over 3G mobile networks,’’ Proc. IEE International Conference on 3G Mobile Communication Technologies, London, UK, pp. 366 370, May 2002. [90] Z. Lei and N.D. Georganas, ‘‘Video transcoding gateway for wireless video access,’’ Proc. IEEE Canadian Conference on Electrical and Computing Engineering, Montreal, Canada, Vol. 3, pp. 1775 1778, May 2003.

Context based Visual Media Content Adaptation 557 [91] E. Amir, S. McCanne, and R.H. Katz, ‘‘An active service framework and its application to real time multimedia transcoding,’’ Proc. ACM Conference on Applications, Technologies, Architectures, and Protocols for [92] Computer Communication, Vancouver, Canada, pp. 178 189, Aug. Sep. 1998. [93] M. Ott, G. Welling, S. Mathur, D. Reininger, and R. Izmailov, ‘‘The JOURNEY, active network model,’’ IEEE [94] J. Sel. Areas Commun., Vol. 19, No. 3, pp. 527 536, Mar. 2001. K.B. Shimoga, ‘‘Region of interest based video image transcoding for heterogeneous client displays,’’ Proc. [95] Packet Video Workshop, Pittsburgh, PA, Apr. 2002. [96] A. Sinha, G. Agarwal, and A. Anbu, ‘‘Region of interest based compressed domain video transcoding [97] scheme,’’ Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, Montreal, [98] Canada, Vol. 3, pp. 161 164, May 2004. [99] A. Vetro, H. Sun, and Y. Wang, ‘‘Object based transcoding for adaptable video content delivery,’’ IEEE Trans. Circuits Syst. Video Technol., Vol. 11, No. 3, pp. 387 401, Mar. 2001. [100] L. Itti, C. Koch, and E. Niebur, ‘‘A model of saliency based visual attention for rapid scene analysis,’’ IEEE [101] Transactions on Pattern Analysis and Machine Intelligence, Vol. 20, No. 11, pp. 1254 1259, Nov. 1998. [102] W. Osberger and A.J. Maeder, ‘‘Automatic identification of perceptually important regions in an image,’’ Proc. [103] IEEE International Conference on Pattern Recognition, Brisbane, Australia, Aug. 1998. [104] L.Q. Chen, X. Xie, X. Fan, W.Y. Ma, H.J. Zhang, and H.Q. Zhou, ‘‘Avisual attention model for adapting images on small displays,’’ ACM Multimedia Syst. J., Vol. 9, No. 4, pp. 353 364, Oct. 2003. [105] D.S. Cruz, T. Ebrahimi, M. Larsson, J. Askelof, and C. Christopoulos, ‘‘Region of interest coding in JPEG 2000 for interactive client/server applications,’’ Proc. IEEE International Workshop on Multimedia Signal [106] Processing, Copenhagen, Denmark, pp. 389 394, Sep. 1999. A.P. Bradley and F.W.M. Stentiford, ‘‘Visual attention for region of interest coding in JPEG 2000,’’ Journal of [107] Visual Communication and Image Representation, Vol. 14, No. 3, pp. 232 250, Sep. 2003. [108] L. Favalli, A. Mecocci, and F. Moschetti, ‘‘Object tracking for retrieval applications in MPEG 2,’’ IEEE Trans. Circuits Syst. Video Technol., Vol. 10, No. 3, pp. 468 476, Jun. 1997. [109] K. Yoon, D. DeMenthon, and D. Doermann, ‘‘Event detection from MPEG video in the compressed domain,’’ Proc. IEEE International Conference on Pattern Recognition, Barcelona, Spain, Sep. 2000. [110] B.S. Manjunath, J.R. Ohm, V.V. Vasudevan, and A. Yamada, ‘‘Color and texture descriptors,’’ IEEE Trans. [111] Circuits Syst. Video Technol., Vol. 11, No. 6, pp. 703 715, Jan. 2001. G. Agarwal, A. Anbu, and A. Sinha, ‘‘A fast algorithm to find the region of interest in the compressed MPEG [112] domain,’’ Proc. IEEE International Conference on Multimedia and Expo, Baltimore, MD, Vol. 2, pp. 133 136, Jul. 2003. V. Mezaris, I. Kompatsiaris, and M.G. Strintzis, ‘‘Compressed domain object detection for video understand ing,’’ Proc. 5th International Workshop on Image Analysis for Multimedia Interactive Services, Lisbon, Portugal, Apr. 2004. M. Soleimanipour, W. Zhuang, and G.H. Freeman, ‘‘Modeling and resource allocation in wireless multimedia CDMA systems,’’ Proc. 48th IEEE Vehicular Technology Conference, Vol. 2, Nos 18 21, pp. 1279 1283, May 1998. S. Zhao, Z. Xiong, and X. Wang, ‘‘Optimal resource allocation for wireless video over CDMA networks,’’ Proc. International Conference on Multimedia and Expo, Vol. 1, pp. 277 80, Jul. 2003. H. Schwarz, D. Marpe, and T. Wiegand, ‘‘Overview of the scalable video coding extension of the H.264/AVC standard,’’ Joint Video Team (JVT) of ISO/IEC MPEG & ITU T VCEG (ISO/IEC JTC1/SC29/WG11 and ITU T SG16 Q.6), 23rd meeting, San Jose, CA, Input Doc., Apr. 2007. D.D. Schrijver, W.D. Neve, D.V. Deursen, S.D. Bruyne, and R. van der Walle, ‘‘Exploitation of interactive region of interest scalability in scalable video coding by using an XML driven adaptation framework,’’ Proc. 2nd International Conference on Automated Production of Cross Media Content for Multi channel Distri bution, Leeds, UK, pp. 223 231, Dec. 2006. ‘‘Advanced video coding for generic audiovisual services,’’ ITU T Recommendation H.264, Mar. 2005. M.H. Lee, H.W. Sun, D. Ichimura, Y. Honda, and S.M. Shen, ‘‘ROI slice SEI message,’’ Joint Video Team (JVT) of ISO/IEC MPEG & ITU T VCEG (ISO/IEC JTC1/SC29/WG11 and ITU T SG16 Q.6), Geneva, Switzerland, Input Doc. JVT S054, Apr. 2006. ‘‘Applications and requirements for scalable video coding,’’ Tech. Rep. ISO/IEC JTC/SC29/WG11/N6880, Jan. 2005.

