MXPA01001945A - Dynamic bit allocation for statistical multiplexing of compressed and uncompressed digital video signals - Google Patents

Dynamic bit allocation for statistical multiplexing of compressed and uncompressed digital video signals

Info

Publication number
MXPA01001945A
MXPA01001945A MXPA/A/2001/001945A MXPA01001945A MXPA01001945A MX PA01001945 A MXPA01001945 A MX PA01001945A MX PA01001945 A MXPA01001945 A MX PA01001945A MX PA01001945 A MXPA01001945 A MX PA01001945A
Authority
MX
Mexico
Prior art keywords
bits
images
image
superframe
gop
Prior art date
Application number
MXPA/A/2001/001945A
Other languages
Spanish (es)
Inventor
Limin Wang
Ajay Luthra
Original Assignee
General Instrument Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by General Instrument Corporation filed Critical General Instrument Corporation
Publication of MXPA01001945A publication Critical patent/MXPA01001945A/en

Links

Abstract

The present invention relates to a method and apparatus for allocating bits in a statistical multiplexing system (stat mux). A statistical multiplexer (stat mux) (600) accommodates both compressed and uncompressed video programs using transcoding (640, 650) and encoding (620, 630), respectively. Hierarchical dynamic bit allocation is used, starting from a super GOP level (700, 702, 704, 800, 900), then to a super frame level (902;904;908;909), and then to the regular (individual) frame level (910, 920, 930, 990;912, 922, 932, 992;916, 926, 936, 996;918, 928, 938, 998). At each level, a target number of bits (T1, T2, TL-1, TL) is determined. A target number of bits for a super frame, which is a collection of frames across all channels at a given frame instance (t1, 2t1, 3t1...), is adaptive and is able to address any combination of picture types. Frames of the same picture type for a program are generallyassigned the same (or similar) number of bits. Relative program quality can be controlled using a program priority weighting factor (w). Additionally, constraints on target bit rates and minimum and maximum bit rates are provided.

