CN101112101A - Rate control techniques for video encoding using parametric equations - Google Patents
Rate control techniques for video encoding using parametric equations Download PDFInfo
- Publication number
- CN101112101A CN101112101A CN 200580047232 CN200580047232A CN101112101A CN 101112101 A CN101112101 A CN 101112101A CN 200580047232 CN200580047232 CN 200580047232 CN 200580047232 A CN200580047232 A CN 200580047232A CN 101112101 A CN101112101 A CN 101112101A
- Authority
- CN
- China
- Prior art keywords
- video
- defines
- parameter
- prediction
- frame
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
This disclosure describes rate control techniques that can improve video encoding. The described rate control techniques exploit relationships between the number of bits encoded per frame and the umber of non-zero coefficients of the video blocks after quantization is referred to as rho (p). The value of p is generally proportional to the number of bits used in the video encoding. This disclosure utilizes a relationship between p and a quantization parameter (QP) in order to achieve rate controlled video encoding. More specifically, this disclosure exploits a parametric equation to map a value oe a predicted p to a QP.
Description
Technical field
The application's case is advocated the right based on the 60/631st, No. 835 U.S. patent application case of filing an application on November 29th, 2004.
The present invention relates to Digital Video Processing, and more specifically, relate to rate controlled coding video sequence.
Background technology
The digital video function can be contained in the various devices, comprises Digital Television, the direct broadcast system of numeral, radio communication device, PDA(Personal Digital Assistant), laptop computer, desktop PC, digital camera, digital recording device, mobile phone or satelline radio phone, reaches in the similar device.Digital video apparatus can provide the remarkable improvement that is better than the conventional simulation video system aspect the full dynamic video sequence forming, revise, transmit, store, write down and play.
People have worked out the some kinds of different video coding standards that are used for the encoded digital video sequence.For example, animation expert group (the Moving Picture Experts Group MPEG) has developed some kinds of standards, comprises MPEG-1, MPEG-2 and MPEG-4.Other standard comprise International Telecommunications Union (International TelecommunicationUnion, ITU) H.263 standard, by the QuickTime of the Apple Computer company exploitation that is positioned at Cupertino California
TMTechnology, by being positioned at Redmond, the Video forWindows of the Microsoft company of Washington exploitation
TM, by the Indeo of Intel Company exploitation
TM, from being positioned at Seattle, the RealVideo of the RealNetworks company of Washington
TM, and by the Cinepak of SuperMac company exploitation
TMIn addition, new standard is also emerging in large numbers and evolution constantly, comprises ITU H.264 standard and some kinds of proprietary standards.
Many video encoding standards can realize the video sequence transmission rate that improves by with compress mode data being encoded.Compression can reduce to need the data total amount of transmission so that transmission of video frame effectively.For example, most of video encoding standards are utilized figure and video compression technology, and described figure and video compression technology are designed to be beneficial to one and need not to compress transmission of video and image on the bandwidth that narrows down that can obtain.For example, mpeg standard and ITU H.263 reach ITU H.264 standard support those to utilize similitude between each succession frame of video (be called time or interframe relevant) that the video coding technique of interframe compression is provided.The compression of this kind interframe normally realizes by estimation and motion Compensated Coding of Image.In addition, some video coding technique can utilize similitude in the frame (being called in space or the frame relevant) to come further compressed video frame.
People have developed the some kinds of rate control techniques that are used for video coding.Rate control techniques is even more important for the real-time Transmission of video sequence for the benefit of, sets but also can be used for the non real-time coding.For rate controlled, described coding techniques is dynamically adjusted the coded bit quantity of every frame.Particularly, rate controlled can limit the coded bit quantity of every frame, can set speed effectively video sequence be implemented to encode and therefore transmit on the bandwidth in distribution so that guarantee.If described coding techniques can not be in response to the scene change of video sequence, then the bit rate of video sequence real-time Transmission can the marked change along with scene change.Therefore, for defining a substantially invariable bit rate, during encoding, dynamically adjust the bit quantity of every frame.
A kind ofly realize that rate controlled Methods for Coding is to allow to be adjusted at quantization parameter used in the video coding process (QP).QP is defined in the quantification that occurs during the video coding, and directly influences the coded bit quantity of per second.When QP increased, the quality of preserving less data and video coding may reduce.And when QP reduced, the quality of preserving more data and video coding improved.Yet if QP is too small, the coded bit quantity of per second may exceed the bandwidth of being distributed, and infringement transmits the ability of frame in the bandwidth of this limited size.By selecting QP, can make the bit rate substantially constant of frame of video transmission with a dynamical fashion.For some Video Applications, expect that very bit rate has little variation.
Summary of the invention
The present invention's explanation can improve the rate control techniques of video coding.Particularly, described rate control techniques is to utilize the coded bit quantity of every frame and quantize relation between the nonzero coefficient quantity of rear video piece.The nonzero coefficient quantity that quantizes the video blocks of back frame is called rho (ρ).The value of ρ is proportional to the quantity of used position in video coding substantially.For realizing rate controlled video coding, the present invention utilizes the relation between a ρ and the quantization parameter (QP).More specifically, the present invention utilizes the parametric equation formula that the value of the ρ of prediction is mapped to a QP.This parametric equation formula can be following form:
ρ＝nc×e
-f(qp)，
Wherein e represents Euler (Euler) number (constant 2.71828182845904523536028747135 ...), the nonzero coefficient quantity of frame video blocks after quantification of ρ representative prediction, nc represents the number of coefficients (for example quantity of coefficient in a frame or the video blocks) with the unit representation of being considered, and f (qp) is the parametric function of this quantization parameter.For example, f (qp) can comprise a simple multinomial or an exponential function.In some cases, can use different constants to replace Euler's numbers (e), for example be worth 2 or may other constant (k).
In the rate controlled video encoding process, can define the programmable variable of f (qp) with this parametric equation formula initialization according to being used to.Then, can calculate the actual value (actual ρ) of ρ at selected QP value.Then, can adjust the programmable variable that is used to define f (qp), so that more preferably approach the ρ of prediction and the relation between the QP.Described technology can be applicable to video blocks stage speed control (for example macro-block level) or the control of frame stage speed.In a preferable embodiment, use the control of video blocks stage speed, but adjust variable in this parametric equation formula in the frame level.
In one embodiment, the invention provides a kind of video coding apparatus, it comprises the memory and an encoder of a stores video sequences, this encoder operation parameter equation selects to be used for this video sequence is carried out the quantization parameter (QP) of rate controlled video encoding, and the pre-quantitation (ρ of prediction) that this parametric equation formula will quantize the back nonzero coefficient maps to quantization parameter (QP).
In another embodiment, the invention provides a kind of rate controlled video encoding method, it comprises: the parametric equation formula of using a pre-quantitation (ρ of prediction) that will quantize the back nonzero coefficient to map to quantization parameter (QP) is selected a quantization parameter (QP), and uses this selected QP that one or more video blocks are implemented coding.