558 Visual Media Coding and Transmission [113] P. Lambert, D.D. Schrijver, D.V. Deursen, W.D. Neve, Y. Dhondt, and R. van der Walle, ‘‘A real time content adaptation framework for exploiting ROI scalability in H.264/AVC,’’ Lecture Notes in Computer Science, [114] Springer, Germany, pp. 442 453, 2006. [115] M.T. Andrade, P. Bretillon, H. Castro, P. Carvalho, and B. Feiten, ‘‘Context aware content adaptation: a [116] systems approach,’’ Proc. European Symposium Mobile Media Delivery, Sardinia, Italy, Sep. 2006. F. Pereira and I. Burnett, ‘‘Universal multimedia experiences for tomorrow,’’ IEEE Signal Processing Mag., [117] Vol. 20, pp. 63 73, Mar. 2003. [118] M.T. Andrade, H.K. Arachchi, S. Nasir, S. Dogan, H. Uzuner, A.M. Kondoz,et al., ‘‘Using context to assist the [119] adaptation of protected multimedia content in virtual collaboration applications,’’ Proc. 3rd IEEE Interna [120] tional Conference on Collaborative Computing: Networking, Applications and Worksharing, New York, NY, [121] Nov. 2007. A. Farquhar, R. Fikes, W. Pratt, and J. Rice, ‘‘Collaborative ontology construction for information integration,’’ [122] Stanford University, Tech. Rep. KSL 95 63, Aug. 1995. [123] D. Reynolds, ‘‘Jena 2 inference support,’’ Jul. 2007, http://jena.sourceforge.net/inference/. [124] ‘‘Creative Commons,’’ http://www.creativecommons.org/. J. Delgado and E. Rodrıguez, ‘‘Towards the interoperability between MPEG 21 REL and Creative Commons [125] licenses,’’ ISO/IEC JTC 1/SC 29/WG 11/M13118, Montreux, Switzerland, Apr. 2006. [126] T. Kim, J. Delgado, F. Schreiner, C. Barlas, and X. Wang, ‘‘Information technology: multimedia framework [127] (MPEG 21): part 5: rights expression language, AMENDMENT 3: ORC (Open Release Content) profile: a [128] proposal for ISO/IEC standardization,’’ ISO/IEC 21000 5:2004/FPDAM 3, Apr. 2007. [129] M. Bystrom and T. Stockhammer, ‘‘Dependent source and channel rate allocation for video transmission,’’ [130] IEEE Transactions on Wireless Communications, Vol. 3, No. 1, pp. 258 268, Jan. 2004. [131] P. Cherriman, E.L. Kuan, and L. Hanzo, ‘‘Burst by burst adaptive joint detection CDMA/H.263 based video telephony,’’ IEEE Trans. Circuits Syst. Video Technol., Vol. 12, No. 5, pp. 342 348, May 2002. C. Kodikara, S. Worrall, S. Fabri, and A.M. Kondoz, ‘‘Performance evaluation of MPEG 4 video telephony over UMTS,’’ Proc. 4th IEE International Conference on 3G Mobile Communication Technologies, London, UK, pp. 73 77, Jun. 2003. R.E. Dyck and D.J. Miller, ‘‘Transport of wireless video using separate, concatenated, and joint source channel coding,’’ Proc. IEEE, Vol. 87, No. 10, pp. 1734 1750, Oct. 1999. I.M. Kim and H.M. Kim, ‘‘An optimum power management scheme for wireless video service in CDMA systems,’’ IEEE Transactions on Wireless Communications, Vol. 2, No. 1, pp. 81 91, Jan. 2003. M.M. Hannuksela, Y.K. Wang, and M. Gabbouj, ‘‘Sub picture: ROI coding and unequal error protection,’’ Proc. IEEE International Conference on Image Processing, Rochester, NY, Vol. 3, pp. 537 540, Sep. 2002. ‘‘Coding of audio visual objects: part 2: visual: second edition,’’ ISO/IEC standard, ISO/IEC FDIS 14496 2 (MPEG 4 Visual), Dec. 2001. S. Worrall, S. Fabri, A. Sadka, and A. Kondoz, ‘‘Prioritization of data partitioned MPEG 4 over mobile networks,’’ Europ. Trans. Telecom., Vol. 12, No. 3, May Jun. 2001. ‘‘The SVT high definition multi format test set,’’ ftp://vqeg.its.bldrdoc.gov/HDTV/SVT MultiFormat. J. Vieron, M. Wien, and H. Schwarz,‘‘JSVM 7 software,’’ Joint Video Team (JVT) of ISO/IEC MPEG & ITU T VCEG (ISO/IEC JTC1/SC29/WG11 and ITU T SG16 Q.6), 20th Meeting, Klagenfurt, Austria, Output Doc. JVT T203, Jul. 2006.

