- On the other hand, uncontrolled reduction of the output bit rate of a video coder leads to unnecessary quality degradation and inefficient use of available bandwidth re- sources. - The status of the channel is reported back to the video source by a number of receivers that have special traffic data compilation capabilities. - 3.2 Bit Rate Variability of Video Coders. - Time Bit rate. - Bit rate. - Figure 3.1 Relationship between quality and bit rate. - output bit rate a variable function of time. - The number of uncoded MBs in predicted frames is certainly a function of the temporal correlations in the video content. - The variability of the number of coded MBs in predicted frames certainly leads to a variable output bit rate. - On the other hand, the spatial correlations between pixels of the same video frame dictate the number of bits required to encode the 64 transform coefficients of each 8 . - Table 3.1 lists the fixed and variable-length video parameters of the H.263 compression algorithm. - shows that most of the bits of an H.263 stream, for the Foreman sequence coded at 30 kbit/s, are due to the variable-length codes. - Therefore, in addition to buffering the video data, other measures need to be taken in order to reduce the burstiness of the output flow of video coders.. - The most commonly used technique is to adjust some video encoding par- ameters as a function of the buffer fullness, i.e. - In the next section, we describe the response of the video coder to feedback or picture activity feed-forward messages.. - The adjustment of the parameters is usually done in line with the channel status that is periodically reported to the video source. - The traditional approach to regulate the output bit rate of a video source is to adjust the quantisation step size of the next frame, GOB or MB, based on the local buffer occupancy that is essentially dictated by the status of the network. - ameters are updated based on the encoding results of the current frame. - Then, the rate control stage sets the reference value of the quantisation parameter for each MB by means of a virtual buffer. - Finally, the adaptive quantisation stage modulates the reference value according to the spatial activity in the MB to derive the value of the quantisation parameter used to quantise the MB. - In the target bit rate allocation stage, TM5 calculates the bit allocation of the next frame using the global complexity measure X. - R G : bit rate . - Before encoding MB j ( j P 1), the fullness of the appropriate virtual buffer is computed based on the picture type:. - The final fullness of the virtual buffer (d GH , d NH and d @H for j : MB-cnt) is used as d G , d N and d @ for encoding the next picture of the same type.. - bit rate. - H is the fullness of the appropriate virtual buffer. - Finally, TM5 finds the value of the quantisation parameter mquant. - As is obvious from the two rate control algorithms described above, the quan- tisation parameter of the current frame (MB or slice) is decided based on the number of bits taken by coding the previous frame. - In traditional variable-quantisation rate control algorithms, the quantisation parameter of the next frame is determined based on the number of bits generated by the previous frame. - In feed-forward rate control schemes, the quantisation parameter is determined based on the number of bits required to code the prediction error of the current frame, GOB or MB. - In feed-forward algorithms, an initial Qp value of a frame is selected based on Qp and the bit rate of the last coded frame. - With the selected Qp, the number of bits required to code the transform coefficients of the current difference frame is estimated using the prediction error per block and the bit per error curves of Figure 3.6. - This bit rate is then compared to the target bit rate per frame. - Qp is increased when the predicted bit rate is higher than the target bit rate of the frame, or vice versa.. - This bit rate control algorithm gives an accuracy of <. - The achieved bit rates of the feed-forward rate control scheme are very close to the pre-defined target rates and the quality degradation is minimal. - Figure 3.7 shows the variations in the output bit rate for the Foreman sequence using both rate control algorithms. - The efficiency of the feed-forward rate control algorithm can be seen in the smooth bit rate variations achieved in comparison with the fluctuat-. - Bit rate per frame (b/s). - A penalty to this rate control scheme is, as expected, a less stable perceptual video quality, as