Description

DISTRIBUTION OF DYNAMIC BIT FOR ULTIP EXACION STATISTICS OF DIGITAL VIDEO SIGNALS COMPRESSED AND NOT COMPRESSED FIELD OF THE INVENTION The present invention relates to a method and apparatus for distributing bits in a system of statistical multiplexing. In particular, an architecture is described for the statistical multiplexing (stat mux) of both compressed and uncompressed video signals. Dynamic bit distribution and rate control are provided. In addition, the upper and lower program bit rates are specified to prevent the encoder and decoder buffers from overflowing or sub-flowing.
BACKGROUND OF THE INVENTION With recent advances in digital video compression, such as those used in the MPEG-2 standard, and digital data transmission techniques, it is possible to send several compressed video programs diqi tally in the same width of band currently occupied by a single analog television (TV) channel. These facilities provide opportunities for programming service providers (e.g., television broadcasters such as CNN, ABC), communications network operators (e.g., owners of cable and satellite networks), and end users. In a multiple program transmission environment, several programs (for example, channels) are coded, multiplexed and transmitted through a single communications channel. Since these programs share a limited channel capacity, the aggregate bit rate of the programs should not be greater than the communication channel rate. This can be achieved by controlling either each individual program bit rate using independent coding, or the aggregate bitrate using statistical multiplexing, also known as associated coding. A statistical multiplexer is referred to herein as "stat mux", whereas the statistical multiplexing is referred to as "stat muxing". With independent coding, the control rate can only be realized through time and the spatial dimensions of a program. However, in the statistical multiplexing or associated coding, the control is extended to an additional dimension; that is, the dimension of the program. As a result, there is greater freedom to distribute channel capacity between programs and therefore more image quality control between programs as well as within a program. However, such systems generally process an image at a time from each channel (e.g., in a common frame instance), and do not represent the Image Group (GOP) settings of the image type data streams. A GOP is a group of one or more consecutive images. For example, a GOP may contain internally encoded images (I images), predictive encoded images (P images) and / or bidirectional predictive encoded images (B images). Different channels can have different GOP configurations and lengths. A GOP may also consist of progressively refreshed images where there are no I images. However, in a P image, a portion, for example, a slice of the image is coded as blocks. The distribution of blocks I changes from one picture P to another. In addition, video materials such as movies and the like can be precompressed and stored for subsequent transmission. These precompressed videos can be encoded at a constant bit rate (CBR) or at a variable bit rate (VBR). This presents difficulties when the statistical multiplexer tries to integrate the bit streams of the precompressed program with the uncompressed, uncompressed digital video streams. Accordingly, it would be desirable to have a statistical multiplexer system that is capable of handling precompressed data that is found either at a constant bit rate (CBR) or at a variable bit rate (VBR), along with uncompressed video data. The statistical multiplexer system should use the GOP structure of the video channels to provide an efficient bit distribution technique. The statistical multiplexer system must also represent the type of image in each GOP to distribute the bits. The statistical multiplexer system must assign the same or similar number of bits to frames of the same type of image for continuous scenes. The statistical multiplexer system must also represent a relative priority of the channels, as well as the level of complexity of each frame. The statistical multiplexer system must be compatible with existing digital video standards such as MPEG-2. The statistical multiplexer system must prevent the encoder or decoder from overflowing or sub-flowing. The statistical multiplexer system must provide restrictions on the target bit rates, which include restrictions on the general minimum and maximum bit rates. In addition, for super GOP and superframe bit distribution schemes, the statistical multiplexer system must provide target bit rates, and restrictions on target bit rates, for super GOPs, superframes, and regular frames, as well as also restrictions on the general minimum and maximum bit rates. The present invention provides a system having the above advantages and still others.
BRIEF DESCRIPTION OF THE INVENTION The present invention relates to a method and apparatus for distributing bits in a statistical multiplexer system. The invention provides a statistical multiplexer that accommodates both precompressed and uncompressed video programs using transcoding and coding, respectively. In addition, the hierarchical dynamic bit distribution is used, starting from a super GOP level, then to a superframe level, and then to the regular (individual) frame level. The concept may further extend to a sub-frame level, where the bits are distributed for a portion of a frame such as a slice, or for a plane of video objects (VOP) as described in the MPEG-4 standard, for example. In each hierarchical layer, an objective number of bits is determined. An objective number of bits (Tn) for a superframe n, which is a collection of frames across all channels in a given frame instance n, is adaptive and represents any combination of image types. Furthermore, although it is not necessary, it is desirable to assign the same (or similar) number of bits to the frames of the same type of image for a program (for continuous scenes). To achieve this adaptation in the bit distribution, the invention provides a dynamic bit distribution strategy that determines an objective number of bits for each program on a frame-by-frame basis according to the above coding information, such as the parameters of quantification used, and the resulting number of bits. Furthermore, in order to prevent both the encoder and decoder buffers from having overflow or underflow, restrictions are imposed on the compressed bit rate of each program in the multi-program transmission environment. In addition, the quality of the program can be controlled during statistical multiplexing according to a program priority weighting factor. In addition, target bit rates, and restrictions on target bit rates, are provided for super GOPS, superframes, and regular frames. Restrictions on the general minimum and maximum bit rates are also provided. A particular bit distribution method for digital video according to the present invention processes a plurality L of video programs (eg, channels) in a coder, where each program has successive groups of images (GOPs). Each group of images has an associated number of images, typically 10.-20. The term "image" refers to a plot or a field. A "super group of images" is provided which comprises at least one group of images from each of the video programs L, and which have a length of N images. A first target number of bits, T, is calculated to encode the super group of images according to the number of images in the super group of images, L * N, and an available capacity of a channel on which the programs are transmitted of video, such as a cable television network or a satellite television network. In addition, each super group of images comprises a plurality N of "supert branches", each superframe having L images at a common temporal reference point. A second target number of bits, Tn, is computed to encode each nes? M superframe of images, where n = l, ..., N, according to the first objective number of bits, T, and a measure of complexity of every 2? ma image in the n?? ma superframe, where 1 = 1,. . . , L. Is a third objective number of bits, T?, N, calculated to code each day? a image in the n th associated superframe according to the second objective bit number and the complexity measurement, and in inverse proportion to the sum of the complexity measurements for each image in the superframe néslma. The long N of the super GOP is preferably a least common multiple of the associated number of images in each of the groups of images. For example, the least common multiple for GOPs with respective lengths of nine and fifteen frames is forty-five. When different types of images are provided in at least one of the groups of images, the method comprises the additional steps to: provide respective different weighting factors K for the different types of images; and calculating the third target number of bits to encode each? e s? ma image in the associated superframe weight according to the respective weighting factor thereof. For example, different types of images may include I images, P images, and / or B images. In some cases, a time limit of a group of images of at least one of the programs is misaligned with a time limit of the super. group of images in such a way that the super group of images comprises a fractional portion of the misaligned group of images. But, the super GOP lengths are always multiple of the program's GOP lengths. The invention can be used even when the program GOPs are asynchronized with each other and / or a super GOP. The method may include the additional steps to provide respective weighting factors, w, for the different video programs according to a relative priority thereof, and calculate the third target number of bits to encode each lousy image in? ü is the associated superframe according to the respective weighting factor of the associated video program. Preferably, the same complexity measurement is used for each image with a common image type (e.g., I, P or B) in at least one of the video programs of the super group of images to calculate the second and second bit numbers. third. In particular, the method may include the additional steps to: define the respective complexity measurements for each type of image in each program, and use the respective complexity measurements to calculate the second and third bit numbers. In addition, the respective complexity measurements can be updated after encoding each image. The method may include the additional steps of: calculating a remaining number of available bits, Tr, to encode a remainder of the superframes not yet encoded in the super group of images after encoding the 1 images in one of the superframes, and encoding each remaining superframe not yet encoded in proportion to the remaining number of available bits, Tr. A buffer associated with the encoder receives the encoded data from the video programs. According to the above, it is important to keep the level of the encoder buffer between a minimum and maximum level. Accordingly, the method may include the additional steps to adjust the second associated bit number, if necessary, to prevent it from falling below a minimum level, -I R tunal, h, pl,) - J fí-J n1 -i, r before encoding the nési? Na superframe with the second associated bit number of the bits. O , ,, is a number of average bits per image transmitted through the channel; and η is a fullness level of the buffer after the previous superframe has been coded (eg, ((n - 1) th).) For the maximum encoder buffer level, the method may include the additional step to set the second associated bit target number, if necessary, to avoid exceeding the ± / J? ' mux -? fíJ'-l,, 'before encoding the nésl? na superframe with the second target number of associated bits. Here, ßL ", is a maximum capacity of the buffer. In addition, video programs are transmitted through a channel to a decoder, so it is important to keep the buffer level of a decoder within acceptable limits.