Index 2 Access links, 215, 345 2D + t, 40, 41, 65, 66, 67, 69, 96 ACELP, 221, 341 2D to 3D conversion algorithms, 44 AceMEDIA, 460 2G MSC, 231 Adaptation authorization, 459, 460, 481, 2G SGSN, 231 525, 527 3 Adaptation Decision Engine, 5, 488 3D, 1, 2, 3, 30, 36, 37, 39, 40, 44, 45, 46, 54, 56, Adaptation Engines Stack, 5, 495 Adaptation resource utility, 463, 464 57, 58, 60, 62, 65, 69, 71, 80, 82, 85, 87, 88, Adaptive arithmetic codec, 51 95, 96, 191, 192, 193, 200, 201, 202, 213, 230, Adaptive spreading gain control 398, 471 3D video, 1, 2, 3, 30, 36, 39, 44, 45, 46, 56, 58, techniques, 379 62, 80, 85, 87, 88, 95, 96, 193, 213, 230, 398 Adaptive QoS, 397, 398 3D wavelet, 39, 40, 69, 95, 96 ADE, 5, 434, 435, 453, 461, 464, 465, 466, 469, 3G, 2, 3, 4, 214, 221, 231, 260, 308, 313, 315, 341, 393, 394, 396, 525 481, 482, 487, 488, 492, 493, 495, 501, 512, 3GPP, 3, 212, 220, 221, 233, 234, 237, 239, 256, 521, 523, 524, 525, 526, 527, 530 259, 260, 263, 266, 267, 268, 270, 271, 272, ADE framework on MPEG XE ‘‘MPEG’’ 21, 277, 278, 279, 281, 284, 285, 297, 323, 341, 465 342, 343, 403, 413, 508 ADE Interface, 525 3GPP AMR, 221, 341 ADMITS, 459 3rd Generation Partnership Project, 212, 239, ADTE, 465 260, 267, 268, 277, 278, 284, 395, 508 Advanced forward error correction, 221, 222 Advanced video compression, 16 6 AE, 435, 448, 452, 466, 467, 468, 469, 470, 481, 64 QAM, 304, 305, 308, 312, 313 495, 498, 500, 501, 502, 510, 511, 512, 515, 520, 523, 524 8 AES, 495, 500, 523, 524, 526 8 PSK schemes, 245, 246 AIR, 318, 325, 347, 348, 350, 367, 372, 373, 374, 384, 395, 505, 508 A AMC permutation, 306 AA, 5, 434, 482, 493, 525, 527 AMR WB, 221, 315, 341, 342, 343, 344 AC coefficients, 13 API, 442, 488 Application QoS, 399, 401, 402 AQoS, 446, 448, 450, 451, 453, 458, 464, 465, 488 Visual Media Coding and Transmission Ahmet Kondoz © 2009 John Wiley & Sons, Ltd. ISBN: 978-0-470-74057-6

560 Index AR(1) model, 118 CCSR, 241, 242, 243, 245, 246, 247, 248, 279, ARE, 118 280, 281, 297, 299, 300, 336 ASC, 439 ASC model, 439 CD, 28, 88, 466 Audio Visual, 1 CD ROM, 28 Autoregressive model, 117 CGS, 42 AV, 450, 453, 468, 472, 477 CIF, 37, 61, 68, 69, 70, 81, 82, 84, 88, 91, 161, AV content analysis, 468 AVC, 1, 2, 19, 28, 29, 30, 36, 40, 41, 42, 44, 58, 162, 168, 169, 506, 508, 511, 514, 515 CIR, 318, 325, 358, 359, 360, 361, 363, 364, 59, 60, 61, 62, 63, 64, 65, 73, 75, 79, 95, 96, 97, 112, 120, 127, 129, 131, 132, 137, 142, 365, 366, 367, 368, 369, 370, 373, 374, 375, 152, 153, 154, 155, 156, 157, 158, 170, 173, 376, 377, 378, 384 180, 186, 190, 192, 193, 200, 392, 468, 469, CMP, 112, 113 495, 511, 512, 514, 517, 518, 520, 521, Coarse grain scalability, 42 522, 527 CoBrA, 442, 444 AWGN source, 241, 263, 278, 405, 406 CoDAMoS, 441 Coding mode control, 160 B CoGITO, 441 B frames, 17, 18, 38, 79 Committee draft, 88 Base transceiver station, 98, 231 Complementary alpha plane, 176 Bit Allocation, 153, 160, 162, 163, 164, 165 Complementary shape data, 175, 176, 177, 179 Bit Plane Extraction, 104 Complex FIR filter, 274 Bitstream Syntax Description Language, 449 Composite capability profiles, 445 BLER, 240, 241, 242, 243, 244, 245, 246, 248, Computational power, 39 Conditional entropies, 101 249, 250, 251, 277, 279, 280, 281, 284, 285, Content Adaptation, 433, 466, 481, 482, 501, 287, 288, 291, 353, 354, 355, 366, 367, 374, 510, 512, 522 375, 376, 378, 384, 394, 423, 508, 509, 510 Context Broker Architecture, 442 Block matching, 14, 36, 194 Context brokers, 440 Block transform, 20 Context classes, 486 BlueSpace, 444 Context definition, 436 BSD, 451, 452, 520 Context interpreter, 440, 443, 444 BSD tools, 452 Context mediation, 439, 441 BSDL, 449, 452 Context of usage, 435, 440, 446, 464, 473 BSS, 231, 233 Context ontology Languag, 439 BTS, 98, 231 Context providers, 5, 472, 485 Buffer Control, 153, 162 Context space, 439 Context toolkit, 439, 442, 443 C Context adaptive binary arithmetic coding, 51 CAE method, 48 Context