These and other technology as herein described can be implemented in the digital video apparatus by example, in hardware, form of software, form of firmware or its arbitrary combining form.If implement with form of software, then this software can be carried out in for example digital signal processor programmable processors such as (DSP).In this kind situation, the software that is used for carrying out described technology can at first be stored in computer-readable media, load in this programmable processor and in this programmable processor and carry out, to implement effective rate controlled coding in digital video apparatus.
Other details of each embodiment will be at accompanying drawings and are hereinafter mentioned in the explanation.According to this explanation and graphic and, will easily know further feature, purpose and advantage according to claims.
Description of drawings
Fig. 1 is a calcspar, its show one wherein the source digital video apparatus to the example system of the encoded video data sequences of receiving digital video device transmission;
Fig. 2 is the exemplary calcspar of digital video apparatus according to an embodiment of the invention;
Fig. 3 is a flow chart, the technology that its parametric equation formula that shows that a kind of use one predicted value with ρ maps to quantization parameter (QP) is carried out the control of frame stage speed;
Fig. 4 is a flow chart, the technology that its parametric equation formula that shows that a kind of use one predicted value with ρ maps to quantization parameter (QP) is carried out the control of video blocks stage speed;
Fig. 5 is a flow chart, the technology that its parametric equation formula that shows that a kind of use one is adjusted in the frame level is carried out the control of video blocks stage speed.
Embodiment
The present invention's explanation can be used for improveing the rate control techniques of video coding.Described technology can with various video encoding standards (MPEG-4 standard for example, ITU H.263, ITU is standard or other standard H.264) in any standard use together.Described technology is utilized the coded bit quantity of every frame (code rate) and is quantized relation between the nonzero coefficient quantity of video blocks of back frame.The nonzero coefficient quantity that quantizes the video blocks of back frame is called rho (ρ).The value of ρ is proportional to quantization parameter used in video coding (QP) substantially.Correspondingly, the present invention illustrates that relation between a kind of ρ of utilization and the QP realizes the technology of rate controlled video encoding.
More specifically, the technology described in the present invention utilizes a predicted value with ρ to map to the parametric equation formula of a quantization parameter (QP).Because ρ is proportional to code rate substantially, thereby can defines a required ρ and select a desired QP that should produce required code rate with it.Can be with this parametric equation formula initialization, but in video coding process, adjusted then, so that the ρ of the prediction of being defined by this parametric equation formula more preferably approaches the actual ρ that obtains from a selected QP.By this kind mode, in video coding process, dynamically adjust this parametric equation formula, so that selected QP maps to actual ρ more accurately and therefore obtains required code rate more accurately.
The parametric equation formula that is used to define the ρ of prediction can be following form:
ρ＝nc×e
-f(qp)，
Wherein e represents the arithmetic constant of a so-called Euler (Euler) number (2.71828182845904523536028747135...), the nonzero coefficient quantity of the video blocks of frame after the quantification of ρ representative prediction, nc represents the number of coefficients (for example quantity of coefficient in a frame or the video blocks) with the unit representation of being considered, and f (qp) is the parametric function of this quantization parameter.For example, f (qp) can comprise a simple multinomial or an exponential function.In some cases, can use different constants to replace the value of e, for example use value 2 or more generally use another constant (k).In a word, the example of f applicatory (qp) comprises as follows:
f(qp)＝α×qp，
f(qp)＝α×qp+β，
f(qp)＝α×qp
2+β×qp+δ，
Perhaps exponential function:
f(qp)＝α×qp
β。
In these examples, qp represents quantization parameter (being also referred to as QP in this article).
Variable α, β and δ comprise programmable parameter, and these programmable parameters can be adjusted in the rate control techniques process, more preferably to approach the relation between the ρ and qp in this parametric equation formula.Particularly, during cataloged procedure, adjust described programmable parameter, so that for different QP, the ρ of prediction is in fact corresponding to actual ρ.Set forth some examples of the parametric equation formula that can be used by technology of the present invention herein, many other parametric equation formulas are also applicable certainly.
Fig. 1 is a calcspar, and the example system 10 of communication link 15 to the encoded video data sequences of receiving system 14 transmission passed through in its demonstration one wherein source apparatus 12.The two is digital video apparatus source apparatus 12 and receiving system 14.Particularly, source apparatus 12 according to a kind of video standard that utilizes rate controlled video encoding (MPEG-4 standard for example, ITU H.263, ITU is any standard in standard or various other standard H.264) to video data encoding.As hereinafter being described in more detail, one in the device 12,14 of system 10 or the two are all carried out rate control techniques, so that the improvement video coding process.These rate control techniques are particularly useful for real-time Transmission video sequence on finite bandwidth communication links 15 such as for example wireless communication link.
Again for example, video capture device 23 can comprise video converter, is used for for example converting analog video data to digital of digital video data from television set, video tape recorder, video camera or another video-unit.In certain embodiments, source apparatus 12 can be configured by communication link 15 transmission real-time video sequences.In this kind situation, receiving system 14 can receive described real-time video sequence and show described video sequence to the user.Another is chosen as, and source apparatus 12 can be caught and encode and is sent to the video sequence of receiving system 14 with video data file (promptly not being real-time form) form.Thereby source apparatus 12 and receiving system 14 can for example supported application such as for example video communication, video clipping broadcast, video-mail or video conference in the mobile wireless network.Device 12 and 14 can comprise not concrete various other elements that show in Fig. 1.
Receiving system 14 can be any and can receive and the digital video apparatus form of decode video data.For example, receiving system 14 can comprise receiver 22, to receive encoded digital video sequences via for example intermediate line link, router, other network equipment and analog spontaneous emission device 20.Receiving system 14 also can comprise Video Decoder 24, an and display unit 26 that is used for showing to the user described sequence that is used to decode described sequence.Yet in certain embodiments, receiving system 14 can not comprise integrated display unit 14.In these cases, receiving system 14 can be used as receiver and decodes institute's receiving video data to drive a discrete readout device, for example television set or monitor.
The example device of source apparatus 12 and receiving system 14 comprises the server, work station or other desktop calculation element that are positioned on the computer network, and mobile computing device such as laptop computer or PDA(Personal Digital Assistant) for example.Other example comprise the digital television broadcasting satellite and for example receiving systems such as Digital Television, digital camera, digital camera or other digital recorder, for example have the digital video phones such as mobile phone of video capability, direct and two-way communicator, other wireless video device, an and similar device with video capability.
