WO2012174840A1 - Codage/décodage prédictif inter-image et procédé et appareil de codage/décodage de numéro d'ordre d'image de référence associés - Google Patents

Codage/décodage prédictif inter-image et procédé et appareil de codage/décodage de numéro d'ordre d'image de référence associés Download PDF

Info

Publication number
WO2012174840A1
WO2012174840A1 PCT/CN2011/084466 CN2011084466W WO2012174840A1 WO 2012174840 A1 WO2012174840 A1 WO 2012174840A1 CN 2011084466 W CN2011084466 W CN 2011084466W WO 2012174840 A1 WO2012174840 A1 WO 2012174840A1
Authority
WO
WIPO (PCT)
Prior art keywords
reference frame
optimal
prediction unit
current prediction
probable
Prior art date
Application number
PCT/CN2011/084466
Other languages
English (en)
Inventor
Jie Jia
Yong Ho Cho
Wencheng JIN
Original Assignee
Lg Electronics(China) R&D Center Co., Ltd
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
Priority claimed from CN2011101676234A external-priority patent/CN102843554A/zh
Priority claimed from CN201110167639.5A external-priority patent/CN102843561B/zh
Application filed by Lg Electronics(China) R&D Center Co., Ltd filed Critical Lg Electronics(China) R&D Center Co., Ltd
Publication of WO2012174840A1 publication Critical patent/WO2012174840A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/46—Embedding additional information in the video signal during the compression process
    • H04N19/463—Embedding additional information in the video signal during the compression process by compressing encoding parameters before transmission
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/103—Selection of coding mode or of prediction mode
    • H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/189—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding
    • H04N19/196—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards

Definitions

  • the present invention relates to the technical field of picture video encoding/decoding and inter picture prediction, and more particularly, to inter picture prediction encoding/decoding and method and apparatus for reference frame serial number encoding/decoding thereof.
  • Picture video compression encoding/decoding is generally divided into two prediction technologies for encoding and decoding: inter picture encoding/decoding and intra picture encoding/decoding.
  • the technology of intra picture encoding/decoding performs predictive encoding/decoding on the currently encoded/decoded picture unit by using the information from the encoded/decoded picture unit within the current frame.
  • the technology of inter picture encoding/decoding performs predictive encoding/decoding on the currently encoded/decoded picture by using the information of a picture already encoded/decoded.
  • Embodiments of the present invention provide an inter picture prediction encoding method to enhance inter picture prediction encoding efficiency.
  • the method comprises the steps of obtaining optimal motion data of a current prediction unit in a currently encoded frame, which optimal motion data includes an optimal motion vector and an optimal reference frame serial number; selecting optimal prediction data from prediction data of the optimal motion data; obtaining a residual signal of the optimal motion data, and encoding the residual signal; encoding a serial number of the optimal prediction data.
  • Embodiments of the present invention further provide an inter picture prediction decoding method to enhance inter picture prediction decoding efficiency.
  • the method comprises the steps of decoding to obtain a residual signal of motion data and a serial number of prediction data of a current prediction unit in a currently decoded frame, which motion data includes a motion vector and a reference frame serial number; obtaining the prediction data according to the serial number of the prediction data; obtaining the motion data according to the prediction data and the residual signal.
  • Embodiments of the present invention further provide a video encoder to enhance inter picture prediction encoding efficiency.
  • the video encoder comprises an obtaining module, for obtaining optimal motion data of a current prediction unit in a currently encoded frame, which optimal motion data includes an optimal motion vector and an optimal reference frame serial number; a selecting module, for selecting optimal prediction data from prediction data of the optimal motion data; a first encoding module, for obtaining a residual signal of the optimal motion data, and encoding the residual signal; and a second encoding module, for encoding a serial number of the optimal prediction data.
  • Embodiments of the present invention further provide a video decoder to enhance inter picture prediction decoding efficiency.
  • the video decoder comprises a decoding module, for decoding to obtain a residual signal of motion data and a serial number of prediction data of a current prediction unit in a currently decoded frame, which motion data includes a motion vector and a reference frame serial number; a prediction data obtaining module, for obtaining the prediction data according to the serial number of the prediction data; and a motion data obtaining module, for obtaining the motion data according to the prediction data and the residual signal.