aware content adaptation, 4, 433, 434, Call admission control, 214, 397 435, 445, 470, 479, 480, 481, 482, 486, 495, Call request, 385, 386, 387, 389, 390 530, 531 Carrier Frequency, 241, 392, 405, 406 Contextual information, 436, 437, 470, 476, Carrier to interference, 234, 245, 255, 263, 361, 485, 500 Contour matching, 49 376, 403 CoOL, 439 Catalan Integrated Project, 453 Core network, 371 CC, 284, 286, 313, 322, 326, 327, 330, 332, 342, Correlated frame structure, 73, 74 COSSAP, 234, 240, 403 353, 355, 383, 424, 425, 426, 427, 428, 429, COSSAP hierarchical models, 240 430, 431, 445, 455, 456, 493, 508, 509 COST 231 Walfish Ikegami model, 356, 361 CC/PP, 445, 455, 456 Creation of Smart Local City Services, 441 CC/PP eXAMPLE, 457

Index 561 CROSLOCIS, 441 Doppler spectrum, 239 CS/H.264/AVC, 495, 500, 501 Downsampling, 8, 204, 205 Curvature scale space image, 46, 47 DPCH fields, 270 CxP…. 5, 434, 444, 473, 482, 486, 487, 488, DPDCH, 262, 270, 271, 292, 293, 317 DPRL, 459 493, 523, 526, 530 DRM, 4, 5, 433, 434, 435, 436, 458, 459, 469, D 470, 473, 481, 482, 485, 512, 522, 531 DANAE, 453, 459 DRM based adaptation, 435, 531 dB, 32, 34, 52, 53, 64, 70, 75, 77, 82, 84, 85, 87, DRM enabled adaptation, 5, 434, 470, 473 DS, 171, 449 92, 96, 112, 113, 125, 126, 129, 130, 133, 134, DSL, 3, 37, 213 135, 136, 142, 143, 148, 161, 162, 168, 169, 170, 178, 188, 189, 202, 205, 239, 240, 241, E 242, 243, 244, 245, 246, 247, 248, 249, 250, EBCOT algorithm, 51, 52 251, 252, 255, 272, 276, 277, 278, 279, 281, EC, 277, 278, 314, 453 285, 287, 288, 292, 294, 299, 305, 307, 312, EDDD, 145 319, 320, 326, 327, 328, 329, 332, 333, 335, EDGE, 2, 3, 211, 212, 231, 234, 246, 248, 344, 337, 338, 339, 340, 341, 344, 345, 346, 351, 353, 354, 355, 357, 359, 360, 361, 364, 365, 401, 403, 404, 405, 421, 422, 424, 425, 431, 432 366, 367, 368, 369, 370, 374, 375, 376, 377, EDGE Emulator, 404 378, 379, 380, 381, 383, 387, 388, 390, 392, EDGE to UMTS System, 424 393, 394, 395, 405, 406, 422, 423, 424, 425, EDGE to UMTS transmission, 421, 424, 431 426, 427, 428, 429, 430, 431, 488, 509, 510, Editing like content adaptation, 452 511, 518, 521, 522 EGPRS data flow simulator, 403 DCH, 262, 284, 317 EGPRS physical link layer model, 236, DCT, 1, 9, 10, 11, 12, 13, 20, 28, 34, 35, 36, 41, 69, 103, 113, 120, 121, 122, 123, 133, 134, 248, 252 143, 144, 145, 146, 147, 159, 171, 172, 173, EGPRS protocol layer, 362 220, 346, 348, 349, 466, 506 EIMS, 453 Deblocking filters, 21 Encoder complexity, 99, 193 Decoder complexity, 39, 98 Encoder control, 15, 16 Depth map, 45, 46, 57 ENTHRONE project, 453 DI, 64, 115, 120, 165, 446, 447, 448, 449, 461, Entropy coder, 15, 159 472, 520, 524, 525, 526, 527, 530 EPB, 90, 91, 92, 93, 96 DIA, 445, 446, 448, 449, 453, 458, 472, 473, Equalization, 219 482, 486, 496, 497, 498, 499, 528, 529 Error concealment, 143, 153, 173, 175, 176, DIA tools, 445, 458 DIBR, 45, 64, 65, 80, 87, 88 182, 183, 185, 186 DID, 446, 472 Error patterns, 308 DIDL, 446 Error propagation, 24 Differential pulse code modulation, 13, 28, 39, Error resilience, 153, 155, 181, 182 40, 41 Error robust, 71 Differentiated services, 397 Error tracking, 223 Digital Rights Management, 4, 458 Error prone environments, 155, 180, 186, 218, DIP, 453 Disparity information, 196 220, 221, 326, 355 Distributed adaptation, 453 Error robustness algorithm, 74, 75 Distributed Systems and Computer ERTERM, 91 Networks, 441 ETSI EFR, 221, 341 DistriNet, 441 Event detection, 468 DMIF, 220 Excerpt of a network UED, 451 Exclusive OR, 74 Extensible context ontology, 439 Extensible markup language, 439