Before encoding the? Th image in the Oési to superframe with the associated third target bit number, the method may include the additional step to adjust the associated third target bit number, if necessary, to avoid that exceeds a maximum level,? R ?. -Big T; in where Yes? / «'Is ^ a sum of the number of transmitted bits for níslma up to (n + N ') these images for the bad video program; N 'is a decoder decoding delay; and Be?, n-? is a fullness level of the buffer after the ith image has been detected in the (n - l) s ima superframe An.tes to encode the 2nd image in the super7 r7 picture with the third target number of In the case of bits, the method may include additional steps to adjust the third target number of bits, if necessary, to prevent it from falling below a minimum level, n + N '2-j J? .l ,,, - £ > l, n- \ JtS? where ? i? ", 'is ^ a sum of the number of bits transmitted n' = n by the nth to (n + N ') th images for the bad video program; N' is a decoder decoding delay; ?, n-? is a fullness level of the buffer after the - th image in the n - l) superframe has been detected; B: is the maximum capacity of the decoder's buffer. In addition, in many cases it is desirable to keep the bit rate for each program within the predetermined limits. For a minimum rate, the method comprises the additional steps to: determine a minimum average number of Rmin bits to encode N > 1 images; and before encoding the image in the bad hyperframe with the third associated bit number, if necessary, to prevent it from falling below where 2jRl, p 'is the sum of the number of bits rí = nN "transmitted for the (nN' ') s ima until (nl) és? M images for the? ß s: Lm ° video program For a maximum rate, the method comprises the additional steps for: determining a maximum average number of bits Rm? to code N ''> 1 images, and before encoding the image in the bad superframe with the third associated bit number, adjust the third associated bit number, if necessary, to avoid exceeding where nl 2-yRl.n 'is the sum of the number of bits n'-nN "transmitted for the (n-N' ') th to (nl) these images for the S 2-th video program. In a particular application of the invention, the video programs are adapted for communication through a broadband communications network to a decoder population, the method can include the step to transcode the pre-compressed video bit stream of A particular video program of the plurality L of video programs in another bit stream, wherein the pre-compressed video bit stream is provided at a different bit rate after transcoding .This transcoding process allows the use of both of uncompressed video source data as precompressed in a statistical multiplexer Another method of the present invention is presented for encoding non-compressed video source data, and transcoding the source data of precompressed video. The method includes the steps to: partially decompress the precompressed video source data to obtain the corresponding video data partially uncompressed; distributing the bits to encode the uncompressed video source data according to a scheme of many statistical typing; and distributing the bits to transcode the video data partially uncompressed according to the scheme of many statistical typing. The precompressed image data is transcoded in such a way that a bit rate of the precompressed image data is different than a bit rate provided by the associated distributed bits. The structures of the corresponding apparatuses are also presented.
BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 illustrates a coding and decoding system according to the present invention. Figure 2 illustrates an MPEG encoder for use with uncompressed video data in accordance with the present invention. Figure 3 illustrates a transcoder according to the present invention. Figure 4 illustrates a cascaded MPEG decoder / encoder for use with precompressed video data in accordance with the present invention.
Figure 5 illustrates a simplified transcoder in accordance with the present invention. Figure 6 illustrates a statistical multiplexing coder according to the present invention. Figure 7 illustrates a super GOP construction with program GOPs aligned in accordance with the present invention. Figure 8 illustrates a super GOP construction with non-aligned program GOPs in accordance with the present invention. Figure 9 illustrates a superframe construction according to the present invention. Figure 10 illustrates a decoder according to the present invention. Figure 11 illustrates the data stored in an encoder buffer according to the present invention.
DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method and apparatus for distributing bits in a statistical system.
Figure 1 illustrates a coding and decoding system according to the present invention. One problem with precompressed materials is that they could be precoded at any bit rate, either a constant bit rate (CBR) or a variable bit rate.
(VBR) Therefore, to accommodate the bit streams of the pre-compressed programs in a statistical multiplexer system, the corresponding rates have to be variable. Transcoding is the operation of converting one bit stream into another bit stream at a different rate. In Figure 1, encoders 105 and 110 are provided, along with the transcoders 115 and 120. At a transmission site 100, such as a hub, the encoders 105 and 110 receive the uncompressed digital video data, for example. , in the domain of the pixels, coming from respective programs or video channels, program 1 and program 2. The transcoders 115 and 120 receive precompressed digital video data from respective compressed programs (for example, bit streams) Ll and L L is the total number of channels or programs in the encoder. The encoded data is provided to a MUX 125 for transmission over a channel, such as a cable television network or a satellite television transmission network, using for example, conventional multiplexing techniques by time and / or multiplexing. by frequency. In a decoding site 150, the channels are collapsed? multiplexed in a DEMUX 155 and can be provided to the decoders 160, 165, 170, 175. Typically, currently only one channel is decoded because the viewer generally sees only one channel at a time. However, split-screen and image-in-picture techniques allow the viewer to see two or more images at the same time, in which case each channel must be decoded. In addition, the decoders shown can be independent or implemented using common hardware. For illustrative purposes, decoders 160 and 165 provide the decoded data for programs 1 and 2, respectively, while decoders 170 and 175 provide the decoded data for programs L-1 and L, respectively. The encoders 105 and 110 may be provided as described in Figure 2, while the transcoders 115 and 120 may be provided as described in Figures 3-5. Figure 2 illustrates an MPEG encoder for use with uncompressed video data in accordance with the present invention. In the encoder 200, a frame decomposition processor 205 decomposes an input video frame into segments such as slices and macroblocks. The pixel data is then provided to an intra / inter mode switch 210, a summer 215, and a Motion Estimation (ME) 220 function. The switch 210 selects either the current pixel data, or the difference between the current pixel data and pixel data from a previous frame, to be processed by a Discrete Cosine Transform (DCT) 222 function, quantizer 225, and a Long Variable Coding (VLC) function 230. The output of the VLC 230 function is a bit stream that is transmitted to a decoder. The bitstream includes data from the Movement Vector (MV) from the ME 220 function. In a feedback path, the inverse quantization is performed in a function 235 and an inverse DCT in a function 240 to recover the data of the domain of the pixels. These data are summed with the motion compensated data or a null signal in an adder 245, and the sum thereof is given to a current Frame Intermediate Memory (FB). The data from the current FB 250 and a previous FB 255 are provided to the ME function and to a Motion Compensation (MC) 260 function. A switch 265 directs either a null signal or the output of the MC function. 260 to the adder 245 in response to the intra / inter mode switch control signal. Figure 3 illustrates a transcoder according to the present invention. The uncompressed video is compressed in a coder 305 at a quantization level (e.g., step size) Qi to produce data at a bit rate Rx. These data, which can be stored in a storage medium for subsequent recovery, are provided to a transcoder 310, which comprises a cascaded decoder 315 and an encoder 317. The data is decoded in the decoder 315 to retrieve the video from input, or a close approximation to it, due to the lossy nature of the encoding in the encoder 305. The decoded or reconstructed data is then encoded in the encoder 317 at a different quantization level Q2 to produce data at a rate of bits R2. Commonly, R2 = R ?. For example, R = 50 Mbps and R2 = 3 Mbps. The data encoded at a rate R2 is then communicated through a channel to a decoder 320 to provide the decoded video output, for example, for display on a television. The Figure illustrates a cascaded MPEG decoder / encoder for use with the precompressed video data in accordance with the present invention. The cascaded MPEG decoder / encoder includes a decoder 400 and an encoder 450. The similarly numbered elements of the encoder 450 correspond to the encoder of Figure 2. A compressed video bit stream is the input to a Variable Long Decoder.
(VLD) 410, which is a counterpart of the VLC 330. An inverse quantizer function 420 processes the output of the VLD 410 using a first quantization step size, Qi. An Inverse DCT function (IDCT) 430 processes the output of the inverse quantizer 420 to provide the domain data of the pixels to an adder 440. This data is summed with either a motion compensation difference signal from the MC 455 or with a null signal, according to the position of a switch 460. The output of the adder 440 is provided to the encoder 450 and a current FB 480 of the decoder 400. The MC 455 function uses the data from the current FB 480 and a previous FB 470 together with the MV data from the VLD 410. The bit output rate of the transcoder 450 is adjusted by changing Q2. Figure 5 illustrates a simplified transcoder in accordance with the present invention. The similarly numbered elements of the transcoder 500 correspond to those in the transcoder of Figure 4. The performance of the simplified transcoder is very close to that of the cascaded transcoder of Figure 4. The transcoder saves processing and hardware steps by using only a current FB (Intermediate Memory) 480 and previous FB 470, and a DCT function 510 and an IDCT function 520. An adder 530 provides a transform domain difference signal to the IDCT 520. A switch 540 selects either the signal of transform domain from the DCT 510 function or a null signal according to an intra / inter mode control signal. Figure 6 illustrates a statistical multiplexing system according to the present invention. The inputs to the encoder 600 may include uncompressed digital video sequences and / or precompressed bit streams. The uncompressed digital video streams, for example, programs 1 and 2, are encoded by the MPEG encoders 620 and 630, respectively, for example. L-1 and L are processed by transcoders 640 and 650, respectively. The simplified transcoder configuration of Figure 5 can be used. The encoded data is provided to a MUX 660 and to the encoder buffer 670 before being transmitted in a channel. The precompressed program data can be recovered from a storage means 645, such as magnetic tape or compact disc, or can be received in real time, for example, from a satellite transmission. The encoder buffer 670 sends a fullness level signal to the rate control processor 610. A user interface 608 may communicate with the rate control processor 610, for example to provide information regarding the length of the GOP and the types of images in the different program flows. The basic requirement in the statistical multiplexer encoder is to provide a relatively uniform image quality within a program, and if necessary, also between the programs. To achieve this goal, channel capacity is distributed dynamically between programs according to a program priority as well as a frame level program complexity measurement. Each MPEG encoder 620, 630 or transcoder, 640, 650 receives an objective number of bits, i, T2, TL-i / and TL, respectively, from a rate control processor 610 in each frame. The rate control processor 610 includes a super GOP level processing function 606, a super-frame level processing function 604, a frame level processing function 602, and a complexity processor 605. These functions of processing can share common hardware such as integrated processing circuits and memories, but they are shown individually for simplicity. The target number of bits for each frame of a program is fulfilled by adjusting the quantization parameter in the MPEG encoder or transcoder. The resulting number of compressed bits, R, as well as the average quantization parameter, Q, used for each frame is then sent to the rate control processor 610. Specifically, the encoder 620, the encoder 630, the transcoder 640 and the transcoder 650 produce the bits Ri, R2, RL-I and RL respectively, using quantization parameters Qi, Q2, QL-I and QLT-respectively. The complexity processor 605 calculates the respective complexity values C using R and Q for each program. The rate control processor 610 then determines a new target number of bits for each new program or image frame based on the complexity of the program at the frame level. Next, the bit distribution procedure is described in detail. Additional restrictions on the target number of bits determined by the rate control processor 610 are also described in order to prevent the overflow and underflow of the buffer from occurring in the buffers of the encoder and decoder. Figure 7 illustrates a super GOP construction according to the present invention. To summarize the problem, the L video programs (for example, channels or programming services) need to be sent over a network with a fixed channel rate, Rcanai - The L programs can use either precompressed program bit streams, or uncompressed digital video programs. In addition, the L programs can use any GOP structure, or long program GOP. Additionally, the distance between the I or P images may be different for the different programs. The GOP length for each program is considered available at the coding site.