  • Embodiments of the present invention provide a reference frame serial number encoding method of inter picture prediction encoding to enhance inter picture prediction encoding efficiency.
  • the method comprises the steps of performing motion estimation on a current prediction unit to obtain an optimal reference frame of the current prediction unit; calculating a most probable reference frame of the current prediction unit; indicating, if the most probable reference frame is the same as the optimal reference frame, that the most probable reference frame is the same as the optimal reference frame, and terminating reference frame serial number encoding; indicating, if the most probable reference frame is not the same as the optimal reference frame, that the most probable reference frame is not the same as the optimal reference frame, and terminating reference frame serial number encoding after encoding residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame.
  • Embodiments of the present invention further provide a reference frame serial number decoding method in inter picture prediction decoding to enhance inter picture prediction decoding efficiency.
  • the method comprises the steps of decoding to determine whether a most probable reference frame of a current prediction unit is the same as an optimal reference frame of the current prediction unit; calculating the most probable reference frame; determining the most probable reference frame as the optimal reference frame if the most probable reference frame is the same as the optimal reference frame; decoding, if the most probable reference frame is not the same as the optimal reference frame, to obtain residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, and obtaining the optimal reference frame according to the residual values and the most probable reference frame.
  • Embodiments of the present invention further provide a video encoder to enhance inter picture prediction encoding efficiency.
  • the video encoder comprises a motion estimating module, for performing motion estimation on a current prediction unit to obtain an optimal reference frame of the current prediction unit; a calculating module, for calculating a most probable reference frame of the current prediction unit; and an encoding processing module, for indicating, when the most probable reference frame is the same as the optimal reference frame, that the most probable reference frame is the same as the optimal reference frame, and terminating reference frame serial number encoding; indicating, when the most probable reference frame is not the same as the optimal reference frame, that the most probable reference frame is not the same as the optimal reference frame, and terminating reference frame serial number encoding after encoding residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame.
  • Embodiments of the present invention further provide a video decoder to enhance inter picture prediction decoding efficiency.
  • the video decoder comprises a determining module, for decoding to determine whether a most probable reference frame of a current prediction unit is the same as an optimal reference frame of the current prediction unit; a calculating module, for calculating the most probable reference frame; and a decoding processing module, for determining the most probable reference frame as the optimal reference frame when the most probable reference frame is the same as the optimal reference frame; decoding, when the most probable reference frame is not the same as the optimal reference frame, to obtain residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, and obtaining the optimal reference frame according to the residual values and the most probable reference frame.
  • Inter picture prediction encoding obtains optimal motion data of a current prediction unit in a currently encoded frame, which optimal motion data includes an optimal motion vector and an optimal reference frame serial number; selects optimal prediction data from prediction data of the optimal motion data; obtains a residual signal of the optimal motion data, and encodes the residual signal; encodes a serial number of the optimal prediction data, thereby introducing encoding of the reference frame serial number based on the predicted residual into inter picture prediction encoding, and enhancing the inter picture prediction encoding efficiency.
  • Inter picture prediction decoding decodes to obtain a residual signal of motion data and a serial number of prediction data of a current prediction unit in a currently decoded frame, which motion data includes a motion vector and a reference frame serial number; obtains the prediction data according to the serial number of the prediction data; obtains the motion data according to the prediction data and the residual signal, thereby introducing decoding of the reference frame serial number based on the predicted residual into inter picture prediction decoding, and enhancing the inter picture prediction decoding efficiency.
  • Reference frame serial number encoding of inter picture prediction encoding performs motion estimation on a current prediction unit to obtain an optimal reference frame of the current prediction unit; calculates a most probable reference frame of the current prediction unit; indicates, if the most probable reference frame is the same as the optimal reference frame, that the most probable reference frame is the same as the optimal reference frame, and terminates reference frame serial number encoding; indicates, if the most probable reference frame is not the same as the optimal reference frame, that the most probable reference frame is not the same as the optimal reference frame, and terminates reference frame serial number encoding after encoding residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, thereby improving reference frame serial number encoding of existing inter picture prediction encoding, and enhancing the inter picture prediction encoding efficiency.