In some cases, source apparatus 12 and receiving system 14 comprise that separately one is used for digital of digital video data is implemented the encoder/decoder (CODEC) of coding and decoding.Particularly, source apparatus 12 and receiving system 14 the two all can comprise reflector and receiver and memory and display.Hereinafter the many kinds in the coding techniques of being summarized are to be illustrated in comprising the digital video apparatus context of encoder.Yet, should be appreciated that this encoder can constitute the part of CODEC.In this kind situation, this CODEC can be implemented in hardware, software, firmware, DSP, microprocessor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), discrete hardware components or its various combinations.
Block of pixels in 18 pairs one sequence of frames of video of video encoder in the source apparatus 12 is implemented to handle, so that encode this video data.For example, video encoder 18 can be carried out estimation and motion compensation technique-wherein that tendency to develop is defeated frame of video is divided into several pixel blocks (being called video blocks).For ease of illustration, described video blocks can comprise the piece of any size and can change to some extent in a set video sequence.As an example, ITU H.264 standard supports 16 * 16 video blocks, 16 * 8 video blocks, 8 * 16 video blocks, 8 * 8 video blocks, 8 * 4 video blocks, 4 * 8 video blocks and 4 * 4 video blocks.In video coding, use the little video blocks that heals in coding, to produce the good resolution that heals, and can be specifically designed to the frame of video position that comprises the high the level of detail of healing.In addition, video encoder 18 can be designed to 4 * 4 video blocks are implemented to handle, and as required from the bigger video blocks of described 4 * 4 video blocks reconstruct.For standards such as for example MPEG-4, video blocks can comprise 16 * 16 video blocks-be called sometimes " macro block ".
Each pixel in the video blocks can be represented by the various n-place values (for example 8 positions) that define the visible characteristic (for example color and the brightness of representing with colourity and luminance value) of pixel.Each pixel all can have one or more the two 8-place values of colourity and luminosity that are used for.Yet technology of the present invention is not limited to pixel format, and can be simpler still less position pixel format or more complicated more multidigit pixel format expand.Can be to define pixel also according to other color coordinates.In addition, the DCT of the described video blocks that can be represented by pixel value by discrete cosine transform (DCT) coefficient heel of video blocks represents.Another is chosen as, and video blocks can be represented by the wavelet transformation of non-zero small echo value heel pixel value.Briefly, rate control techniques as herein described can be applicable to various coding standards, pixel format and video block sizes and shape.Selected quantization parameter (QP) defines and is limited in the quantity of a set frame of video being implemented in the coding nonzero coefficient that will produce.
Described rate control techniques can be carried out in video coding process, and can correct or adjustment in this kind cataloged procedure.For frame of video is implemented coding, for each video blocks in this frame of video, the video encoder 18 of source apparatus 12 all can be implemented estimation in the following way: search the video blocks that is stored in the memory 16 and have or not one or more previous frame of video (perhaps subsequent video frame) of transmission already, to discern a similar video blocks-a be called predicted video block.In some cases, this predicted video block can comprise from last or " optimum prediction " back one frame of video, and the present invention is not limited to this aspect certainly.Video encoder 18 is implemented motion compensation and is created the current video block of an expression desire coding and the difference block of the difference between this optimum prediction.Motion compensation typically refer to use a motion vector take this optimum prediction, in an input block, deduct this optimum prediction to produce the operation of a difference block then.Described rate control techniques defines the quantization parameter (QP) that is used for being used to quantize at video coding process encoded video blocks.As described herein, the present invention is contained the operation parameter equation and is defined in the QP that carries out the speed controlled encoding on video blocks level or the frame level.
After this movement compensation process has been created this difference block, implement a series of extra coding steps usually this difference block is encoded.These extra coding steps can depend upon used coding standard.In meeting the encoder of MPEG-4, for example, described extra coding step can comprise one 8 * 8 discrete cosine transforms, implement scale subsequently and quantize, implement subsequently again a grating to zigzag and reorder, implement subsequently run length encoding, implement huffman coding more subsequently.Yet, in some cases, can use wavelet transformation or other conversion to replace dct transform to video blocks.The quantity that quantizes the back nonzero coefficient is important problem very, because should value can influence for one set or frame are implemented the required bit quantity of coding.
After coding, encoded difference block can be transmitted together with motion vector, this motion vector can distinguish this video blocks and the former frame that is used to encode (or back one frame).By this kind mode, be not with each frame as one independently picture implement coding, video encoder 18 is that the difference between the consecutive frame is implemented coding.These technology can obviously be reduced to accurately represents the required data volume of each frame in the video sequence.
This motion vector can define the location of pixels about the upper left corner of coded video blocks, also can use the motion vector of other form certainly.In a word, video blocks is implemented coding, can obviously be reduced to the required bandwidth of transmitting video data stream by using motion vector.
In some cases, video encoder 18 also can be supported intraframe coding except that supporting the interframe encode.Intraframe coding is to utilize the similitude (being called in space or the frame relevant) in the frame to come further compressed video frame.Compression is normally based on the texture coding that is used to compress rest image in the frame, and for example discrete cosine transform (DCT) is encoded.Compression usually is used in combination with the interframe compression in the frame.
The receiver 22 of receiving system 14 can receive the encoded video data that is the motion vector form and can represent coded video blocks and the encoded difference block of the encoded difference between the used optimum prediction in estimation.Yet, in some cases, be not the transmission motion vector, but the difference between a transmitting moving vector and the motion vector prediction value (MVP) that had before calculated.In a word, decoder 24 can be carried out video decode, comes to be shown to the user by display unit 26 so that produce video sequence.The decoder 24 of receiving system 14 also can be configured to an encoder/decoder (CODEC).In this kind situation, the two all can implement source apparatus 12 and receiving system 14 coding, transmission to digital video sequences, receive and decoding.
According to the present invention, video encoder 18 selects to be used for the quantization parameter (QP) of video coding with a kind of dynamical fashion, so that realize rate controlled coding.QP can define when video blocks is encoded the bit quantity that produces.Thereby selected QP can directly influence the coded bit quantity of per second.When QP increased, the quality of preserving less data and video coding may reduce.And when QP reduced, the quality of preserving more data and video coding improved.Yet if QP is too small, the coded bit quantity of per second may exceed a bandwidth of distributing, and infringement transmits the ability of frame in the bandwidth of this limited size.By selecting QP, can make the bit rate substantially constant of frame of video transmission with a dynamical fashion.
More specifically, video encoder 18 is selected QP according to a parametric equation formula, so that the value of rho (ρ) is mapped to quantization parameter used in video coding (QP).Equally, amount (ρ) representative quantizes the quantity of the nonzero coefficient of rear video piece, and is proportional to the quantity of used position in video coding substantially.Correspondingly, video encoder 18 utilizes the relation between ρ and the QP, so that realize rate controlled video encoding.Because ρ is approximate and the bit rate linear correlation, thereby this ρ-QP parametric equation formula can be used for the QP that selection can obtain the effective speed controlled video encoding.