Super GOP and Target Bit Rate The dynamic bit distribution system is hierarchical. At the top level, a "super GOP" is provided. The input programs are divided into super GOPs that have the same number of I, P or B images, and therefore each super GOP is assigned the same number of bits. A "superframe" is then provided in each frame instance as a collection of frames, namely a frame coming from each of the programs at the same instant in time. The bit distribution for superframes is based on a program complexity measurement. At a frame level, each regular frame (image) receives an objective number of bits proportional to its complexity measurement. To ensure that the encoder and decoder buffers never exhibit overflow and subflow, and to limit each individual bit rate within a specific range, restrictions are applied to the target number of bits for the superframe as well as for each image. The L programs shown in Figure 7, for example, programs 1, 2, ..., L, are conceptually divided into identical groups, designated super GO P (L, N), in terms of the number of frames of each type of image so that the same number of bits can be assigned to each super GOP. The GOPs of a program do not have to be aligned with the super GOPs, as described in connection with Figure 8, below. That is, a super GOP limit can be cut through a program GOP. But, in any case, each super GOP contains the same number of frames of each type of image coming from a program. For example, each super GOP can include two I images, eight P images and twenty B images from a specific program. The determination of a super GOP can be done using a prior knowledge of the GOP length for each program. An operator can have the data relevant to the encoder input using an interface such as a keyboard. Or, it is possible to preprocess the different program streams in an encoder using an appropriate buffer capacity to determine the GOP length. For example, in Figure 7, a first super GOP 700 includes the data frames from program 1 (710), program 2 (720), ..., and program L (790). For example, each program segment, 710, 720, and 790, contains a number of complete frames from one or more GOPs as described below. A second super GOP 702 includes data frames from program 1 (712), program 2 (722), ..., and program L (792). A third super GOP 704 includes data frames from program 1 (714), program 2 (724), ..., and program L (794). In Figure 7, the limits of the program GOPs do not have to match the limits of the super GOPs. Figure 8 illustrates a super GOP construction with non-aligned program GOPs in accordance with the present invention. Here, the GOPs in each program 1, 2, 3, 4, ..., L are not aligned with the left and right limits of the super GOP 800. For example, the GOPs for program 1 (GOPí) are aligned with the limits of the super GOP 800. However, the GOPs for programs 2, 3, 4, ..., L, for example, GOP2, GOP3, GOP4, ..., and GOPL, respectively, are not aligned with the limits of the super GOP 800. Each super GOP includes at least one GOP from each program, and may also include fractional portions of the GOPs of each program. However, note that the fractional portions of the first and last GOPs for programs included in a super GOP can now be considered as a full GOP, as shown in Figure 8. In other words, a super GOP always contains a whole number of GOPs of each program. The length of super GOP is multiple of program GOP lengths. In addition, it is assumed that the distribution of the types of images in the GOPs for programs is the same. Here, L is the number of programs and N is the length of each super GOP. Each super GO P (, N) contains L * N frames. In addition, there may be many different Ns that can make super GO P s (L, N) identical in terms of the number of images of each type. However, from a deployment point of view, a small super GOP is preferred. Let Nl r 1 = 1,2, ..., L, the length of GOP for program 1.
N is set equal to the least common multiple (L.C.M.) of Nlr 1 = 1,2, ..., L, that is, (1) N = L. C.M. (? IJ, N2, ... N) N, defined in equation (1), is the smallest number which can be divided by all Nj. , 1 = 1,2, ...,. Therefore, the super GOPs (L,?) Are the smallest identical groups that contain the same number of frames of each type of image coming from each program. For example, considering only two different GOP lengths for? programs, say nine and fifteen, then the length of super GOP? = 45 (because 9 * 15 = 135, the smallest integer that divides 135 is 3, and 135/3 = 45). Note that if all programs 1, 2, 3, ..., L have the same length of GOP, N, the length of super GOP will be equal to N, regardless of whether the program GOPs have the same patterns of I, P, and B, or if they find synchronized. Because the super GOPs (L, N) with defined N when using equation (1) contain the same number of images I, P and B, the same number of bits, T, is assigned to each super GOP, ie , TE where Rcanai is in bits / sec, and frame_rate is in frames / sec. Analogous to the superframe concept of the Figure 9 below, the frame in each GOP program intersected by the super GOP limit is selected as a super GOP limit frame.
Superframe and Target Rate Figure 9 illustrates a superframe construction in accordance with the present invention. Given the objective number of bits, T, for a super group (L, N) 900, the next step is to determine the distribution of T through the super GOP frames. A "superframe" is defined as a collection of L frames, one from each of the L programs taken at the same moment in time, or common temporary reference point. For example, at time ti, a superframe 902 includes frames 910, 920, 930, ..., and 990. Similarly, at time 2tx, a superframe 904 includes frames 912, 922, 932, .. ., and 992. At time (? -l) t ?, a superframe 908 includes frames 916, 926, 936, ..., and 996. At time Nti, a superframe 909 includes frames 918, 928, 938, ..., and 998. Each super GOP contains N superframes.
For example, the super GOP 900 includes the superframes 902, 904, 906, 908, and 909. Note that because L programs can have any GOP structure of the program, the frames in a superframe can have any type of image, either I, P or B, for example. In addition, the program frame complexity can be used to determine an objective number of bits for a superframe and an individual frame. A complexity measurement, C, can be defined for a frame as the product of the quantization level, Q, used for the frame and the number of bits, R, generated for the frame when using the Q, that is, (3) C = RQ However, it should be understood that any available complexity measurement can be used. Let Q?, N, t and R? A, t, respectively, be the quantization parameter used for frame n of program 1 and the corresponding number of bits generated for the frame using Q? rn, tr where n ranges from one to the number of superframes in a super GOP (for example, five in the simplified example of Figure 9), 1 ranges from one to L, and t corresponds to the image type, I, P or B. For example, Q2 / 3, B corresponds to program 2, third frame, for example, frame 924, assuming it is an image B. For a superframe n, there may be L different frame complexity measurements, one for each regular plot, that is, (4) C - l ,, n, l = 30yC l, n tf- Rl, .n ,? 1 = 1, '2,'. . . , 'L Also, let Tn be the target number of bits for superframe n. The total number of bits generated from the L regular frames within the superframe n must be close to Tn, that is, Similarly, the total number of bits generated for all N superframes in a super GOP must be close to the target number of bits, T, assigned for each super GOP, is from ci r, (6) T =? Tn In order for the statistical multiplexer system to achieve a more uniform image quality, ideally, the same quantization parameter must be applied to all frames. Note that quantization is the only lossy operation in MPEG encoding and plays a critical role in controlling both image quality and bit rate. However, in order to represent the different types of images (I, P and B), a constant weighting factor, K?, N, t; for each type of image, that is, K, for the image I ^ Kt, ", < Kr for the image PK * for the image B The "t" in subscript is an index for the type of image, for example, I, P or B. This subscript can be omitted in some of the equations in the present when it is not necessary Your presence. It is possible to use the following weighting factors: K? = KP = l and KB = 1.4. In addition, in many cases, it is desirable for some of the programs to determine a higher priority (eg, relative number of bits) than another. Therefore, the level of quality between the programs can be controlled. Again, because quantification is the primary loss operation, the quality level of the program can be controlled by controlling the quantification. Specifically, the quantization parameter for program 1 can be further modulated by a weighting factor, wl f as given) QIM - WIKIMQ For programs with a higher priority, a smaller weighting factor is used, and for lower priority programs, a larger weighting factor is used. A larger quantization level results in thicker step sizes (eg, a lower quality image), while a smaller quantization level results in finer step sizes (eg, an image of higher quality). Therefore, by controlling the weighting factor, W? R in Eq. (7a) the quantization parameter is controlled and, therefore, the quality. From equations (5) - (17), the target number of bits for the superframe n, Tn, is where the numerator is the sum of the complexity measurements for the L regular frames in the superframe n, and the denominator is the sum of the complexity measurements for all the frames in the current super GOP. In other words, the target number of bits for the superframe n, Tn, is proportional to its complexity. With the aforementioned bit distribution equation, the computation of the target number of bits for a superframe requires the complexity measurements for all the L * N frames within the current super GOP (L, N), that is, C?, N , tr 1 = 1, 2, ..., L and n = 1,2, ..., N. However, this may not be practical or desirable in some cases due to the required memory capacity and potential processing delays. According to the above, a simplified, alternative bit distribution scheme is provided. First, in each superframe n ', only 1 distribution of the remaining bits, Tr, defined as through the remaining superframes from n 'to N in the super GOP (L N). This leads to the distribution of bits for the superframe n 'as The complexity measurements for the previous frame from 1 to n '- 1 are no longer necessary when computing the target number of bits for the superframe n', Tn '- At the beginning of the processing, a new super GOP, Tr, is restored. .explain below: (12) Tr = Tr + T, where T is the target number of bits for a new super GOP, and Tr, on the right side of the equation, which can be either a positive number or negative, is the number of bits left by the previous super GOP. Secondly, according to the present invention, it is assumed that all future frames of the same type of image in a program (in a super GOP) have the same complexity measurement, ie, (13) C?, N, t = C?, N; t n '= n = N, which is a reasonable presumption for continuous scenes. For example, the complexity measurement of an image P in the super video program of a super GOP can be used as the complexity measurement for the next image P in the very last video program in the same super GOP. Advantageously, there is no need to calculate a separate complexity measurement for each image. In addition, the present inventors have determined that satisfactory results are achieved when it is assumed that all future frames of the same type of image in a program have the same complexity measurement. Now, for each program, only three measurements of frame complexity are required, each one corresponding to one of the three types of images, I, P or B, that is, C? r I, C? , p, C? f B, respectively. The three complexity measurements for a program are updated after encoding each frame n '(see Eq. (4)), at least for each image in a program before the last image in a super GOP. Complexity measurements can be estimated or calculated based on the average quantization parameter used for the previous frame of the same type (for example, averaged over an image), and the number of bits generated for the frame (see eq. 3) ) . In other words, complexity measurements in the current frame are available for each program. The dynamic distribution strategy for superframe n is consequently Here, 1. Cj, n, t is the complexity measurement corresponding to the type of image t e. { I, P, B } of the frame n for program 1. 2. K? , n, t is a constant factor used to compensate the type of image t of frame n of program 1. It can be either Kt, KP or KB, depending on the type of image. 3. Nl t I, N, P and N1 / B are, respectively, the remaining number of I, P or B images for program 1 in the super GOP in superframe n. Four . wi is the quality weighting factor for program 1, and is determined by the program or network service provider. Note that the numerator on the right side of equation (14) is the sum of the complexity measurements for all the frames in superframe n. It can be considered a complexity measurement for superframe n. On the other hand, the denominator can be considered as a complexity measurement for the complete set of remaining frames in the super GOP. Therefore, equation (14) assigns to a superframe an objective number of bits proportional to the complexity measurement of the superframe.