  • Reference frame serial number decoding in inter picture prediction decoding decodes to determine whether a most probable reference frame of a current prediction unit is the same as an optimal reference frame of the current prediction unit; calculates the most probable reference frame; determines the most probable reference frame as the optimal reference frame if the most probable reference frame is the same as the optimal reference frame; decodes, if the most probable reference frame is not the same as the optimal reference frame, to obtain residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, and obtains the optimal reference frame according to the residual values and the most probable reference frame, thereby improving reference frame serial number decoding in existing inter picture prediction decoding, and enhancing the inter picture prediction decoding efficiency.
  • Fig. 1 is a process flowchart illustrating an inter picture prediction encoding method according to Embodiment One of the present invention
  • Fig. 2 is a process flowchart illustrating an inter picture prediction decoding method according to Embodiment One of the present invention
  • Fig. 3 is a schematic diagram illustrating the structure of a video encoder according to Embodiment One of the present invention.
  • Fig. 4 is a schematic diagram illustrating the structure of a concrete example of the video encoder according to Embodiment One of the present invention.
  • Fig. 5 is a schematic diagram illustrating the structure of a video decoder according to Embodiment One of the present invention
  • Fig. 6 is a process flowchart illustrating a reference frame serial number encoding method of inter picture prediction encoding according to Embodiment Two of the present invention
  • Fig. 7 is a schematic diagram illustrating the positions of a current prediction unit and neighboring reference units as well as reference frame serial numbers according to Embodiment Two of the present invention.
  • Fig. 8 is a process flowchart illustrating a reference frame serial number decoding method in inter picture prediction decoding according to Embodiment Two of the present invention.
  • Fig. 9 is a schematic diagram illustrating the structure of a video encoder according to Embodiment Two of the present invention.
  • Fig. 10 is a schematic diagram illustrating the structure of a video decoder according to Embodiment Two of the present invention.
  • this embodiment introduces encoding/decoding of the reference frame serial number based on the predicted residual into inter picture prediction encoding/decoding.
  • the process flow of the inter picture prediction encoding method according to this embodiment may include the following steps.
  • Step 101 - obtaining optimal motion data of a current prediction unit (PU) in a currently encoded frame, which optimal motion data includes an optimal motion vector and an optimal reference frame serial number;
  • Step 102 selecting optimal prediction data from prediction data of the optimal motion data; during specific execution, the prediction data of the optimal motion data can be regarded as candidates of the prediction data, and the optimal prediction data is selected among these candidates;
  • Step 103 - obtaining a residual signal of the optimal motion data, and encoding the residual signal
  • Step 104 encoding a serial number of the optimal prediction data. Execution of this step enables the decoding end to obtain the prediction data to which the serial number corresponds, and hence to obtain motion data from the prediction data and the residual signal.
  • inter picture prediction encoding obtains optimal motion data of a current prediction unit in a currently encoded frame, which optimal motion data includes an optimal motion vector and an optimal reference frame serial number; selects optimal prediction data from prediction data of the optimal motion data; obtains a residual signal of the optimal motion data, and encodes the residual signal; encodes a serial number of the optimal prediction data, thereby introducing encoding of the reference frame serial number based on the predicted residual into inter picture prediction encoding, and enhancing the inter picture prediction encoding efficiency.
  • the step of selecting optimal prediction data from prediction data of the optimal motion data may include predicting the optimal motion data of the current prediction unit in the currently encoded frame from neighboring motion data, and selecting motion data with the least cost (namely reference motion data with the least cost) in the neighboring motion data as the optimal prediction data.