Can be with this parametric equation formula initialization, but in video coding process, adjusted then, so that the ρ of the prediction of being defined by this parametric equation formula more preferably approaches the actual ρ that obtains from a selected QP.This kind adjustment to this parametric equation formula can be carried out to each video blocks or to each frame.In a preferable embodiment, adopt the control of video blocks stage speed, but on the frame level, adjust the variable in this parametric equation formula.
Fig. 2 is the exemplary calcspar of a device 30, and device 30 can be corresponding to source apparatus 12.Generally speaking, device 30 comprises that one can carry out the digital video apparatus of rate controlled coding techniques as herein described.Device 30 can meet a kind of video encoding standard, for example MPEG-4, ITU H.263, ITU H.264 or the another kind of video encoding standard that can benefit from rate control techniques described herein.
As shown in Figure 2, device 30 comprise one be used for to video sequence implement coding video coding apparatus 32, and one be used for before coding and after store described video sequence video memory 34.Device 30 also can comprise one be used for encoded sequence transmission to the reflector 36 of another device, and may one be used for capture video sequences and institute's capture video sequences be stored in the video capture device 38 (for example video camera) of memory 34.Device 30 various elements can couple with communication mode mutually via a communication bus 35.Also can comprise various other elements at device in 30, for example intraframe coding element, various filter or other element, but specifically show for simplicity's sake.
This searches the subset of pixels that the space can comprise one or more previous frame of video (or subsequent frame).Selected subclass can be identified as in advance a possible position be used to discern one with the closely optimum prediction of coupling of the current video block that will encode.In addition, if use the different levels of searching, then this is searched the space and may be in motion estimation process changes to some extent.In this kind situation, this is searched the space and can diminish gradually aspect space big or small searching, and wherein searching of these back is that resolution with higher than searching of front is implemented.
At last, exercise estimator identification one " optimum prediction "-it is and the video blocks that will the encode candidate video blocks of tight coupling.Yet, should be appreciated that, in many situations, can before optimum prediction, position, and in these cases, can use appropriate matcher to implement coding appropriate matcher.Equally, a predicted video block is meant an appropriate matcher, and this appropriate matcher can be optimum prediction.
In case exercise estimator 40 is for video blocks identifies optimum prediction, motion compensator 42 promptly forms the difference block of the difference between an indication current video block and this optimum prediction.Difference block encoder 44 can be further to this difference block coding, to compress this difference block, and switchable encoded difference block, so that it is transferred to another device together with a motion vector (perhaps this motion vector and a motion vector prediction value is poor), search to discern this which candidate video blocks once was used for implementing coding in space.For simplicity's sake, other assembly that will be used for after motion compensation implementing coding reduces difference block encoder 44, and is different because concrete assembly will look the concrete standard supported.In other words, difference block encoder 44 can be to as described herein and difference block that produce is carried out one or more conventional coding techniquess.
In cataloged procedure, rate controlled can guarantee to be used for can not surpassing the bit quantity that a set sequence implements to encode the bandwidth constraints of a communication port that is associated with reflector 36.For carrying out this kind rate controlled, video encoder 28 comprises a rate controlled unit 46.Rate controlled unit 46 utilizes the coded bit quantity of every frame and quantizes relation between the nonzero coefficient quantity of rear video piece.The representative of the value of ρ quantizes the nonzero coefficient quantity of rear video piece, and it is proportional to the coded bit quantity of every frame (and thereby be proportional to substantially the coded bit quantity of per second under a constant frame rate) substantially.Correspondingly, for realizing rate controlled video encoding, rate controlled unit 46 utilizes the relation between ρ and the QP.
More specifically, the parametric equation formula that a value with ρ maps to different Q P is used in rate controlled unit 46.Then, can be corresponding to the ρ value of the desired speed of video coding and select QP.During cataloged procedure, renewable this parametric equation formula is more preferably to reflect the actual relationship between ρ and the QP.In other words, can be with this parametric equation formula initialization, but in video coding process, adjusted then, so that the ρ of the prediction of being defined by this parametric equation formula more preferably approaches the actual ρ that obtains from a selected QP.This kind adjustment to this parametric equation formula can be carried out to each video blocks or to each frame.In a preferable embodiment, adopt the control of video blocks stage speed, but on the frame level, adjust the variable in this parametric equation formula.
The parametric equation formula that is used to define the ρ of prediction by rate controlled unit 46 can be following form:
ρ＝nc×e
-f(qp)，
Wherein e represents the arithmetic constant of a so-called Euler (Euler) number (2.71828182845904523536028747135), the nonzero coefficient quantity of the video blocks of frame after the quantification of ρ representative prediction, nc represents the number of coefficients (for example quantity of coefficient in a frame or the video blocks) with the unit representation of being considered, and f (qp) is the parametric function of this quantization parameter.For example, f (qp) can comprise a simple multinomial or an exponential function.In some cases, can use different constants to replace the value of e, for example use value 2 or similar value.
Equally, the example of f applicatory (qp) comprises as follows:
f(qp)＝α×qp，
f(qp)＝α×qp+β，
f(qp)＝α×qp
2+β×qp+δ，
Perhaps exponential function:
f(qp)＝α×qp
β。
In these examples, qp represents quantization parameter (being also referred to as QP in this article).Variable α, β and δ comprise programmable parameter, and these programmable parameters can be adjusted in the rate control techniques process, more preferably to approach the relation between ρ and the qp.Particularly, during cataloged procedure, adjust described programmable parameter, so that for different QP, the ρ of prediction is in fact corresponding to actual ρ.
In another example, the used parametric equation formula in rate controlled unit 46 can be in fact corresponding to:
ρ＝nc×2
-α×qp-β，
Wherein ρ defines the ρ of prediction, and nc representative is with the number of coefficients of the unit representation considered and define the maximum of the ρ of prediction, and α and β comprise adjustable parameter.This parametric equation formula can be particularly useful for MPEG-4 and reach H.263 rate controlled.In this kind situation,, this embodiment replaces equation ρ=nc * e with constant 2 by any heavy losses in the curve precision for more effectively can not caused
-f (qp)In Euler's numbers (e).
In addition,, can have precedence over other candidate item of f (qp) and use f (qp)=α * qp+ β, because it represent trading off between curve accuracy and this equational complexity of enforcement also for MPEG-4 and rate controlled H.263.Once the accuracy to other member in the parametric equation formula family (for example exponential equation formula, quadratic polynomial and a polynomial equation formula) was implemented relatively in various simulations.Although find that exponential curve is the most accurate, yet also find to compare with other equations, exponential curve has high-order complexity.Primary parameter equation f (qp)=α * qp+ β, and the difference aspect accuracy can be ignored between the quadratic equation.Yet quadratic equation has 3 unknown quantitys and its enforcement is more complicated, and simple equation only has 2 unknown quantitys and also more uncomplicated.