Restriction to the Superframe Target Rate Referring now to the buffers of the encoder 670 of Figure 6, the fullness of the encoder buffer in the frame n, ßL, can be calculated as is) B B + KJ- Rt anal (bpj) channel _r ate where Rcam¡Kbp) ~ '~ s. - is the number of average frame_ rate bits per frame (bpf) transmitted through the channel. Let -L fíJX max the maximum size of the encoder buffer. Then, to ensure that the encoder buffer never has an overflow or subflow, the buffer fullness, ß, must be restricted within the range [0, # 1 'that is, (v16) O = ß-J' n = -βLJC max From equations (15) and (16), we have L (17) U - £) "_! + 2-ll-l, n, t ~ Jt-ccmallbpf) ~ t) max 1 = 1 (18) jf? M / (í /! /] JD -I - Zi? -lnl ~ JLYcanaHbpJ) t) max t n- \ 1 = According to the present invention, this is a restriction on the total number of bits generated for the superframe n for a determined. If the number of bits required can be controlled to meet the target number of bits, that is, the restriction to the total number of bits for a superframe n becomes the restriction to the target number of bits for the superframe, that is, (0) jfc channel (bpj) -) "_ i -" - channel (bpf) ÍJ max ~~ D n - l Therefore, before beginning to encode each superframe n, the target rate determined by the equation is verified (14) to determine if it is in the appropriate range, and if not, the target rate is adjusted as cited below: tVLanal (bpj) JD n-V 1 n) J3 n-l (2 1) Tn = I anuH p /) LJ max D n-C 1 n? -? unal. { bpj) J-J mux D n- - TL n otherwise Target Rate for Regular Frames Once an objective number of bits has been established for a superframe, what remains is to distribute the bits in the regular frames in the superframe. According to the present invention, the target number of bits for the frame n of the program 1, T? , n, is where the numerator on the right is the complexity measurement for frame n of program 1, and the denominator is the complexity measurement for superframe n. The distribution of Tn allocated for the superframe n over the L regular frames of the superframe is again based on the complexities of the program frames.
Restriction to the Target Rate for the Regular Screen In applications of multiple program television stations, several video programs are multiplexed into a single fixed rate transmission channel. The service information included in the bit stream, such as packet identifiers (PIDs), provides the necessary navigation information to allow a central node server to select the desired programs for transmission, and allow a decoder placed in the In the home of a user, tune to the appropriate channel and extract (demill and iplex) the packets corresponding to the selected programs, as shown in Figure 10. Figure 10 illustrates a decoder according to the present invention. The multiplexed data from the transmission channel, such as a cable television network and / or a satellite distribution network, is received by a DEMUX 1000. Based on a default or user selection, a particular program is selected in the multiplexer to decode and display. The data coming from successive frames of the selected program is provided by the DEMUX 1000 to the buffer of the decoder 1010, and then to an MPEG decoder 1020, for example, to retrieve the program data in the domain of the pixels, for example , for a program 1. Decoder 1020 may correspond to decoder 400 of Figure 4. The tiplexed demultip stream is at a variable rate. To ensure that the buffer of the decoder 1010 does not show an overflow or subflow when any of the programs is selected, additional restrictions must be imposed on the encoder. Specifically, assume that program 1 is selected and that the decoding delay is N 'frames. The decoding delay is the time a frame lasts in the decoder's buffer before being decoded. Let Rc?, N be the number of bits transmitted by program 1 during the nth frame period, where the subscript "c" refers to "channel". Then, the decoder's buffer will be filled to a level before extracting any bit for decoding. The subscript "d" refers to the decoder 'During the frame period n > N ', the decoder buffer introduces R?, N, t bits to the decoder, and receives Rc?, N + N' bits from the channel. The fullness of the buffer in the frame n > N 'is determined therefore as (24) Bn? JRl, ~ Bln-l ~ Rl, n, t 'where can be considered as the fullness of a virtual encoder buffer for program 1 in frame n. The subscript "e" refers to "encoder". Note that Rc?, N 'is the number of bits transmitted during the interval of frame n' of program 1, not the number of bits used to encode the frame. Be Jí-? mu.x the maximum decoder buffer size. To ensure that the decoder buffer never has an overflow or subflow, the fullness of the buffer memory, B n, must lie within the range -1 0 [0, 'J phíma .x 1 J,' that is, A part of the equations (24) and (26) we have '-B?, N-? According to the present invention, this is a restriction on the number of bits generated for frame n of program 1. Again, we assume that the bit rate for each program can be controlled to meet its target rate, that is, The restriction to the number of bits for each individual frame (1, n) becomes the restriction to the target number of bits for the frame, that is, Therefore, before starting to codify each frame n of program 1, we need to verify if its target rate is within the appropriate range, and if not, adjust the target rate as explained below (3 1 ) LJ m ?.
Notice that Be?, N_? is the fullness of a virtual encoder buffer for program 1 in frame n - 1 and is therefore available in frame n. However, R ° ?, n < , n '= n, n + l, ..., n + N', are the bit numbers that will be transmitted for program 1 during the intervals of the current and future frames n, n + 1,. . . , n + N '. With a constant bit / frame rate, Rc ?? n ', n' = n, n + 1, .. ,, nfN 'can be measured in the encoder's buffer, as shown in Figure 11. Figure 11 illustrates the data stored in an encoder buffer according to the present invention. An encoder buffer may be provided after the VLC 330 in Figures 3-5, for example. Within each interval window n = n, n + l,. . . , n + N r of JR - C channel (bpf) bits in the encoder buffer, the number of bits for program 1 is Rc? , n - Notice that Rc? , n can be zero. For example, the windows of interval n + N '(110), n + l (1120), and n (1130) are shown. The interval window n (1130) includes the intervals Rc? + 1, n (1132), Rc 1 / n (1134), and Rc? -l f n (1136). When a frame n is encoded, all the compressed bits are entered into the buffer of the encoder. This can be done on a program-by-program basis by compressing the entire frame of a program in frame n, entering all the bits in the encoder's buffer, and processing the next program in the same frame rate n, and so on. 0, processing can occur when processing a portion of a plot, such as a macroblock or slice, at a time. What is important when modeling the encoder buffer is the number of bits generated for a program in a frame interval.
Restriction to the Maximum and Minimum Bit Rate The average bit rate through a certain number of frames, N '', can be controlled by limiting the target number of bits for each frame within a specific range. This may be desirable, for example, to avoid large fluctuations in the bandwidth consumed by each program. Let Rma X and ^ min be the number of average maximum and minimum bits, respectively, through each N '' frame. Therefore, the number of average bits per N '' frames up to a frame n has to be in the range of [• Rmin -Rmax], that is, This is another restriction on the number of bits per frame n of program 1. Note that R-? , n ', t, n' = n -N '', -N 'r,. . . , n - l, are all available for the encoder in frame n. Again, assume that the current rate can be made close to the target rate by appropriate control of the target number of bits, ie, the additional restriction to the current number of bits for the frame n then becomes a restriction to its target number of bits , that is, nl (5) N "Rmin- SRl, n ,, = Tl, n = NRm * < - SR.n ,, n '= nN" rl = nN "In summary, according to the present invention , the target number of bits for each individual frame is adjusted as follows: (36) nl N "Rmin- S Rl. l YES Tl, n < N "Rmin- SRl.n, n '= n-N" rí = n-N "T l, n N" R, nu.X S Rl, l * • Tl, n > N "Rm - SRl.n, n '= - n ~ N" n' = n-N "Tl, n otherwise Control Rate Now, the target number of bits must be met for each frame of a program. This can be achieved by adjusting the quantization parameter, Q, in the MPEG encoder (Figure 3) and Q2 in the transcoder (Figures 4 and 5). The rate control scheme described in ISO / IEC (MPEG-2) "Test Model 5" (TM5), April 1993, for example, can be used. An alternative rate control scheme that is more accurate than TM5 can be found in L. Wang, "Rate Control for MPEG Video Coding," Proc. SPIE on Vi s ua l Comm uni ca t i on s and Ima ge Pro cess i n g, p. 53-64, Taiwan, May 1995, which requires multiple VLC and Q operations for each frame. In accordance with the foregoing, it can be seen that the present invention provides a method and apparatus for distributing bits in a system of statistical multiplexing. The invention extends the concept of the statistical multiplexer to accommodate both compressed and uncompressed video programs using transcoding and encoding, respectively. In addition, for superframe and super GOP bit distribution schemes, the invention provides target bit rates, and restrictions on target bit rates, for super GOPS, superframes, and regular frames, as well as rate restrictions of minimum and maximum bits. Additional efficiencies are achieved by assuming that all images of the same type in a super GOP program have the same complexity, thus avoiding the need to calculate and maintain the complexity for each image when determining the target number (first, second and third) of bits. An objective number of bits, Tn, for a superframe, which is a collection of frames across all the channels in a given frame instance, is adaptive and is capable of addressing any combination of types of images. Frames of the same type of image for a program are assigned the same number (or similar) of bits. To achieve this adaptation in the bit distribution, the invention provides a dynamic bit distribution strategy which determines an objective number of bits for each program on a frame-by-frame basis according to the above coding information, such as parameters of quantification used, and the resulting number of bits. The hierarchical dynamic bit distribution is used starting from a super GOP level, then to a superframe level, and then to the individual frame level (image). The concept can also be extended to a sub-frame level, where the bits are distributed for a portion of a frame such as a slice or Video Object Plane (VOP). At each level, an objective number of bits is determined. In addition, to prevent the encoder and decoder buffers from having an overflow or underflow, restrictions are imposed on the compressed bit rate of each program in the multi-program transmission environment. In addition, the quality of the program can be controlled according to a program priority weighting factor. Although the invention has been described in connection with various specific embodiments, those skilled in the art will appreciate that numerous adaptations and modifications may be made thereto without being insulated from the spirit and scope of the invention as set forth in the claims. For example, although the invention was described in terms of the MPEG-2 standard, it can be adapted for use with other standards which use groups of images or analog constructions with different types of images