  • the step of selecting motion data with the least cost in the neighboring motion data as the optimal prediction data can be executed in various modes, for instance, calculating a sum of absolute difference (SAD) between counterparts of the optimal motion data (MVx, MVy, ref_idx) and the neighboring motion data (namely reference motion data); selecting, as the optimal prediction data, motion data in the neighboring motion data to which the least sum of absolute difference (SAD) corresponds (namely the reference motion data to which the least sum of absolute difference corresponds), for further instance, calculating a sum of squared error (SSE) between counterparts of the optimal motion data (MVx, MVy, refjdx) and the neighboring motion data; selecting, as the optimal prediction data, motion data in the neighboring motion data to which the least sum of squared error (SSE) corresponds (namely the reference motion data to which the least sum of squared error corresponds).
  • SAD sum of absolute difference
  • SSE sum of squared error
  • the step of selecting motion data with the least cost in the neighboring motion data as the optimal prediction data can also be calculating rate distortion costs of counterparts of the optimal motion data (MVx, MVy, refjdx) and the neighboring motion data, and simultaneously calculating a bit of an encoding prediction serial number; selecting, as the optimal prediction data, motion data in the neighboring motion data to which the least rate distortion cost corresponds (namely selecting the reference motion data to which the least rate distortion cost corresponds).
  • the bit of the encoding prediction serial number may also include a step of storing the calculated bit of the encoding prediction serial number, in which case, while encoding the serial number of the optimal prediction data, it is possible to directly extract the stored bit of the encoding prediction serial number to encode the serial number of the optimal prediction data, without having to repeatedly calculate the bit of the encoding prediction serial number while encoding the serial number of the optimal prediction data.
  • the inter picture prediction encoding method has corresponding syntax modification of the prediction unit in the high efficiency video coding (HEVC) working scheme - see the detailed modification in Table 1 , in which modified lines are marked up, and encoding of the original reference frame serial numbers is modified to encoding based on predicted residual in the modified sections.
  • HEVC high efficiency video coding
  • Table 1 Syntax Modification of the Prediction Unit prediction_unit( x0, yO, log2PUWidth, log2PUHeight, Partldx , Descriptor
  • This embodiment further provides an inter picture prediction decoding method, as described below. Since the inter picture prediction decoding method is similar to the inter picture prediction encoding method in terms of the principles for solving problems, implementation of the inter picture prediction decoding method may refer to implementation of the inter picture prediction encoding method, as repetition will not be made here.
  • Step 201 - decoding to obtain a residual signal of motion data and a serial number of prediction data of a current prediction unit in a currently decoded frame, which motion data includes a motion vector and a reference frame serial number;
  • Step 202 - obtaining the prediction data according to the serial number of the prediction data
  • Step 203 - obtaining the motion data according to the prediction data and the residual signal.
  • inter picture prediction decoding decodes to obtain a residual signal of motion data and a serial number of prediction data of a current prediction unit in a currently decoded frame, which motion data includes a motion vector and a reference frame serial number; obtains the prediction data according to the serial number of the prediction data; obtains the motion data according to the prediction data and the residual signal, thereby introducing decoding of the reference frame serial number based on the predicted residual into inter picture prediction decoding, and enhancing the inter picture prediction decoding efficiency.
  • the step of obtaining the motion data may include adding the prediction data to the residual signal to obtain the motion data.
  • This embodiment further provides a video encoder and a video decoder, as described below. Since the video encoder and the video decoder are similar to the inter picture prediction encoding and decoding methods in terms of the principles for solving problems, implementations of the video encoder and the video decoder may refer to implementations of the inter picture prediction encoding and decoding methods, as repetition will not be made here.
  • the video encoder may include: an obtaining module 301 , for obtaining optimal motion data of a current prediction unit in a currently encoded frame, which optimal motion data includes an optimal motion vector and an optimal reference frame serial number; a selecting module 302, for selecting optimal prediction data from prediction data of the optimal motion data; a first encoding module 303, for obtaining a residual signal of the optimal motion data, and encoding the residual signal; and a second encoding module 304, for encoding a serial number of the optimal prediction data.
  • the selecting module 302 is specifically used for predicting the optimal motion data of the current prediction unit in the currently encoded frame from neighboring motion data, selecting motion data with the least cost in the neighboring motion data as the optimal prediction data.