In certain embodiments, if video encoder produces one ρ-QP table already, technology then of the present invention can be utilized this ρ-QP table.For example, if two unknown quantity α and β need be found the solution in rate controlled unit 48, then it can be at two some places to one ρ-QP table sampling (if obtaining this table already).This first can comprise QP, and this second can be approximately 2 * QP.The exception of this of rate control techniques kind of version can be the situation when QP＞25, for example for MPEG-4 and H.263, and in this kind situation, the point and β can be set at zero of only taking a sample.Other exceptional situation also can be handled in rate controlled unit 48, for example the situation when the tabular value of the tabular value of second sampling place or these two samplings place is zero.Can these two unknown quantity α and β be implemented once to upgrade every frame, enforcement that certainly also can be more frequent or more not frequent in different embodiments is upgraded.
Yet ρ-QP table is not necessary for carrying out rate control techniques described herein.If for example use leaving over of one ρ-QP table can obtain ρ-QP table in the encoder traditionally, then still can utilize ρ-QP to show to quicken selection to adjustable parameter in the parametric equation formula one.Yet, need not to use this table can easily determine described adjustable parameter.
In fact, a main distinction between conventional rate control technology and the parameter rate control techniques described herein is: be not to produce and search in one ρ-QP question blank (LUT) to obtain the optimum value of qp, but can use a for example following parametric equation formula to calculate the value that quantizes parameter q p:
qp＝(-log2(ρ/nc)-β)/α，
Wherein ρ defines the ρ of prediction, and nc representative is with the number of coefficients (for example quantity of coefficient in the frame) of the unit representation considered and define the maximum of the ρ of prediction, and α and β comprise adjustable parameter.Equally, during cataloged procedure, programmable parameter is adjusted, so that for different QP, the ρ of prediction is all in fact corresponding to actual ρ.If obtain one ρ-QP table already, then can use this ρ-QP to show to adjust described programmable parameter, if but do not have one ρ-QP table, then can select described programmable parameter according to the specific assignment sampling point that is used to define the actual ρ that obtains from specific QP.
The similitude of half that f (qp) is used that exponential function intuitive knowledge behind is based on that ρ-QP curve and a Generalized Gaussian (Gaussian) distribute is used once to reach quadratic polynomial function intuitive knowledge behind and then correspond respectively to the curve that use one is similar to laplacian distribution (Laplacian-distribution) or Gaussian Profile.
Fig. 3 is a flow chart, and it shows that a kind of use one predicted value with ρ maps to the frame stage speed control technology of quantization parameter (QP).As shown in Figure 3, rate controlled unit 46 uses, the one parametric equation formula of video encoder 28 is that a frame is determined a QP (301).At first, can use initialization value, so that this parametric equation formula maps to QP with the predicted value of ρ to the programming of this parametric equation formula.Then, the 46 pairs of video blocks in rate controlled unit are implemented coding (302).If this frame comprises the more video blocks that will encode (not (no) branches of 303), then rate controlled unit 46 uses the quantization parameter of this frame that other video blocks are implemented coding (302).
In case arrive the end (303 be (yes) branch) of this frame, just next frame be loaded on and implement coding (304) in the local storage 37.At this moment, rate controlled unit 46 undated parameter equations (305).Particularly, rate controlled unit 46 can determine the to do for oneself actual value of the ρ that the selected QP of video blocks of former frame obtains, the actual value of this ρ can be different from the prediction ρ that is mapped to selected QP by this parametric equation formula slightly.By this kind mode, will use one more the precise parameters equation next frame is implemented coding.
After upgrading this parametric equation formula (305), any subsequent frame is repeated this process (306 be (yes) branch).In this kind situation, use the updated parameters equation to determine a QP (301) as next frame.Can repeat this process to all frames in the video sequence, to realize frame stage speed controlled video encoding to this sequence.Equally, if desired, then this process can be implemented multipass to a set frame of video, for example, when this process begins, does not use the initialized equations formula to come the actual coding video blocks so that define this parametric equation formula more accurately.Yet in most of situations, after several frames, this parametric equation formula will restrain rapidly, thereby need not each frame enforcement multipass coding.
Fig. 4 is a flow chart, and it shows that a kind of use one predicted value with ρ maps to the technology that quantization parameter (QP) carries out the control of video blocks stage speed.As shown in Figure 4, the rate controlled unit 46 operation parameter equations of video encoder 28 are that video blocks is determined a QP (401).At first, can use initialization value, so that this parametric equation formula maps to QP with the predicted value of ρ to the programming of this parametric equation formula.Then, the 46 pairs of video blocks in rate controlled unit are implemented coding (402), and undated parameter equation (403).Particularly, rate controlled unit 46 can determine the to do for oneself actual value of the ρ that one or more fronts selected QP of video blocks obtains in this frame, the actual value of this ρ can be different from the prediction ρ that is mapped to selected QP by this parametric equation formula slightly.By this kind mode, will use one more the precise parameters equation subsequent video blocks is implemented coding.
If this frame comprises the more video blocks that will encode (403 not (no) branches), then rate controlled unit 46 uses the quantization parameter of determining by undated parameter equation that other video blocks are implemented coding (402).Can repeat this process to all video blocks in the frame, to realize video blocks stage speed controlled video encoding to this sequence.As Fig. 3, process shown in Figure 4 can be that a set video blocks is implemented the multipass coding in this kind situation, for example through adjusting to implement the multipass coding, when this process begins, do not use the initialized equations formula to come the actual coding video blocks so that define this parametric equation formula more accurately.Yet in most of situations, after several frames, this parametric equation formula will restrain rapidly, thereby need not each video blocks enforcement multipass coding.
In case arrive the end (404 be (yes) branch) of this frame, then next frame be loaded on and implement coding (405) in the local storage 37.At this moment, the 46 pairs of any subsequent frames in rate controlled unit repeat this process (406 be (yes) branch).For next frame, use nearest institute updated parameters equation to determine a QP (401).
Fig. 5 is a flow chart, the technology that a kind of use of its explanation one parametric equation formula of being adjusted on the frame level is carried out the control of video blocks stage speed.For for example and illustration purpose, will be with parametric equation formula ρ=nc * e
-f (qp)In the situation of the rate controlled unit 46 of programming Fig. 5 is described, wherein e represents Euler's numbers, the pre-quantitation of the nonzero coefficient of the video blocks of frame after the ρ representative quantizes, nc represents the number of coefficients in the video blocks, and f (qp) is the parametric function by f (qp)=α * quantization parameter that qp+ β is given, wherein qp is a quantization parameter, and α and β define the adjustable parameter of this parametric equation formula.Certainly, also can use many other parametric equation formulas, for example above one of each equation that defines or a different parametric equation formula.