Claims (36)

  1. CLAIMS Having described the invention as an antecedent, the content of the following claims is claimed as property: 1. A method of distributing bits for digital video, characterized in that it comprises the steps for: providing a plurality L of video programs to an encoder, each program having successive groups of images (GOPs); each GOP having an associated number of images; providing a super GOP comprising at least one GOP from each of said L video programs, and having a length of N images; calculate a first bitjs target number, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; define a respective complexity measurement for each type of image in each bad program;
  2. calculate a second objective number of bits to encode each n th superframe of images, where n = l, ..., N, according to the first objective number of bits, T, and the complexity measurement of each? th image in the nth associated superframe, where 1 = 1, ..., L; and calculating a third target number of bits to encode each? this image in the associated superframe n in accordance with the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the terrible super-drama; where: the same complexity measurement is used for each image with a common type of image in at least one of the video programs of the super GOP to calculate the target numbers of second and third bits. The method according to claim 1, characterized in that: the same complexity measurement is used for each image with a common image type in each of the video programs of the super GOP to calculate the target numbers of second and third bits.
  3. 3. The method according to the claim
    1, characterized in that: the video programs of the super GOP have a plurality of different types of images.
  4. 4. The method according to the claim
  5. 1, characterized in that: the complexity measurement of a first image of a certain type in at least one of the video programs is used for each subsequent image of the same type in at least one of the video programs of the super GOP to calculate the target numbers of second and third bits. The method according to claim 1, characterized in that: the video programs are adapted for communication through a broadband communications network to a decoder population.
  6. 6. The method according to claim 1, characterized in that: the respective complexity measurement for each of the types of images in each program is updated after each image in the program before it is programmed. encode the last super GOP image
  7. 7. A method of distributing bits for digital video, characterized by comprising the steps for: providing a plurality L of video programs in an encoder, each program having successive groups of images (GOPs); each GOP having an associated number of images; provide a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the video programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; define a respective complexity measurement for each type of image in each of the very last program; calculate a second target number of bits to code each n: s -ma superframe of images, where n = l, ..., N, according to the first objective number of bits, T, and the complexity measurement of every j? es image in the nes ma associated superframe, where 1 = 1, ..., L; calculate a third target number of bits to encode each such image in the nested associated super-frame according to the second number "bit target and the associated complexity measurement, and the sum of the complexity measurements for each image in the bad associate SUpertrama, and provide respective weighting factors, w, for the different video programs according to a relative priority thereof, wherein: the third bit number to encode every third image in the associated n superframe is calculates according to the respective weighting factor of the associated video program
  8. 8. A bit distribution method for digital video, characterized in that it comprises the steps for: providing a plurality L of video programs in an encoder , each program having successive groups of images (GOPs), comprising at least one of the video programs, uncompressed video data; GOP gives an associated number of images; processing precompressed image data to obtain partially uncompressed data from at least one particular video program of the plurality L of video programs; provide a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the video programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; define a respective complexity measurement for each type of image in each of the first program;
    calculate a second target number of bits to encode each nth superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the complexity measurement of each j? es? ma image in the Oth associated superframe, where 1 = 1, ..., L; and calculate a third target number of bits to code each j. s? to image in the Oés? associated super-frame according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the nth associated super-frame; wherein the precompressed image data is transcoded in such a way that a bit rate of the precompressed image data is different than a bit rate provided by the associated third associated bit number.
  9. 9. A bit distribution method for digital video, characterized in that it comprises the steps for: providing a plurality L of video programs in an encoder, each program having successive groups of images (GOPs); wherein a buffer associated with the encoder. receives encoded data from the video programs, each GOP having an associated number of images'; provide a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; calculating a first target number of bits, T, to encode the super GOP according to the number of images of the super GOP, L * N, and the available capacity of a channel through which the video programs are transmitted;
    wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; define a respective complexity measurement for each type of image in each tenth program; calculate a second target number of bits to encode each Oleslma superframe of images, where N = l, ..., N, according to the first target number of bits, T, and the complexity measurement of each j? es? ma image in the nes? ma associated superframe, where 1 = 1, ..., L; calculate a third target number of bits to encode each ies? ma image in the n ß s? the associated superframe according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the associated lousy SUpertrama; and before encoding the n s lma superframe with the second associated bit number, at least one of the steps for; (a) set the second target bit number associated, if necessary, to avoid xx cais 3a p -or should -jto of J / -J? t? -, 1, 'where
    R channel bpf) e s a number of average bits per image transmitted through the channel; and ßl is a fullness level of the buffer after the n-superframe has been coded; and (b) setting the second associated bit number, if necessary, to avoid exceeding;
    where J fí-J ma .x is the maximum capacity of the buffer memory.
  10. 10. A method of distributing bits for digital video, characterized in that it comprises the steps to: provide a plurality L of video programs in an encoder, each program having successive groups of images (GOPs); wherein a buffer associated with the encoder receives encoded data from the video programs, and the video programs are transmitted through a channel to a decoder and stored in a buffer therein;
    each GOP having an associated number of images; provide a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of the channel; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; define a respective complexity measurement for each type of image in each I? mo program;
    calculate a second target number of bits to encode each nésl to a superframe of images, where n = l, ..., N, according to the first objective bit number, T, and the complexity measurement of each j? s? ma image in the Oes? ma associated superframe, where 1 = 1, ..., L; calculate a third target number of bits to encode each? this image in the nesx ma associated superframe according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the bad associated SUpertrama; and before encoding the 1 st image in the lousy superframe with the associated third target bit number, at least one of the steps for: (a) setting the associated third target bit number, if necessary, to prevent it from falling under
    where: 77 + .V 2y, R? , f is the sum of the number of bits transmitted
    for the n th to (n + N is ima images for the th video program; N 'is a decoder decoding delay; Be?, n-? is a fullness level of the encoder's buffer after it is has encoded the 2nd image in the n - the superframe, and Jfí- 1max is the cap-maximum capacity of the decoder's buffer, and (b) set the third associated bit number, if it is necessary, to avoid exceeding
  11. 11. A bit distribution method for digital video, characterized in that it comprises the steps for: providing a plurality L of video programs in an encoder, each program having successive groups of images (GOPs); each GOP having an associated number of images; provide a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; calculating a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and an available capacity of a channel through which the video programs are transmitted;
    wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; define a respective complexity measurement for each type of image in each last program; calculating a second target number of bits to encode each nth superframe of images, where n = l, .., N, according to the first target number of bits, T, and the complexity measurement of each jth image in the nési to the associated superframe, where 1 = 1, ..., L; calculating a third target number of bits to encode each program in the very poor associated superframe according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the lousy SUper-branch associated; wherein Rmin is a minimum average number of bits to encode N '' > 1 images; and before encoding the j? es image in the lousy superframe with the third associated bit number, at least one of the steps for: (a) setting the third associated bit number, if necessary, to prevent it from falling below
    where 77-1 Ri.rí is the sum of the number of bits n '= nh' transmitted for the (nN /, ith up to (nl s s ima images for the 2és? m0 video program, and (b) adjust the third objective bit number associated, if necessary, to avoid exceeding
    where Rma? is a maximum average number of bits for encoding N '' > 1 images 12. A method for encoding uncompressed video source data, and transcribing precompressed video source data, characterized in that it comprises the steps to: partially decompress the precompressed video source data to obtain partially uncompressed video data; distribute the bits to encode the uncompressed video source data according to a scheme of many statistical typing; and distributing the bits to transcode the partially uncompressed video data according to the scheme of statistical multiplexing; wherein the precompressed image data is transcoded such that a bit rate of the precompressed image data is different than a bit rate provided by the associated distributed bits. 13. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs is provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, characterized in that comprises: means for providing a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; means for calculating a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the video programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; means to define a respective complexity measurement for each type of image in each _7_ s? m ° program; means to calculate a second target number of bits to encode each nesiM superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the complexity measurement of each jes? ma image in the n? a associated superframe, where 1 = 1, .., L; and means for calculating a third target number of bits to encode each Ies? "to image in the associated superframe according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the very bad associated superframe, where: the same complexity measurement is used for each image with one type of image in at least one of the video programs of the super GOP to calculate the target numbers of second and third bits. The apparatus according to claim 13, characterized in that: the same complexity measurement is used with a common image type in each of the video programs of the -super GOP to calculate the target numbers of second and third bits. The apparatus according to the claim
    13, characterized in that: the video programs of the super GOP have a plurality of different types of images. 16. The apparatus according to the claim
    13, characterized in that: the complexity measurement of a first image of a certain type is used in at least one of the video programs for each of the following images of the same type in at least one of the video programs of the super GOP to calculate the target numbers of second and third bits. 17. The apparatus according to claim 13, characterized in that: the video programs are adapted for communication through a broadband communications network to a decoder population. 18. The apparatus according to the claim
    13, characterized in that: the respective complexity measurement for each of the types of images in each 10th program is updated after each image in the same program before the last image of the super GOP is coded. 19. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs is provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, characterized in that comprises: means for providing a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; means for calculating a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the video programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; means for defining a respective complexity measurement for each type of image in each program; means for calculating a second target number of bits to encode each nes? M superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the complexity measurement of each _jes? ma image in ns l-m3- associated superframe, where 1 = 1, .., L; and means for calculating a third target number of bits to encode each _ £ is? ma image in the nth associated superframe according to the second objective number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the n? m associated superframe; and means for providing respective weighting factors w, for the different video programs according to a relative priority thereof; wherein: the third target number of bits to encode each lousy image in the n siM associated superframe is calculated according to the respective weighting factor of the lousy associated video program. 20. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs is provided in an encoder, each program has successive groups of images (GOPs) _, each GOP has an associated number of images, and minus one of the video programs comprises uncompressed video data, characterized in that it comprises: means for processing precompressed image data to obtain partially uncompressed data in at least one particular video program of the plurality L of video programs; means for providing a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; means for calculating a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the video programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; means to define a respective complexity measurement for each type of image in each? é? mo program; means for calculating a second target number of bits to encode each néslM superframe of images, where n = l, ..., N, according to the first objective number of bits, T, and the complexity measurement of each one? ma image in the Oés? ma associated superframe, where 1 = 1, .., L; and means for calculating a third target number of bits to encode each le3in? a image in the n? s associated superframe in accordance with the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the associated superframe n ^ 1"13, wherein: the precompressed image data is transcoded such that a bit rate of the precompressed image data is different from a bit rate provided by the third target number associated bits 21. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs is provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, and wherein a buffer associated with the encoder receives the encoded data from the video programs, characterized in that it comprises: means for proportion a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; means for calculating a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the ideo programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; means for defining a respective complexity measurement for each type of image in each syllabus; means for calculating a second target number of bits to encode each i] és? p, superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the measurement of complexity of each jés? ma image in the nsta associated superframe, where 1 = 1,. . , L; and means for calculating a third target number of bits to encode each les? p? a image in the nslM associated superframe according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the n? a associated superframe; and at least one of: (a) means for adjusting the second associated target number of bits, if necessary, to prevent it from falling below
    J R- Vcanal "(b, .pf,) .- J / -J? n -, \ 'before the n th is encoded
    superframe with the second associated bit number, where:
    J RL - ca "l! Í (bAp" j /) is a number of bits p ^ half per image transmitted through the channel; and Bn-? it is a level of fullness of the buffer after the n-l and x superframe has been coded; and (b) means for adjusting the associated second target bit number, if necessary, to avoid exceeding
    J R V anaH /, h_p,), + J / -J? ma • x ~ J fí-J n- il 'before the bad superframe was encoded with the second associated bit number, where J fí-J "mux is the maximum cap capacity of the buffer 22. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs is provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, and in where a buffer associated with the encoder receives the encoded data from the video programs, and the video programs are transmitted through a channel to a decoder and stored in a buffer therein, characterized in that it comprises: means to provide a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; means for calculating a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of the channel; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; means for defining a respective complexity measurement for each type of image in each i? a program; means for calculating a second target number of bits to encode each n? to a superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the complexity measurement of each image in the footprint associated super-frame, where means for calculating a third target number of bits to encode each I? image in the n "to associated super-frame according to the second target number of bits and the complexity measurement associated, and the sum of the complexity measurements for each image in the associated super-frame, and at least one of: (a) means to adjust the associated third target bit number, if necessary, to prevent it from falling through under