shown in the luminance PSNR values of the Foreman sequence in Figure 3.8. - Figure 3.8 Y-PSNR values of the Foreman sequence encoded at 20 kbit/s and 7.5 f/s using different rate control algorithms. - D for the face region at Qp 9 2 and that of the background Qp LD at Qp . - This simple rate control technique with ROI coding produces satisfactory results, as shown in Table 3.4 where 150 frames of the Miss America sequences are coded at various target bit rates. - Table 3.4 150 frames of the Miss America sequence encoded at different bit rates with two different rate control algorithms. - Figure 3.9 Frame of the Miss America sequence encoded at 14.4 kbit/s: (a) conventional variable-Qp TMN5 rate control, (b) ROI coding for enhanced-face rate control PSNR levels around the face without disturbance to the rate control efficiency.. - The subjective improvement achieved by this rate control algo- rithm is depicted in Figure 3.9 which shows a frame of the Miss America sequence encoded at 14.4 kbit/s using the traditional TMN5 and enhanced-face ROI rate control algorithms. - On the objective scales, Figures 3.10 and 3.11 show the number of bits per frame and luminance PSNR values, respectively, for 150 frames of the sequence encoded at 20 kbit/s.. - Although the above ROI rate control technique achieves its objective in enhanc- ing the perceptual quality of the region of interest in the sequence while maintain- ing the resultant bit rate close to the target value, it is still in need of improvement, since the bit rate per frame is still highly fluctuating, as shown in Figure 3.10. - Initially, the minimum size of the gap between the two step sizes is set to g, i.e.. - Figure 3.11 Y-PSNR for the facial region of 150 frames of the Miss America sequence coded at 20 kbit/s using conventional and ROI rate control algorithms. - Bit rate per frame (bits). - Figure 3.13 shows the luminance PSNR values of the Foreman sequence using three different rate control algo- rithms. - It can be seen that the overall PSNR of the stabilised rate control algorithm is generally always higher than that of ROI coding, as shown in Figure 3.13(a). - However, the PSNR values around the face are always better than those achieved by a stabilised bit rate algorithm due to the enhanced face coding of the algorithm, as shown in Figure 3.13(b).. - Each VLC represents a particular piece of information related to the temporal and spatial details of the video sequence.. - Figure 3.13 Y-PSNR values of the Foreman sequence coded at 48 kbit/s using three different rate control algorithms: (a) overall PSNR (b) facial region PSNR. - These video codewords are ordered according to the syntax of the video coding algorithm and then sent to a local buffer before transmission. - This prompts the decoder to skip the information of the current frame until the next error-free synch word in the bit stream. - Figures 3.14 and 3.15 show the subjective and objective effects, respectively, of dropping INTER MBs off the local buffer of the video encoder to simulate the effect of a network congestion. - It is calculated as the ratio of the number of dropped bits over the total number of bits in the coded video stream. - On the other hand, transform coefficients occupy a larger portion of the coded bit stream, as indicated in Section 3.2. - therefore, dropping these coefficients in the case of congestion helps reduce the flow rate of the video encoder with only a graceful degradation of the reconstructed video quality. - Figure 3.16 Frame 150 of the Suzie sequence encoded with H.263 at 55 kbit/s and subject to 5 per cent information drop: (a) on the sub-stream of DCT coefficients of a P-frame, (b) on the MV sub-stream. - Figure 3.18 depicts a 3-D reference model that illustrates the dependency of the dynamic prioritisation scheme on the time-varying channel conditions and the sensitivity of video parameters to information loss.. - To regulate its output bit rate, the encoder makes an estimate of the number of bits it has to drop by analysing the statistics data of the channel feedback reports.. - Therefore, the bit rate is controlled without adjusting any video encoding parameter, thereby achieving a consistent quantisation process regardless of the network state.. - A mismatch between these two frames leads to the accumulative damage of the decoded video frames.. - The block diagram of the feedback-controlled H.263 video encoder is depicted in Figure 3.19.. - However, the negoti- able options contribute to the reduction of the output bit rate, thereby