562 Index Extensible stylesheet language H T transformation, 452 H.264, 1, 2, 8, 16, 17, 18, 19, 20, 21, 22, 23, 24, Extrapolation orientations, 73 25, 26, 27, 28, 30, 34, 36, 40, 41, 42, 44, 53, 54, 55, 56, 58, 59, 60, 61, 62, 63, 64, 65, 73, F 75, 79, 80, 81, 88, 95, 96, 97, 98, 112, 120, FA, 512 127, 129, 131, 132, 137, 142, 152, 153, 154, Fast fading, 271, 356, 358, 388 155, 156, 157, 158, 170, 172, 180, 186, 189, FD, 466 190, 200, 202, 203, 392, 468, 469, 495, 510, FEC/ARQ method, 223 511, 512, 514, 517, 518, 520, 521, 522, 527 FIBME algorithm, 67 HardwarePlatform, 454 FIFO buffer, 254 HCI, 436 FIRE, 235, 236 HDTV, 29, 501 FIRE code, 236 Heterogeneous networks, 215, 217, 330, 396, Flexible Macroblock Ordering, 25 397, 427, 466 FlexMux, 220 Hierarchical coding, 23 FMO, 24, 25, 26, 27, 131, 132, 155, 186, 187, HP, 444 HR/DSSS, 3, 212 189, 468, 469 HSCSD, 3, 211, 241, 405 FP, 453 HT, 238, 241, 405 Frame dropping, 463, 466 HTML, 454 FUSC, 302, 305, 306, 307 Huffman codec, 48 G I Gaussian, 106, 115, 138, 139, 140, 172, IBM, 444 IBMH MCTF, 70 233, 234, 240, 263, 272, 281, 290, ICAP, 454 387, 388 Ideal rake receiver, 273, 274, 299 Gaussian probability distribution, 139 IEC, 1, 36, 41, 60, 61, 150, 445, 446 gBS, 449, 452 IEEE802.16e, 85, 301 gBS tools, 449 IETF, 321, 330, 454 gBSDtoBin, 453 IFFT block, 302, 307 GERAN, 221, 231, 232, 233, 239, 254, 312, IMT2000, 270 341, 403 In Band MCTF, 70 GERAN data flow model, 403 Independent Replication, 391 Global motion model, 177, 184, 196 Info pyramid, 462, 463 Global motion parameters, 176 In loop filter, 22 GMSK modulation scheme, 236, 249 Integer transform, 20 Gold codes, 269 Interleaving, 88, 90, 94, 99, 219, 236, 238, 240, GOP, 52, 53, 71, 72, 73, 75, 76, 77, 96, 109, 112, 114, 115, 116, 117, 118, 120, 121, 122, 264, 267, 268, 278, 285, 289, 291, 301, 302, 124, 126, 129, 130, 133, 134, 142, 151, 154, 304, 325, 406 198, 202, 203, 206, 208 Internet Content Adaptation Protocol, 454 GOS, 160, 163, 164, 165, 167 Inter networked QoS, 400, 397, 399, 401 GPRS, 3, 211, 212, 215, 231, 233, 234, 235, 236, Inter view Prediction, 196, 206 237, 240, 241, 242, 243, 244, 245, 252, 253, Interworked heterogeneous networks, 215 254, 255, 256, 257, 260, 312, 315, 332, 333, Intra frames, 14 334, 335, 336, 337, 339, 344, 345, 363, 371, Intra prediction, 21 403, 405, 411, 412, 417, 425, 427, 428, 430 Inverse Quantization, 108 GPRS PDTCHs, 233, 236 IP video conferencing, 221, 341 GPRS SNDC, 252, 253, 403 IPR, 459 GSM 05.05, 240, 241, 243, 405 IROI, 468, 498, 500, 511, 517, 522 GSM SACCH, 235

Index 563 IROI adaptation, 468, 498, 517 MAD, 154, 166 IS 95, 3, 212 MALR, 111, 112, 113 ISO, 1, 28, 35, 36, 41, 60, 61, 150, 191, 192, Maximum likelihood, 137 MB, 8, 14, 15, 17, 18, 19, 20, 21, 22, 25, 26, 28, 212, 445, 446 IST, 38, 161, 168, 169, 385, 453 29, 35, 62, 145, 148, 149, 150, 154, 155, 158, ITEC, 453 159, 160, 162, 166, 169, 187, 188, 190, 347, ITU Standards, 28 350, 507, 512, 513, 515, 517 ITU T G.722.2, 221, 341 MB motion vectors, 160 ITU T G.729, 221 MC EZBC codec, 65 MCS 1, 233, 234, 236, 237, 238, 241, 246, 247, J 248, 250, 339, 340, 341, 363, 364, 365, 367, JAS UEP scheme, 381, 383 370, 373, 405, 406, 412, 413, 414, 415, 422, Joint MVS Encoding, 153, 156 423, 424, 425, 426, 427, 428, 429 Joint MVS Transcoding, 153, 170 MCS 4, 234, 237, 241, 246, 247, 248, 250, 251, JPEG, 11, 13, 14, 36, 54, 69, 88, 89, 90, 91, 93, 339, 340, 363, 364, 413, 414, 415 MCS 9, 234, 237, 246, 339, 340, 363, 364, 413, 94, 96, 446 414, 415, 422 JPEG 2000, 36, 69, 88, 89, 90, 91, 93, 94, 96 MDC, 2, 33, 36, 37, 39, 42, 43, 44, 79, 80, 81, JPWL EPB, 91, 92 82, 83, 84, 85, 86, 87, 88, 96, 223 JPWL system description, 89 MDC decoder, 44, 81, 82 JSVM, 60, 75, 76, 77, 79, 82, 84, 96, 193, 197, MDC stream, 44, 80, 81, 82, 85 MDS, 445, 464 198, 514, 517, 518, 520 MDS media characteristics, 464 Mean absolute log likelihood ratio, 111 K Medium grain scalability, 42 Kalman filtering, 116, 117 Mesh based temporal filtering, 51 Key frames, 71, 72, 73, 74, 76, 77, 96, 106, 107, MGS, 42 MIT, 444 109, 110, 112, 114, 116, 117, 120, 127, 129, Mobile Alliance Forum, 454 130, 131, 132, 133, 134, 137, 139, 142, 466 Mobile Station mode, 410 Mobile Terminal, 241, 405, 406 L Mobile terminal power, 227 Lagrangian cost, 19, 73 Mobile terminal velocity, 285 Lagrangian optimization, 19, 