As shown in Figure 5, with rate controlled initialization (501).The initial value of α and β is defined in this initialization, and can according to make and testing apparatus 30 during mean value by determined α of emulation and β.For beginning this rate controlled cataloged procedure, a frame is loaded on local storage 37 (502) from video memory 34.The rate controlled unit uses this parametric equation formula to determine that one is used for video blocks is implemented the QP (503) of coding.Then, rate controlled unit 406 uses institute to determine that QP is to this video blocks enforcement encode (504).If this frame comprises the more video blocks that will encode (not (no) branches of 505), then rate controlled unit 46 uses this parametric equation formula to determine QP (503) separately and use QP separately that described video blocks is implemented coding for each video blocks.On this kind meaning, be on the video blocks level, to determine QP, but for a set frame, the parametric equation formula that is used to define QP does not change.
In case arrive the end (505 be (yes) branch) of this frame, the rate controlled unit promptly can the equational adjustable parameter of undated parameter (α and β) (305).Particularly, rate controlled unit 46 can determine the to do for oneself actual value of the ρ that the selected QP of video blocks of former frame obtains, the actual value of this ρ can be different from the prediction ρ that is mapped to selected QP by this parametric equation formula slightly.Rate controlled unit 46 can select new α and β value so that the actual value of ρ is mapped to more accurately the value of QP by the parametric equation formula.By this kind mode, will use one can reflect more preferably that the more precise parameters equation of actual relationship between ρ and the QP is to next frame enforcement coding.
After upgrading the parameter (506) of this parametric equation formula, any subsequent frame in this video sequence is repeated this process (507 be (yes) branch).In this kind situation, next frame is loaded in the local storage 37 (502), and rate controlled unit 46 operation parameter equations are determined QP (503) separately and are used QP separately that described video blocks is implemented coding (504) for each video blocks.
Can repeat this process to all frames in the video sequence,, wherein all implement the adjustment of parametric equation formula for each frame to realize video blocks stage speed controlled video encoding to this sequence.Equally, this process can be implemented multipass to a set frame of video, for example, when this process begins, does not use the initialized equations formula to come the actual coding video blocks so that define the parametric equation formula of preceding several frames in the sequence more accurately.This can improve the accuracy of rate controlled, and cost is that amount of calculation increases.Yet in most of situations, after several frames being implemented coding, this parametric equation formula will converge to an accurate ρ-QP rapidly and concern expression.Therefore, can need not each frame usually and implement multipass.
So far set forth each embodiment.Particularly, the rate control techniques that various use one parametric equation formulas are represented the mapping between quantization parameter QP and the nonzero coefficient quantity ρ has been proposed.Identified some suitable parameters equational one non-exhaustive list that accurately to represent this kind relation.Described technology can be used for limiting the quantity of the quantization DCT coefficient that is used to define coded video blocks, also can use the coefficient of other type certainly, for example through the coefficient of wavelet transformation.In addition, can be only with regard to P-Frames (frame of prediction) aspect or can adopt technology described herein with two aspects of P-Frames with regard to I-Frames (interframe).For example, such as H.263, MPEG-4 and H.264 to wait many kinds of coding standards be to utilize these I-Frames and P-Frames.In this kind situation, can be only with regard to P-Frames (frame of prediction) aspect or can adopt technology described herein with two aspects of P-Frames with regard to I-Frames (interframe).
Parameter rate controlled described herein has the flexibility that control of frame stage speed or the control of video blocks stage speed can be provided.In a preferred embodiment, be that the control of video blocks stage speed is used with the frame level of parametric equation formula is upgraded.In addition, described video blocks stage speed control technology is configurable with regard to can setting two QP side-play amounts (this determines tolerable variable quantity again).Usually, the value of described side-play amount is higher, can guarantee that promptly the variable quantity level is lower, and cost is that video quality is lower.Emulation shows, by rate control techniques as herein described, can significantly reduce variable quantity, keeps the video quality of level much at one simultaneously.In addition, use a parametric equation formula to calculate the comparable generation of qp and/or to search one ρ-QP table more accurate and have more calculating validity for each video blocks.
The techniques described herein can be implemented in hardware, software, firmware or its arbitrary combination.If be implemented in the software, then described technology can relate to a kind of computer-readable media that comprises program code, and this program code can be carried out one or more methods mentioned above when implementing to carry out in the apparatus for encoding at a pair of video sequence.In this kind situation, this computer-readable media can comprise the Synchronous Dynamic Random Access Memory (SDRAM) of random-access memory (ram)-for example, read-only memory (ROM), nonvolatile RAM (NVRAM), EEPROM (Electrically Erasable Programmable Read Only Memo) (EEPROM), FLASH (quickflashing) memory, and similar memory.
This program code can form of computer readable instructions be stored on the memory.In those kind situation, a processor (for example DSP) can be carried out the instruction that is stored in the memory, so that carry out one or more technology as herein described.In some cases, described technology can come the DSP of speech coding process to carry out by call various nextport hardware component NextPorts for a moment.In other situation, this video encoder can be configured to a microprocessor, one or more application specific integrated circuit (ASIC), one or more field programmable gate array (FPGA) or some other hardware-combination of software.These and other embodiment belongs in the category of following claims.
Claims (33)
1. video coding apparatus, it comprises:
Memory, its stores video sequences; And
Encoder, its operation parameter equation select to be used for described video sequence is carried out the quantization parameter (QP) of rate controlled video encoding, described parametric equation formula with the quantification of pre-quantitation after nonzero coefficient (ρ of prediction) map to quantization parameter (QP).
2. video coding apparatus as claimed in claim 1, wherein said encoder are that each different frame of described video sequence is selected QP separately, and use described different each frame QP separately that described video sequence is implemented frame stage speed controlled video encoding.
3. video coding apparatus as claimed in claim 1, wherein said encoder is that each different video blocks of the frame of video in the described video sequence is selected QP separately, and use described different each video blocks separately, described video sequence implemented video blocks stage speed controlled video encoding.
4. video coding apparatus as claimed in claim 1, wherein said parametric equation formula comprises one or more adjustable parameters, and described encoder is determined nonzero coefficient (actual ρ) after the quantification of the one or more actual quantities produce and is adjusted the described adjustable parameter of described parametric equation formula from the QP of one or more selections, so that the ρ of described prediction is more preferably corresponding to described actual ρ.
5. video coding apparatus as claimed in claim 4, wherein said parametric equation formula in fact corresponding to:
ρ＝nc×e
-f(qp)，
Wherein e represents Euler (Euler) number, and ρ defines the ρ of described prediction, and the nc representative is with the number of coefficients of the unit representation considered and define the maximum of the ρ of described prediction, and f (qp) is the parametric function of described quantization parameter (QP).
6. video coding apparatus as claimed in claim 5, wherein f (qp) comprises one in polynomial function and the exponential function.
7. video coding apparatus as claimed in claim 5, wherein f (qp) comprises in fact a polynomial function corresponding to following formula:
f(qp)＝α×qp，
Wherein qp defines adjustable parameter one time corresponding to described quantization parameter (QP) and α.