  • the selecting module 302 is specifically used for calculating a sum of absolute difference between counterparts of the optimal motion data and the neighboring motion data; selecting, as the optimal prediction data, motion data in the neighboring motion data to which the least sum of absolute difference corresponds.
  • the selecting module 302 is specifically used for calculating a sum of squared error between counterparts of the optimal motion data and the neighboring motion data; selecting, as the optimal prediction data, motion data in the neighboring motion data to which the least sum of squared error corresponds.
  • the selecting module 302 is specifically used for calculating rate distortion costs of counterparts of the optimal motion data and the neighboring motion data, and simultaneously calculating a bit of an encoding prediction serial number; selecting, as the optimal prediction data, motion data in the neighboring motion data to which the least rate distortion cost corresponds.
  • the video encoder as shown in Fig. 3 may further include a storing module 401 , for storing the calculated bit of the encoding prediction serial number.
  • the second encoding module 304 is specifically used for extracting the stored bit of the encoding prediction serial number to encode the serial number of the optimal prediction data.
  • the obtaining module 301 is specifically used for performing motion estimation and motion compensation on the current prediction unit in the currently encoded frame to obtain the optimal motion data.
  • the video decoder may include: a decoding module 501 , for decoding to obtain a residual signal of motion data and a serial number of prediction data of a current prediction unit in a currently decoded frame, which motion data includes a motion vector and a reference frame serial number; a prediction data obtaining module 502, for obtaining the prediction data according to the serial number of the prediction data; and a motion data obtaining module 503, for obtaining the motion data according to the prediction data and the residual signal.
  • the motion data obtaining module 503 is specifically used for adding the prediction data to the residual signal to obtain the motion data.
  • inter picture prediction encoding obtains optimal motion data of a current prediction unit in a currently encoded frame, which optimal motion data includes an optimal motion vector and an optimal reference frame serial number; selects optimal prediction data from prediction data of the optimal motion data; obtains a residual signal of the optimal motion data, and encodes the residual signal; encodes a serial number of the optimal prediction data, thereby introducing encoding of the reference frame serial number based on the predicted residual into inter picture prediction encoding, and enhancing the inter picture prediction encoding efficiency.
  • Inter picture prediction decoding decodes to obtain a residual signal of motion data and a serial number of prediction data of a current prediction unit in a currently decoded frame, which motion data includes a motion vector and a reference frame serial number; obtains the prediction data according to the serial number of the prediction data; obtains the motion data according to the prediction data and the residual signal, thereby introducing decoding of the reference frame serial number based on the predicted residual into inter picture prediction decoding, and enhancing the inter picture prediction decoding efficiency.
  • this embodiment makes improvement over reference frame serial number encoding/decoding in the currently available inter picture prediction encoding/decoding.
  • the process flow of the reference frame serial number encoding method of inter picture prediction encoding may include the following steps:
  • Step 601 performing motion estimation on a current prediction unit to obtain an optimal reference frame of the current prediction unit
  • Step 602 calculating a most probable reference (MPR) frame of the current prediction unit
  • reference frame serial number encoding of inter picture prediction encoding makes improvement over reference frame serial number encoding of the currently available inter picture prediction encoding, and enhances the inter picture prediction encoding efficiency.
  • the step of calculating a most probable reference frame of the current prediction unit can be executed in various modes, for instance, selecting, as the most probable reference frame, a reference frame with the smallest reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, selecting, as the most probable reference frame, a reference frame with a median reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit; alternatively, selecting, as the most probable reference frame, a reference frame with the highest use frequency from reference frames of neighboring reference units of the current prediction unit.
  • Fig. 7 is a schematic diagram illustrating the positions of a current prediction unit and neighboring reference units as well as reference frame serial numbers. Included in Fig. 7 are a current prediction unit (PU) and neighboring reference units (PU_A, PU_B, PU_C, PU_D, PU_E). Reference frame serial numbers of the units include ref _idx_ l0 , ref _ idx _ 11 and ref _ idx _ lc .