    before the image is coded in the nth superframe with the third associated bit number, where:
    is a sum of the number of bits transmitted for the n th th (n + N ') s ima images for the r th video program; N 'is a decoder delay of the decoder; Be?, N-? is a fullness level of the encoder's buffer after the I? ma? image has been encoded in the n? l? s? ma superframe; Y
    ? fíJ d max is the maximum capacity of the decoder's buffer; and (b) means for adjusting the associated third target bit number, s i is n + N 'necessary, to avoid exceeding? R ~ B n-? '
    before the 1st image is encoded in the lousy superframe with the third associated bit number; 23. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs is provided in an encoder, each program has successive groups of images (GOPs), and each GOP 9 -has an associated number of images, characterized in that it comprises: means for providing a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; means for calculating a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through which the video programs are transmitted; wherein each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; means for defining a respective complexity measurement for each type of image in each? this program; means for calculating a second target number of bits to encode each nESiBa superframe of images, where n = l, ..., N, according to the first objective number of bits, T, and the complexity measurement of each jés? ma image in the Ilesian associated superframe, where 1 = 1, • •, L; means for calculating a third target number of bits to encode each les? p? a image in the n? ma associated superframe according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the associated associated super-frame; wherein Rmin is a minimum average number of bits for encoding N '' > 1 images; and at least one of: (a) means for adjusting the associated third target bit number, if necessary, to prevent it from falling below
    77-1 N "Rnlm ~? Rin 'an fc is where the J_thima n' = nN" image is encoded in the IiaslII to superframe with the third associated bit number, where: 77-1? R¡ " ? is the sum of the number of bits p '= n-' "transmitted for (n-N, ',') ess i? mmaa up (n-l) é s ima
    images for the j? es? mo video program; and (b) means for adjusting the associated third target bit number, if necessary, to prevent it from exceeding before encoding the 2nd image in the super-super-frame with the associated third target bit number; wherein Rmax is the maximum average number of bits to encode N '' > 1 images 24. An apparatus for encoding uncompressed video source data, and transcoding precompressed video source data, characterized in that it comprises: means for partially decompressing the precompressed video source data to obtain corresponding partially uncompressed video data; means to distribute bits to code. the video source data not compressed according to a scheme of statistical multiplexing; and means for distributing the bits to transcode the partially uncompressed video data according to the scheme of statistical multiplexing; wherein the precompressed image data is transcoded in such a manner that a bit rate of the precompressed image data is different than a bit rate provided by the associated distributed bits. 25. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs are provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, characterized in that comprises: a rate control processor; a super GOP processor, a superframe processor, a frame processor, and a complexity processor, each of which is associated with the rate control processor; wherein: the rate control processor is responsive to a signal defining a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; the super GOP processor is adapted to calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through the which video programs are transmitted; each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; the complexity processor is adapted to calculate a respective complexity measurement for each type of image in each other program; the superframe processor is adapted to calculate a second target number of bits to encode each nth superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the measurement of associated complexity of each j_éSima image in the
    , is an associated superframe, where 1 = 1 L; and the frame processor is adapted to calculate a third target number of bits to encode each j? s? a image in the n th associated superframe according to the second objective number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the bad associated SUpertrama; where: the same complexity measurement is used for each image with a common type of image in at least one of the video programs of the super GOP to calculate the target numbers of second and third bits. 26. The apparatus according to claim 25, characterized in that: the same complexity measurement is used for each image with a common image type in each of the video programs of the super GOP for calculating the target numbers of second and third bits. 27. The apparatus according to the claim
    25, characterized in that: the video programs of the super GOP have a plurality of different types of images. 28. The apparatus according to the claim
    25, characterized in that: the complexity measurement of a first image of a certain type in at least one of the video programs is used for each subsequent image of the same type in at least one of the video programs of the super GOP to calculate the target numbers of second and third bits. 29. The apparatus according to claim 25, characterized in that: the video programs are adapted for communication through a broadband communications network to a decoder population. 30. The apparatus according to the claim
    25, characterized in that: the respective complexity measurement for each of the types of images in each file is updated after each image in the program or before the last image of the super GOP is coded. 31. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs are provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, characterized in that comprises: a rate control processor; a super GOP processor, a superframe processor, a frame processor, and a complexity processor, each of which is associated with the rate control processor; wherein: the rate control processor is responsive to a signal defining a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; the super GOP processor is adapted to calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through the which video programs are transmitted; each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; the complexity processor is adapted to calculate a respective complexity measurement for each type of image in each program; the superframe processor is adapted to calculate a second target number of bits to encode each lousy superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the " measurement of associated complexity of each I is an image in the nth associated superframe, where 1 = 1 L, the frame processor is adapted to calculate a third objective number of bits to encode each 1 image in the nth associated superframe of according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the super-associated associated super-frame, the rate control processor provides respective weighting factors, w, for the different programs of video according to a relative priority thereof, and the rate control processor is adapted to calculate the third target number of bits to encode each
    1 st image in the n e a xma associated superframe according to the respective weighting factor of the associated video program. 32. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs are provided in an encoder, each program has successive groups of images (GOPs), each GOP has an associated number of images, and at least one of the video programs comprises uncompressed video data, characterized in that it comprises: a transcoding processor for processing precompressed image data to obtain partially uncompressed data of at least one particular video program of the plurality L of video programs; a rate control processor; a super GOP processor, a superframe processor, a frame processor, and a complexity processor, each of which is associated with the rate control processor; wherein: the rate control processor is responsive to a signal defining a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; the super GOP processor is adapted to calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through the which video programs are transmitted; each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; the complexity processor is adapted to calculate a respective complexity measurement for each type of image in each? this program; the superframe processor is adapted to calculate a second target number of bits to encode each nth superframe of images, where
    • n = l, ..., N, according to the first objective number of bits, T, and the associated complexity measurement of each 1th image in the nés ima associated superframe, where 1 = 1, ..., L; the frame processor is adapted to calculate a third objective number of bits to encode each image in the super-associated associated super-frame according to the second target number of bits and the associated complexity measurement, and the sum of the measurements of complexity for each image in the terrible associated SUpertrama; and the precompressed image data is transcoded in such a way that a bit rate of the precompressed image data is different than a bit rate provided by the associated third target bit number. 33. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs are provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, characterized in that comprises: a rate control processor; a super GOP processor, a superframe processor, a frame processor, and a complexity processor, each of which is associated with the rate control processor; wherein: the rate control processor is responsive to a signal defining a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; the super GOP processor is adapted to calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through the which video programs are transmitted; each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; the complexity processor is adapted to calculate a respective complexity measurement for each type of image in each other program; the superframe processor is adapted to calculate a second target number of bits to encode every nth superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the measurement of associated complexity of each j image in the bad associated superframe, where 1 = 1, ..., L; the frame processor is adapted to calculate a third target number of bits to encode every second image in the associated superframe in accordance with the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the nth associated super-frame an intermediate memory associated with the encoder receives encoded data from the video programs; the superframe processor receives a sensitive signal from a fullness level of the buffer; wherein at least one of: (a) the superframe processor is adapted to set the second associated bit number, if necessary, to prevent it from falling below
    JR i uiuiH, h ^ pJ,) - Jfí-J n "-, before g * ue that the superframe is encoded with the second associated bit number of bits; i \ anul (bpj) is a number of average bit per image transmitted through the channel, and ßn is the fullness level of the buffer after which the superframe is encoded, and (b) the superframe processor is adapted to adjust the second associated target number of bits, if necessary, + J fJ {max - U /? ' n-, l before encoding the nés? n? superframe with the second associated bit number, where J fí-J ma .x is the cap Maximum memory capacity 34. A distribution device of bits for digital video, characterized in that a plurality L of video programs are provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, characterized in that it comprises: rate control; a super GOP processor, a superframe processor, a frame processor, and a complexity processor, each of which is associated with the rate control processor; wherein: the rate control processor is responsive to a signal defining a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; the super GOP processor is adapted to calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through the which video programs are transmitted; each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; the complexity processor is adapted to calculate a respective complexity measurement for each type of image in each jés? Itl ° program; the superframe processor is adapted to calculate a second target number of bits to encode each lousy superframe of images, where n = l, ..., N, according to the first target number of bits, T, and the measurement of associated complexity of each _t_és? raa image in the terrible associated superframe, where 1 = 1, ..., L; the frame processor is adapted to calculate a third target number of bits to encode every 1th image in the super-associated associated super-frame according to the second target number of bits and the associated complexity measurement, and the sum of the complexity measurements for each image in the terrible associated SUpertrama; a buffer associated with the encoder receives the encoded data from the video programs, and the video programs are transmitted through the channel to a decoder and stored in a buffer therein; the superframe processor receives a signal indicative of a fullness level of the buffer; wherein at least one of: (a) the superframe processor is adapted to adjust the associated third target bit number, if necessary, to prevent it from falling below before the image is coded? bad superframe with the third associated bit number; n + N '? R / n. s s s s s s s s s s s s s s s s s s s s s s
    transmitted for the lousy up (n + N ') th images for the j? es? or video program; N 'is a decoder decoding delay; Be?, N-? is a level of fullness of the encoder's buffer after the image has been encoded in the n-th superframe; and Bc is the maximum capacity of the decoder's buffer; and (b) the frame procesto adjust the associated third target bit number, if necessary, to avoid exceeding
    ? R) "'~ B'n-' before the t_es is coded?
    image in the child's superframe with the third associated bit number. 35. A bit distribution apparatus for digital video, characterized in that a plurality L of video programs are provided in an encoder, each program has successive groups of images (GOPs), and each GOP has an associated number of images, characterized in that comprises: a rate control proces a super GOP proces a superframe proces a frame proces and a complexity proces each of which is associated with the rate control proces wherein: the rate control procesis responsive to a signal defining a super GOP comprising at least one GOP from each of the L video programs, and having a length of N images; the super GOP procesis adapted to calculate a first target number of bits, T, to encode the super GOP according to the number of images in the super GOP, L * N, and the available capacity of a channel through the which video programs are transmitted; each super GOP comprises a plurality N of superframes, each superframe having L images at a common temporal reference point; the complexity procesis adapted to calculate a respective complexity measurement for each type of image in each? s or program; the superframe procesis adapted to calculate a second target number of bits to encode each p if a superframe of images, where n = l, ..., N, according to the first objective number of bits, T, and the associated complexity measurement of each image in the terrible associated SUpertrama, where 1 = 1, ..., L; the frame procesis adapted to calculate a third target number of bits to encode each _ [this image in the super-associated associated super-frame according to the second target number of bits and the associated complexity measurement, and the sum of the measurements of complexity for each image in the n? ma associated superframe; Rmin is a minimum average number of bits for encoding N '' > 1 images; wherein at least one of: (a) the frame procesis adapted to adjust the associated third number of bits, if necessary, to prevent it from falling below
    before the? th image in the lousy superframe is coded with the third associated bit number; Y
    The sum of the number of bi ts transmitted for the (n-N '') is? ma up to (n-l) is? ma images for? a video program; and (b) the frame procesis adapted to adjust the associated third number of bits, if necessary, to avoid exceeding
    before the image is encoded in the lousy superframe with the associated third bit number; where m is a maximum average number of bits to encode N '' > 1 images 36. An apparatus for encoding uncompressed video source data, and for transcribing precompressed video source data, characterized in that it comprises: an encoder; a transcoder; a rate control procesassociated with the encoder and the transcoder; wherein: the transcoder partially decompresses the precompressed video source data to obtain the corresponding partially uncompressed video data; the rate control procesdistributes the bits to the encoder to encode the video data 'non-compressed' according to a scheme of many statistical typing; the rate control processor distributes the bits to the transcoder to transcode the partially uncompressed video data according to a scheme of many statistical typing; and the precompressed image data is transcoded in such a way that a bit rate of the precompressed image data is different than a bit rate provided by the associated distributed bits.
    RE'STJMEN
    The present invention relates to a method and apparatus for distributing bits in a statistical multiplexing system (stat mux). A statistical multiplexer (stat mux)
    (600) accommodates both compressed and uncompressed video programs using transcoding (640, 650) and encoding
    (620, 630), respectively. Hierarchical dynamic bit distribution is used, starting at a super GOP level (700, 702, 704, 800, 900), then at a super-frame level (902; 904; 908; 909), and then at the regular frame level (individual (910, 920, 930, 990, 912, 922, 932, 992, 916, 926, 936, 996, 918, 928, 938, 998) In each level, an objective number of bits for a superframe, the which is a collection of data through all the channels in a given frame instance (tx, 2t ?, 3tx, ...), is adaptive and is capable of addressing any combination of types of images. An image type for a program is usually assigned the same number (or similar) of bits.The quality of the arlative program can be controlled using a program priority weighting factor (w). target bits and the minimum and maximum bit rates.
MXPA/A/2001/001945A 1998-08-27 2001-02-23 Dynamic bit allocation for statistical multiplexing of compressed and uncompressed digital video signals MXPA01001945A (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US09141265 1998-08-27