8. video coding apparatus as claimed in claim 5, wherein f (qp) comprises in fact the quadratic polynomial function corresponding to following formula:
f(qp)＝α×qp+β，
Wherein qp defines adjustable parameter one time corresponding to described quantization parameter (QP) and α, and β defines the secondary adjustable parameter.
9. video coding apparatus as claimed in claim 5, wherein f (qp) comprises in fact the quadratic polynomial function corresponding to following formula:
f(qp)＝α×qp
2+β×qp+δ，
Wherein qp defines adjustable parameter one time corresponding to described quantization parameter (QP) and α, and β defines the secondary adjustable parameter, and δ defines adjustable parameter three times.
10. video coding apparatus as claimed in claim 5, wherein f (qp) comprises in fact the exponential function corresponding to following formula:
f(qp)＝α×qp
β，
Wherein qp defines described adjustable parameter corresponding to described quantization parameter (QP) and α and β.
11. video coding apparatus as claimed in claim 4, wherein said parametric equation formula in fact corresponding to:
ρ＝nc×2
-α×qp-β，
Wherein ρ defines the ρ of described prediction, and nc representative is with the number of coefficients of the unit representation considered and define the maximum of the ρ of described prediction, and α and β comprise described adjustable parameter.
12. a rate controlled video encoding method, it comprises:
The operation parameter equation is selected quantization parameter (QP), described parametric equation formula with the quantification of pre-quantitation after nonzero coefficient (ρ of prediction) map to quantization parameter (QP); And
Use the QP of described selection that one or more video blocks are encoded.
13. method as claimed in claim 12, it further comprises the different Q P that selects plan to be used for the different video frame of video sequence, described video sequence is carried out frame stage speed controlled video encoding.
14. method as claimed in claim 12, it further comprises the different Q P that selects plan to be used for the different video piece of frame of video, video sequence is carried out video blocks stage speed controlled video encoding.
15. method as claimed in claim 12, wherein said parametric equation formula comprises one or more adjustable parameters, and described method further comprises:
Nonzero coefficient (actual ρ) after the quantification of definite one or more actual quantities that from the QP of one or more selections, produce; And
Adjust the adjustable parameter of described parametric equation formula, so that the β of described prediction is more preferably corresponding to described actual ρ.
16. method as claimed in claim 15, wherein said parametric equation formula in fact corresponding to
ρ＝nc×e
-f(qp)，
Wherein e represents Euler (Euler) number, and ρ defines the ρ of described prediction, and the nc representative is with the number of coefficients of the unit representation considered and define the maximum of the ρ of described prediction, and f (qp) is the parametric function of described quantization parameter (QP).
17. method as claimed in claim 16, wherein f (qp) comprises one in polynomial function and the exponential function.
18. method as claimed in claim 16, wherein f (qp) comprises in fact a polynomial function corresponding to following formula:
f(qp)＝α×qp，
Wherein qp defines adjustable parameter one time corresponding to described quantization parameter (QP) and α.
19. method as claimed in claim 16, wherein f (qp) comprises in fact the quadratic polynomial function corresponding to following formula:
f(qp)＝α×qp+β，
Wherein qp defines adjustable parameter one time corresponding to described quantization parameter (QP) and α, and β defines the secondary adjustable parameter.
20. method as claimed in claim 16, wherein f (qp) comprises in fact the quadratic polynomial function corresponding to following formula:
f(qp)＝α×qp
2+β×qp+δ，
Wherein qp defines adjustable parameter one time corresponding to described quantization parameter (QP) and α, and β defines the secondary adjustable parameter, and δ defines adjustable parameter three times.
21. method as claimed in claim 16, wherein f (qp) comprises in fact the exponential function corresponding to following formula:
f(qp)＝α×qp
β，
Wherein qp is corresponding to described quantization parameter (QP), and α and β define described adjustable parameter.
22. method as claimed in claim 15, wherein said parametric equation formula in fact corresponding to:
ρ＝nc×2
-α×qp-β，
Wherein ρ defines the ρ of described prediction, and nc shows with the number of coefficients of the unit representation of being considered from generation to generation and defines the maximum of the ρ of described prediction, and α and β comprise described adjustable parameter.
23. a computer-readable media that comprises program code, described program code can make programmable processor when being performed:
The operation parameter equation is selected quantization parameter (QP), described parametric equation formula with the quantification of pre-quantitation after nonzero coefficient (ρ of prediction) map to quantization parameter (QP); And
Use the QP of described selection that one or more video blocks are encoded.
24. computer-readable media as claimed in claim 23, described medium comprise that further the different Q P of the different video frame that causes described programmable processor selection plan to be used for video sequence is to carry out the instruction of frame stage speed controlled video encoding to described video sequence.
25. computer-readable media as claimed in claim 23, described medium comprise that further the different Q P of the different video piece that causes described programmable processor selection plan to be used for frame of video is to carry out the instruction of video blocks stage speed controlled video encoding to video sequence.
26. computer-readable media as claimed in claim 23, wherein said parametric equation formula comprises one or more adjustable parameters, and described medium further comprise the instruction that makes described programmable processor carry out following operation:
Nonzero coefficient (actual ρ) after the quantification of definite one or more actual quantities that from the QP of one or more selections, produce; And
Adjust the adjustable parameter of described parametric equation formula, so that the ρ of described prediction is more preferably corresponding to described actual ρ.
27. computer-readable media as claimed in claim 26, wherein said parametric equation formula in fact corresponding to
ρ＝nc×e
-f(qp)，
Wherein e represents Euler (Euler) number, and ρ defines the ρ of described prediction, and the nc representative is with the number of coefficients of the unit representation considered and define the maximum of the ρ of described prediction, and f (qp) is the parametric function of described quantization parameter (QP).
28. computer-readable media as claimed in claim 27, wherein f (qp) comprises one in polynomial function and the exponential function.
29. computer-readable media as claimed in claim 27, wherein f (qp) comprises in fact a polynomial function corresponding to following formula:
f(qp)＝α×qp，
Wherein qp is corresponding to described quantization parameter (QP), and α defines adjustable parameter one time.
30. computer-readable media as claimed in claim 27, wherein f (qp) comprises in fact the quadratic polynomial function corresponding to following formula:
f(qp)＝α×qp+β，
Wherein qp is corresponding to described quantization parameter (QP), and α defines adjustable parameter one time, and β defines the secondary adjustable parameter.
31. computer-readable media as claimed in claim 27, wherein f (qp) comprises in fact the quadratic polynomial function corresponding to following formula:
f(qp)＝α×qp
2+β×qp+δ，
Wherein qp is corresponding to described quantization parameter (QP), and α defines adjustable parameter one time, and β defines the secondary adjustable parameter, and δ defines adjustable parameter three times.