  • a reference frame with the smallest reference frame serial number is selected, as the most probable reference frame, from reference frames of neighboring reference units of the current prediction unit, and is expressible as
  • MPR min(ref _ idx _ lx(A), - - -, ref _ idx _ lx(E)) , where ref _ idx _ lx(A), - - - , ref _ idx _ lx(E) both include the reference frame serial numbers ref _ idx_ l0 , ref _idx_ l ⁇ and ref _ idx_ lc of the corresponding units.
  • a reference frame with a median reference frame serial number is selected, as the most probable reference frame, from reference frames of neighboring reference units of the current prediction unit, and is expressible as
  • MPR median(ref _ idx _ lx(A), ⁇ , ref _ idx _ lx(E)) , where ref _ idx _ lx(A), - - - , ref _ idx _ lx(E) both include the reference frame serial numbers ref _ idx _ l0 , ref _ idx _ l ⁇ and ref _ idx _ lc of the corresponding units.
  • the step of selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit may include: determining a size of the current prediction unit according to number of pixels in the current prediction unit; determining sizes of the neighboring reference units of the current prediction unit according to number of pixels in the neighboring reference units of the current prediction unit; wherein the numbers of pixels are proportional to the sizes of the units; that is to say, the more the pixels there are, the bigger the units are; while the less the pixels there are, the smaller the units are.
  • the current prediction unit includes 4/ x AN number of pixels
  • PU A includes 2N x 2N number of pixels
  • PU_B includes 2N x N number of pixels
  • PU_C includes 2N x N number of pixels
  • PU D includes 2N x 2N number of pixels
  • PU E includes 2N x AN number of pixels.
  • the step of selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit may include: selecting from the neighboring reference units of the current prediction unit: a reference unit sized greater than or equal to the current prediction unit; or a reference unit sized smaller than or equal to the current prediction unit; or a reference unit differently sized within a predetermined range from the current prediction unit; determining a reference frame of the selected reference unit as the most probable reference frame.
  • the reference unit differently sized within a predetermined range from the current prediction unit may for example include a reference unit identically sized with the current prediction unit in horizontal direction and/or vertical direction.
  • the step of indicating that the most probable reference frame is the same as the optimal reference frame may include setting a most probable reference frame flag as 1
  • the step of indicating that the most probable reference frame is not the same as the optimal reference frame may include setting a most probable reference frame flag as 0.
  • the specific numerical value of the most probable reference frame flag can be set upon practical demand, 1 and 0 in this context are set merely by way of examples, and other values can of course be set to indicate whether or not the most probable reference frame is the same as the optimal reference frame.
  • the most probable reference frame flag as mpr _ flag _ lx ; if the most probable reference frame is the same as the optimal reference frame, mpr _ flag _lx is set as 1 , and reference frame serial number encoding is terminated; if the most probable reference frame is not the same as the optimal reference frame, mpr _ flag _ lx is set as 0, and reference frame serial number encoding is terminated after encoding residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame.
  • the process of encoding the reference frame serial number further includes encoding the value of mpr _ flag _ lx after setting the value of the mpr _ flag _ lx .
  • the determination as to whether the most probable reference frame is the same as the optimal reference frame can be based on determining whether the serial number of the most probable reference frame is the same as the serial number of the optimal reference frame. In other words, if the serial number of the most probable reference frame is the same as the serial number of the optimal reference frame, this indicates that the most probable reference frame is the same as the optimal reference frame; if the serial number of the most probable reference frame is not the same as the serial number of the optimal reference frame, this indicates that the most probable reference frame is not the same as the optimal reference frame.
  • the reference frame serial number encoding method of inter picture prediction encoding has corresponding syntax modification of the prediction unit in the high efficiency video coding (HEVC) working scheme - see the detailed modification in Table 2, in which modified lines are marked up, and encoding of the original reference frame serial numbers is modified to encoding based on the most probable reference frame serial number in the modified sections.
  • HEVC high efficiency video coding
  • This embodiment further provides a reference frame serial number decoding method in inter picture prediction decoding, as described below. Since the reference frame serial number decoding method in inter picture prediction decoding is similar to the reference frame serial number encoding method of inter picture prediction encoding in terms of the principles for solving problems, implementation of the reference frame serial number decoding method in inter picture prediction decoding may refer to implementation of the reference frame serial number encoding method of inter picture prediction encoding, as repetition will not be made here.