20 Mode decision, 15 Laplacian, 106, 138, 139, 140, 146, 172 MOS, 343, 344, 345 Laplacian distribution, 140, 146, 172 Motion compensation, 14 LAPP, 110, 111, 147 Motion estimation, 14, 40, 41, 107, 114, 118, 127 Layered transmission, 48 Motion compensated interpolation, 114, 117 Layers of depth images, 64 Motion compensated temporal filtering, 44, Link adaptation, 221, 225, 356, 362, 394 51, 96 LLC PDUs, 233 MP3, 29, 446 LLC PDU, 254, 333, 334, 367 MPEG, 465 LLR, 107, 108, 137, 138, 139 MPEG Ad Hoc Group, 96 Log likelihood ratio, 107 MPEG 21, 96, 445, 446, 447, 448, 451, 452, LogMap algorithm, 274, 281, 288, 299 453, 454, 457, 458, 459, 460, 461, 464, 468, 470, 471, 472, 473, 474, 475, 476, 482, 486, M 488, 493, 495, 523, 525, 526, 527, 528, 530 MAC, 3, 58, 59, 61, 62, 65, 192, 212, 232, 233, MPEG 21 DIA, 445, 446, 447, 451, 452, 453, 459, 460, 464, 468, 470, 471, 472, 473, 474, 238, 239, 240, 243, 252, 253, 256, 258, 259, 475, 476, 488, 493, 523, 525, 527, 528 260, 262, 284, 295, 296, 301, 311, 315, 316, 317, 333, 334, 362, 403, 411, 412, 413, 417 MAC multiplexing, 296, 317 Macroblock Bit Allocation, 166

564 Index MPEG 21 REL, 459, 460, 493 Orthogonality factor, 263, 289, 290, 291, 388, MPEG 4, 2, 8, 18, 19, 27, 28, 29, 30, 39, 48, 50, 392 53, 54, 58, 59, 60, 61, 62, 63, 64, 65, 95, 97, OSCRA, 495, 498, 500 154, 155, 161, 162, 163, 168, 170, 175, 179, Overhead, 61, 71, 74, 76, 79, 94, 96, 112, 194, 180, 190, 192, 219, 220, 222, 225, 252, 253, 254, 255, 256, 294, 296, 315, 318, 319, 320, 202, 220, 322, 331, 332, 346, 382, 417 321, 325, 326, 327, 328, 329, 331, 332, 333, OWL, 439, 442, 455, 481, 492, 531 334, 335, 336, 340, 346, 347, 348, 352, 353, 356, 363, 367, 373, 380, 381, 384, 395, 396, P 401, 407, 408, 409, 410, 416, 421, 423, 424, P frames, 14, 16, 17, 38, 116 426, 431, 447, 466, 468, 469, 503, 505, 506, P2P, 454 507, 527 PACCH, 232 MPEG 4 MAC, 58, 59, 60, 61, 62, 63, 64, 95 Packet Data Block Type 2, 235 MPEG 7 descriptors, 468 Packet Data Block Type 3, 235 MPEG 7 DSs, 450, 472 Packet Data Block Type 4, 236 MQ coder, 91 PAGCH, 232 Multimedia Communications, 214, 217 PAL, 28 Multimedia Description Schemes, 445 Parameter set, 24, 25, 85, 167, 256, 262, 277, Multiple Auxiliary Component, 58 multiple video sequence, 152, 156, 162 282, 283, 286, 292, 296, 312, 313, 315, 321, Multiplexing, 181, 221, 224, 260, 284, 345, 406 322, 325, 344, 346, 405, 410, 421, 424, 431, Multi sequence rate control, 170 517 Multi View Coding, 16, 192, 193, 196, 201, Parity Bit, 106 203, 204, 206, 207, 208, 209, 210 Parity bit puncturer, 106 MVO performance, 160 Parity check, 237, 301, 313 MVO rate control, 163 PCCCH, 232 MVP, 513 PCCPCH, 261 MVS, 2, 152, 153, 156, 157, 158, 159, 160, 162 PDA, 456, 500, 501, 525, 527 MVS Encoding, 153, 162 PDSCH, 262 MVS Encoding Rate Control, 153, 162 PDTCH, 232, 240, 241, 336, 405 Percentage of sign changes, 111 N Performance evaluation objective, 5, 6, 30, 31, NA CDMA IS 127, 221, 341 34, 35, 43, 53, 56, 65, 95, 121, 131, 137, 155, NACK, 224, 412 190, 201, 204, 214, 231, 338, 343, 354, 398, NAL, 468 434, 436, 453, 462, 466, 467, 480, 481, 515, NA TDMA IS 641, 221, 341 517, 520, 521 NoE, 465 Performance evaluation Subjective, 6, 30, 31, Non normative, 2, 152, 154, 155, 190 32, 53, 56, 62, 65, 67, 68, 71, 84, 85, 88, Non normative tools, 2, 152 95, 96, 112, 153, 162, 173, 175, 180, 181, Normative tools, 2, 152 182, 186, 204, 210, 329, 338, 343, 345, 451, 462, 510 O Physical Context, 437, 438, 476 OAVE encoding scheme, 48 PLR, 132, 133, 134, 136, 143, 188 OCNS, 277, 314 Post filter, 22 Odd stream, 81, 82 Power control algorithm, 291, 292, 293, 294, ODRL, 459 328, 376 ODWT, 66, 67, 69, 70 Power limitations, 217 OMA, 459 PPCH, 232 Ontologies, 438, 481, 531 PRACH, 232, 262 Open Digital Rights Language, 459 Prediction step, 66, 69 Preferences Profiles, 445 PRISM codec, 103