32. computer-readable media as claimed in claim 27, wherein f (qp) comprises in fact the exponential function corresponding to following formula:
f(qp)＝α×qp
β，
Wherein qp is corresponding to described quantization parameter (QP), and α and β define described adjustable parameter.
33. computer-readable media as claimed in claim 26, wherein said parametric equation formula in fact corresponding to:
ρ＝nc×2
-α×qp-β，
Wherein ρ defines the ρ of described prediction, and nc shows with the number of coefficients of the unit representation of being considered from generation to generation and defines the maximum of the ρ of described prediction, and α and β comprise described adjustable parameter.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US63183504P | 2004-11-29 | 2004-11-29 | |
| US60/631,835 | 2004-11-29 | ||
| US11/077,697 | 2005-03-10 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101112101A true CN101112101A (en) | 2008-01-23 |
Family
ID=39043163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200580047232 Pending CN101112101A (en) | 2004-11-29 | 2005-11-22 | Rate control techniques for video encoding using parametric equations |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101112101A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012027892A1 (en) * | 2010-09-02 | 2012-03-08 | Intersil Americas Inc. | Rho-domain metrics |
| CN102771123A (en) * | 2010-09-02 | 2012-11-07 | 英特赛尔美国股份有限公司 | Video classification systems and methods |
| CN102036062B (en) * | 2009-09-29 | 2012-12-19 | 华为技术有限公司 | Video coding method and device and electronic equipment |
| CN101656887B (en) * | 2009-09-23 | 2013-04-10 | 杭州华三通信技术有限公司 | Method and device for selecting rate control algorithm |
| CN104769945A (en) * | 2012-09-27 | 2015-07-08 | 奥林奇公司 | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto |
| US9609348B2 (en) | 2010-09-02 | 2017-03-28 | Intersil Americas LLC | Systems and methods for video content analysis |
| CN107454422A (en) * | 2011-11-07 | 2017-12-08 | 英孚布瑞智有限私人贸易公司 | The coding/decoding method of video data |
| CN107465924A (en) * | 2010-06-10 | 2017-12-12 | 汤姆森特许公司 | The method and apparatus that quantization parameter predicted value is determined from multiple adjacent quantization parameters |
-
2005
- 2005-11-22 CN CN 200580047232 patent/CN101112101A/en active Pending
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101656887B (en) * | 2009-09-23 | 2013-04-10 | 杭州华三通信技术有限公司 | Method and device for selecting rate control algorithm |
| CN102036062B (en) * | 2009-09-29 | 2012-12-19 | 华为技术有限公司 | Video coding method and device and electronic equipment |
| CN107465924A (en) * | 2010-06-10 | 2017-12-12 | 汤姆森特许公司 | The method and apparatus that quantization parameter predicted value is determined from multiple adjacent quantization parameters |
| US11722669B2 (en) | 2010-06-10 | 2023-08-08 | Interdigital Vc Holdings, Inc. | Methods and apparatus for determining quantization parameter predictors from a plurality of neighboring quantization parameters |
| US11381818B2 (en) | 2010-06-10 | 2022-07-05 | Interdigital Vc Holdings, Inc. | Methods and apparatus for determining quantization parameter predictors from a plurality of neighboring quantization parameters |
| CN107465924B (en) * | 2010-06-10 | 2021-01-12 | 交互数字Vc控股公司 | Method and apparatus for determining quantization parameter predictor from multiple adjacent quantization parameters |
| US10742981B2 (en) | 2010-06-10 | 2020-08-11 | Interdigital Vc Holdings, Inc. | Methods and apparatus for determining quantization parameter predictors from a plurality of neighboring quantization parameters |
| US9609348B2 (en) | 2010-09-02 | 2017-03-28 | Intersil Americas LLC | Systems and methods for video content analysis |
| WO2012027892A1 (en) * | 2010-09-02 | 2012-03-08 | Intersil Americas Inc. | Rho-domain metrics |
| CN102771123A (en) * | 2010-09-02 | 2012-11-07 | 英特赛尔美国股份有限公司 | Video classification systems and methods |
| CN102714722A (en) * | 2010-09-02 | 2012-10-03 | 英特赛尔美国股份有限公司 | RHO-domain metrics |
| CN107454422A (en) * | 2011-11-07 | 2017-12-08 | 英孚布瑞智有限私人贸易公司 | The coding/decoding method of video data |
| CN107454422B (en) * | 2011-11-07 | 2020-08-07 | 英孚布瑞智有限私人贸易公司 | Method for decoding video data |
| CN104769945B (en) * | 2012-09-27 | 2018-06-22 | 奥林奇公司 | It codes and decodes the method for image, code and decode device, mechanized data medium |
| CN104769945A (en) * | 2012-09-27 | 2015-07-08 | 奥林奇公司 | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100919074B1 (en) | Rate control techniques for video encoding using parametric equations | |
| US11563974B2 (en) | Method and apparatus for video decoding | |
| CN101164345B (en) | Adaptive frame skipping techniques for rate controlled video encoding | |
| CN101112101A (en) | Rate control techniques for video encoding using parametric equations | |
| US11172220B2 (en) | Video encoding method, and storage medium thereof | |
| EP1653744A1 (en) | Non-integer pixel sharing for video encoding | |
| US10873763B2 (en) | Video compression techniques for high dynamic range data | |
| WO2000045602A1 (en) | Optimal encoding scheme for motion compensated video | |
| CN118055253A (en) | Optical flow estimation for motion compensated prediction in video coding | |
| CN101305615A (en) | Multi-dimensional neighboring block prediction for video encoding | |
| KR20020051929A (en) | Method of converting video data streams | |
| US10432952B2 (en) | System and methods for fixed-point approximations in display stream compression (DSC) | |
| CN110876060A (en) | Code rate adjusting method and device in coding process | |
| MXPA05014209A (en) | Multipass video rate control to match sliding window channel constraints. | |
| CN101331773B (en) | Device and method for processing rate controlled for video coding using rate-distortion characteristics | |
| US20100166075A1 (en) | Method and apparatus for coding video image | |
| EP1642464B1 (en) | Method of encoding for handheld apparatuses | |
| CN103650490A (en) | Method and apparatus for motion compensation prediction | |
| US7956898B2 (en) | Digital image stabilization method | |
| CN113473136B (en) | Video encoder and code rate control device thereof | |
| CN100581245C (en) | Efficient rate control techniques for video encoding | |
| CN1848960B (en) | Residual coding in compliance with a video standard using non-standardized vector quantization coder | |
| US6950470B2 (en) | Method and apparatus for optimizing of motion estimation | |
| JP2004518199A (en) | coding | |
| CN100531394C (en) | Method and system for video motion processing in a microprocessor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1110462 Country of ref document: HK |
|
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20080123 |
|
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: WD Ref document number: 1110462 Country of ref document: HK |