ensuring a better coder performance in congestion control terms. - The higher bit rate of the feedback-loop encoder compared to the ordinary H.263 one is mainly due to the updated content of the locally reconstructed picture memory. - Traditional variable-quantisation rate control algorithms proved efficient in re- ducing the fluctuations of the output bit rate of a video encoder. - Y-PSNR P(B) Cb PSNR P(B) Cr PSNR P(B) Total bit rate. - Figure 3.21 Number of bits per frame for 150 frames of the Miss America sequence using the feedback-loop rate control mechanism. - Figure 3.23 Up-sampling of the reconstructed block prediction error: (a) inside a block, (b) at a block boundary. - Bit rate per frame. - Figure 3.25 Bit rate per frame for 150 frames of the Silent Voice sequence coded at 40 kbit/s and 10 f/s. - Figure 3.25 shows the bit rate per frame for 150 frames of the sequence Silent Voice. - Figure 3.26 Y-PSNR values for 150 frames of the Silent Voice sequence coded at 40 kbit/s and 10 f/s. - The background of the sequence is also very detailed. - This happens when the original coder starts to run out of bits due to overspending of the available bit rate at the beginning of the sequence. - On the subjective scale, the reconstructed sequence of the reduced resolution decoder shows some blurriness due to the up-sampling process. - The first frame of the sequence is INTRA coded with a quantisation parameter INTRAQ set to 15.. - Let B be the number of bits spent on the previous encoded frame, R the target bit rate per second, G the frame rate of the original video sequence (f/s), F the target frame rate (f/s), and M the threshold for frame skipping (M : R/G).. - The next ‘‘frame — skip’’ frames of the original video sequence are skipped and the target bit rate for the next frame is B : R/F. - Figure 3.28 Y-PSNR values of the Silent Voice sequence coded at 40 kbit/s using variable frame rate control for both the original H.263 coder and the modified coder with reduced resolution mode. - The use of the reduced resolution controller helps reduce the output bit rate of the modified coder, whilst the original TMN5 fails to reduce its bit rate since the estimated Qp has already attained its maximum (i.e. - As a result, the decoded sequence becomes jerkier, whereas the same frame rate could be maintained using the reduced resolution mode of the modified H.263. - The enhancement layers could then be used as a trade-off between quality and compression efficiency in order to control the output bit rate of the video coder.. - The base layer consists of a low quality image generated at a very low bit rate while the second layer is the difference between the input original picture and the output of the base layer coder. - The base and enhancement layers could be encoded at different VOP rates to achieve the required temporal scaleability of the coded output.. - Y PSNR bit rate Y PSNR bit rate. - This is mainly due to the additional bit overhead required to code the administrative data of the enhancement layers.. - In this case, the multi-layer coder has to adjust the quantisation parameter, spatial and temporal resolutions of the base layer and the number of coded VOPs in the video scene in accordance with the output bit rate. - The BL represents the minimal amount of data required for the reconstruction of the video sequence at the decoder side. - The ELs represent additional information that contributes, if correctly received and decoded, to the enhancement of the decoded video quality. - The granularity is achieved by decreasing the amount of prediction in the pictures of the enhancement layer thereby reducing the coding efficiency of the video coding scheme. - In the former case, the overall frame rate of the transmitted video is equal to that of the base layer regardless of the available bandwidth.. - Moreover, the video encoding parameters, such as the quantisation step size Qp, the motion threshold and the use of variable-length codes, are all factors that contribute to the variability in the output bit rate of the video encoder. - However, when the band- width requirements of the network are time-varying, the output bit rate of a video coder must be adapted to the changing bandwidth conditions at any instant of time. - Furthermore, rate control could be achieved by down-scaling/upscaling the resolution of the video content in accordance with the bandwidth requirements. - of the Packet Video Workshop 2000, Sardinia, Italy, May 2000.
Xem thử không khả dụng, vui lòng xem tại trang nguồn hoặc xem
Tóm tắt