Index 565 Profiling, 434, 467, 480 Rate matching, 267, 278, 285, 406 Project Integration, 460 Rate matching ratio, 285, 406 Propagation model, 238, 239, 245, 248, 254, Rayleigh distribution, 239, 271 Rayleigh fading channel, 141, 265 256, 263, 270, 299, 337, 356, 387, 388 RD, 32, 33, 50, 76, 98, 99, 103, 112, 113, 114, PSCS, 111, 112, 113, 114 PSD, 115, 120 115, 118, 119, 120, 124, 130, 131, 132, 133, PSNR, 31, 32, 33, 34, 52, 53, 55, 61, 64, 68, 70, 134, 136, 137, 142, 143, 152, 154, 155, 156, 157, 159, 172, 208, 513, 515, 517, 518, 519 75, 76, 77, 82, 85, 87, 93, 112, 113, 125, 126, R D performance, 48, 49, 50, 52, 58, 61, 64, 65 129, 132, 133, 134, 135, 136, 142, 148, 149, RDD, 459, 493, 496 161, 162, 168, 169, 170, 178, 188, 189, 202, RDF, 455, 456 207, 318, 319, 320, 321, 322, 325, 326, 327, RDO, 154 328, 329, 332, 333, 335, 337, 338, 340, 341, RDOPT, 518 346, 351, 354, 355, 364, 365, 366, 367, 368, Real time communication, 30 369, 370, 374, 379, 380, 381, 383, 395, 421, Redundant Slices, 27, 44 422, 423, 424, 425, 426, 427, REL, 459, 460, 493, 496 428, 429, 430, 431, 509, 511, 515, 518, 521, 522 Residual Error Descriptor, 90 PUSC, 85, 302, 305, 306, 307, 308, 309, Resource Management, 225, 385 310, 312 Resources context, 437, 476 Resources context category, 476 Q Retransmission techniques, 223 QCIF, 48, 70, 74, 77, 88, 110, 112, 120, 122, RLC/MAC blocks, 233, 236, 243, 295, 296, 334, 413, 417 124, 132, 142, 147, 161, 169, 321, 353, 367, RLC/MAC layer, 254, 294, 296, 323, 362, 409 421, 511, 515, 527, 530 RM8, 154 QoS, 3, 4, 71, 89, 213, 214, 215, 216, 217, 219, Robust FMO Scheme, 153, 186 220, 223, 224, 226, 230, 259, 262, 282, 294, ROI, 467, 468, 469, 470, 471, 498, 500, 501, 296, 298, 315, 316, 345, 347, 371, 372, 378, 507, 510, 511, 512, 514, 515, 516, 517, 526 385, 393, 394, 395, 396, 397, 398, 399, 400, ROI selection algorithm, 467, 516 401, 402, 404, 405, 407, 408, 409, 410, 416, ROPE algorithm, 143, 144 418, 419, 421, 423, 425, 426, 427, 428, 429, RSC encoder, 105, 106 430, 431, 448, 512 RTP/UDP/IP, 220, 233, 252, 254, 294, 403, 410 QoS discrepancies, 216 RTP PDU header, 254 QoS Mapping Emulator, 409 Run level coding, 13 QP, 58, 61, 63, 65, 75, 76, 77, 82, 85, 87, 147, 154, 155, 162, 163, 166, 169, 188, 208, 515, S 517, 518 SA SPECK algorithm, 51 Quantization, 12, 16, 87, 104, 121, 124, 226 Scalable intra shape coding scheme, 47, 48 Quantization bins, 104, 123 Scalable predictive coding scheme, 49 Scalable shape encoding, 46 R SCCPCH, 261, 262 RA, 90, 238, 241, 405 Scene Plane, 163, 164 Random access, 16, 17, 161, 163, 168, 193, 204, SDC stream, 82 SECAS, 440 206, 208, 209, 232 SEGMARK, 91 Random View Access, 204 Selective combining, 273 Randomizer, 302 Sensing higher level context, 461 Rate Control, 134, 136, 153, 156, 159, 253, 295 Sensitivity to Feedback Delay, 375 Rate distortion, 32, 33, 50, 76, 98, 99, 103, 112, Service Oriented Context Aware 113, 114, 115, 118, 119, 120, 124, 130, 131, Middleware, 442 132, 133, 134, 136, 137, 142, 143, 152, 154, 155, 156, 157, 159, 172, 208, 513, 515, 517, 518, 519

566 Index Shape data, 153, 175, 176, 177, 182, 183, 184, SVO rate control algorithm, 161 185, 186 Symmetric Distributed Coding, 126 Shape adaptive bit plane coding, 51, 52 T Side Information, 106, 109, 139 Tailing bits, 238 Signal Processing WorkSystem, 256, 312, 405 TB size, 294, 296, 317, 420 Signal to noise ratios, 163, 403 TBAS, 363, 364, 369, 370, 371, 384 SISO, 107, 108, 110, 111, 112, 137, 140, 141 TCH/FS speech channel, 235 Skip, 19, 513, 515, 517 TDMA, 2, 211, 219, 232, 236, 240 Slice, 17, 24, 25, 26, 27, 147, 155, 186, TDMA frames, 236 TDWZ, 112, 113, 131, 137 187, 188 Temporal filtering, 40, 51, 53, 67, 96 SN DATA PDU, 410 Temporal redundancy, 6, 7 SNDC header, 254 TFCI, 260, 261, 270, 278, 285, 317, 406 SNDCP, 232, 233, 410, 411 TFI, 260, 261 SNR, 28, 43, 70, 85, 87, 139, 142, 143, Thermal noise power, 388 Time Context, 437, 476 275, 312, 379, 380, 381, 382, 383, TM5, 154, 294, 318, 325, 353, 373, 421, 508 490, 500 TMN8, 154 SN UNITDATA PDU, 410 Traffic model, 219, 385 SOAP, 465, 