Publications (1)

Publication Number Publication Date
MXPA01001945A true MXPA01001945A (en) 2002-03-05

Family

ID=

Similar Documents

Publication Publication Date Title
US6167084A (en) Dynamic bit allocation for statistical multiplexing of compressed and uncompressed digital video signals
KR100329892B1 (en) Control strategy for dynamically encoding multiple streams of video data in parallel for multiplexing onto a constant bit rate channel
Tudor et al. Real-time transcoding of MPEG-2 video bit streams
CA2407682C (en) Statistical multiplexer and remultiplexer that accommodates changes in structure of group of pictures
TW529308B (en) Statistical remultiplexing with bandwidth allocation among different transcoding channels
US7170938B1 (en) Rate control method for video transcoding
US5532746A (en) Bit allocation method for controlling transmission rate of video encoder
Pan et al. A study of MPEG-4 rate control scheme and its improvements
EP1032214B1 (en) Method and apparatus for transcoding moving picture data
EP2123040B1 (en) An improved video rate control for video coding standards
EP2088782B1 (en) A method and a device for transcoding video
EP1217841A2 (en) Bitstream separating and merging system, apparatus, method and computer program product
MXPA02010321A (en) Rate control and buffer protection for variable bit rate video programs over a constant rate channel.
US6215824B1 (en) Transcoding method for digital video networking
CA2688249A1 (en) A buffer-based rate control exploiting frame complexity, buffer level and position of intra frames in video coding
Wang et al. Bit allocation and constraints for joint coding of multiple video programs
Lei et al. Rate adaptation transcoding for precoded video streams
KR100498332B1 (en) Apparatus and method for adaptive rate in video transcoder
JPH1013826A (en) Picture encoder, its method, picture decoder, its method, picture transmitter, picture receiver and recording medium
EP0888009A1 (en) Device and method for encoding image data, and image data transmission method
CN112004084B (en) Code rate control optimization method and system by utilizing quantization parameter sequencing
MXPA01001945A (en) Dynamic bit allocation for statistical multiplexing of compressed and uncompressed digital video signals
KR20030000310A (en) Apparatus for transcoding video
KR100286108B1 (en) Method and apparatus for estimating the number of bits of a video signal for real-time processing, method of encoding using the method, and apparatus therefor
Seo et al. Adaptive rate control algorithm based on logarithmic R–Q model for MPEG-1 to MPEG-4 transcoding