523 Training Sequence Codes, 241, 405 SOCAM, 442, 443, 444 Transcoding, 100, 153, 156, 171, 186 Soft QoS control, 399 Transmission Time Interval, 265, 286 SoftwarePlatform, 454 Transmit bit energy, 354, 379, 381, 383 SOP, 91 TransMux layer, 220 Source sequences, 37 TTI, 261, 264, 265, 267, 284, 286, 295, 296, Spaces for adaptation decision, 462 Spatial power spectral density, 115 317, 323, 347, 379, 395, 417, 420 Spatial redundancy, 6, 7, 8, 9, 13, 200 TU, 238, 241, 246, 336, 337, 341, 405, 422 Spatio Temporal, 159 TU3, 241, 242, 245, 361, 363, 365, 366 SPECK algorithm, 51, 52 Turbo coding, 104, 107, 286 Speech compression, 221 Turbo Encoder, 104 SPIHT, 69 Spreading Factor, 285, 286, 326, 380, U 383, 406 UAProf specifications, 456 SQBAS, 363, 365, 367, 368, 369, 370, 371, 384 UCD, 446, 448, 450, 451, 453, 464 SR ARQ, 223 UCDs, 453, 465 SSD, 19 UED, 446, 448, 449, 450, 451, 453, 464, 470, SSIM, 34 Standardization Bodies, 27 472, 473, 482, 486, 487, 489, 523 State of the art, 96, 131, 155, 193, 530, 531 UED ral environment characteristics, 450, 472 State space model, 116, 118 UEP, 27, 58, 90, 91, 92, 93, 96, 187, 222, 347, STC, 302 Stereo Video Sequences, 126 351, 352, 353, 354, 355, 356, 381, 382, 383, Stereoscopic, 44, 56, 57, 58, 60 384, 502, 507 Stereoscopic 3D video, 44 UF, 466 Stereoscopic content, 44 UMA, 53, 453, 457, 460, 470 Stereoscopic Video Conversion, 471 UMA scenario, 460 Streaming, 143, 371 UMTS, 2, 3, 37, 211, 212, 215, 216, 219, 221, SVC, 1, 33, 36, 39, 41, 44, 59, 60, 61, 63, 71, 73, 226, 231, 256, 257, 260, 262, 263, 264, 272, 74, 79, 80, 81, 82, 88, 95, 96, 468, 469, 520, 279, 280, 282, 293, 296, 297, 299, 312, 315, 521, 522 316, 321, 329, 330, 331, 332, 341, 342, 344, SVO encoding, 160, 162 345, 346, 347, 352, 353, 354, 355, 356, 357,

Index 567 378, 379, 380, 381, 382, 383, 384, 385, 395, ViTooKi operating system, 453 401, 403, 404, 405, 406, 407, 408, 409, 410, VLSI, 2, 211 413, 416, 417, 421, 424, 425, 426, 427, 428, VO, 2, 47, 49, 52, 152, 158, 160, 162, 163, 429, 430, 431, 432, 502, 507, 508, 509 UMTS DL Model Verification, 279, 280 164, 165, 166, 168, 170, 173, 174, 175, 176, UMTS emulator, 296, 297, 401, 405, 407, 410, 177, 178, 179, 180, 182, 183, 186, 225, 348, 416, 421, 431 466, 506 UMTS FDD simulator, 297 VOP, 159, 161, 163, 164, 165, 166, 168, 169, Unequal Error Protection, 27, 93 175, 176, 178, 324, 331, 332, 348, 349, 506 User Agent Profile, 454 VOP header, 324, 332, 348, 349, 506 User Context, 437, 438, 476 VS, 152, 157, 160 User Interface, 417 USF bits, 235, 236 W Utility space vehicle, 461, 464 W3C, 438, 443, 445, 454, 455, 456, 458 UTRAN, 221, 256, 257, 258, 259, 260, 271, WAP, 454, 455 285, 292, 295, 296, 300, 315, 321, 322, 325, Wavelet, 36, 40, 41, 44, 65, 66, 67, 68, 69, 71, 329, 330, 341, 342, 344, 385, 395, 403, 508 UTRAN emulator, 300, 325, 344 91, 96 Wavelet analysis, 40, 41 V Wavelet transform, 40, 44, 65, 69 VBV, 161, 163, 164, 167, 168 WCDMA, 91, 221, 231, 257, 258, 290, 355, VCEG, 1, 36, 41, 161, 192 VCS, 477, 478, 479, 480 378, 392, 395, 409, 508 VERT, 73 WDP, 454 Vertex based shape coding, 46, 47, 48 Web Ontology Language, 531 Video Buffering Verifier Control, 160, 167 Web Services technologies, 439 Video Buffering Verifier Control WiMAX, 3, 38, 212, 213, 215, 216, Compensation, 167 231, 300 Video Coding Expert Group, 36 Windows Media DRM 10, 459 Video object planes, 158, 175 Wireless Access Protocol, 454 Video objects, 2, 47, 49, 52, 152, 174, 175, 176, Wireless JPEG 2000, 88 Wireless PC cameras, 99 177, 178, 179, 180, 182, 183, 225, 466 WLAN, 3, 37, 212, 215, 216, 501 Video surveillance, 99 WML, 454 Video traffic model, 219 WMSA, 274, 299 Virtual Classroom, 478 WZ frames, 116, 118, 131, 132, 133, 134 Virtual Collaboration, 477, 480 Virtual Collaboration System, 38, 477 X Virtual Desk, 37 XDSL, 3, 213, 301 Virtual View Generation, 200 XML, 439, 440, 446, 449, 452, 455, 456, 459, VISNET I NoE project, 465 472, 473, 486 XrML, 459 XSLT, 452, 465