  • the process flow of the reference frame serial number decoding method in inter picture prediction decoding may include the following steps: Step 801 - decoding to determine whether a most probable reference frame of a current prediction unit is the same as an optimal reference frame of the current prediction unit;
  • Step 802 calculating the most probable reference frame
  • Step 803 determining the most probable reference frame as the optimal reference frame if the most probable reference frame is the same as the optimal reference frame; decoding, if the most probable reference frame is not the same as the optimal reference frame, to obtain residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, and obtaining the optimal reference frame according to the residual values and the most probable reference frame.
  • reference frame serial number decoding in inter picture prediction decoding makes improvement over reference frame serial number decoding in the currently available inter picture prediction decoding, and enhances the inter picture prediction decoding efficiency.
  • the step of calculating the most probable reference frame may include selecting, as the most probable reference frame, a reference frame with the smallest reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, selecting, as the most probable reference frame, a reference frame with a median reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit; alternatively, selecting, as the most probable reference frame, a reference frame with the highest use frequency from reference frames of neighboring reference units of the current prediction unit.
  • the step of selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit may include determining a size of the current prediction unit according to number of pixels in the current prediction unit; determining sizes of the neighboring reference units of the current prediction unit according to number of pixels in the neighboring reference units of the current prediction unit; wherein the numbers of pixels are proportional to the sizes of the units.
  • the step of selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit may include selecting from the neighboring reference units of the current prediction unit: a reference unit sized greater than or equal to the current prediction unit; or a reference unit sized smaller than or equal to the current prediction unit; or a reference unit differently sized within a predetermined range from the current prediction unit; determining a reference frame of the selected reference unit as the most probable reference frame.
  • the reference unit differently sized within a predetermined range from the current prediction unit may include a reference unit identically sized with the current prediction unit in horizontal direction and/or vertical direction.
  • the step of decoding to determine whether a most probable reference frame of a current prediction unit is the same as an optimal reference frame of the current prediction unit may include determining that the most probable reference frame is the same as the optimal reference frame if a most probable reference frame flag obtained by decoding is 1 ; determining that the most probable reference frame is not the same as the optimal reference frame if the most probable reference frame flag obtained by decoding is 0.
  • This embodiment further provides a video encoder and a video decoder, as described below. Since the video encoder and the video decoder are similar to the reference frame serial number encoding and decoding methods in inter picture prediction encoding and decoding in terms of the principles for solving problems, implementations of the video encoder and the video decoder may refer to implementations of the reference frame serial number encoding and decoding methods in inter picture prediction encoding and decoding, as repetition will not be made here.
  • the video encoder may include: a motion estimating module 901 , for performing motion estimation on a current prediction unit to obtain an optimal reference frame of the current prediction unit; a calculating module 902, for calculating a most probable reference frame of the current prediction unit; and an encoding processing module 903, for indicating, if the most probable reference frame is the same as the optimal reference frame, that the most probable reference frame is the same as the optimal reference frame, and terminating reference frame serial number encoding; indicating, if the most probable reference frame is not the same as the optimal reference frame, that the most probable reference frame is not the same as the optimal reference frame, and terminating reference frame serial number encoding after encoding residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame.
  • the calculating module 902 may include a first calculating unit, for selecting, as the most probable reference frame, a reference frame with the smallest reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, a second calculating unit, for selecting, as the most probable reference frame, a reference frame with a median reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, a third calculating unit, for selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit; alternatively, a fourth calculating unit, for selecting, as the most probable reference frame, a reference frame with the highest use frequency from reference frames of neighboring reference units of the current prediction unit.
  • the third calculating unit is specifically used for determining a size of the current prediction unit according to number of pixels in the current prediction unit; determining sizes of the neighboring reference units of the current prediction unit according to number of pixels in the neighboring reference units of the current prediction unit; wherein the numbers of pixels are proportional to the sizes of the units.
  • the third calculating unit is specifically used for selecting from the neighboring reference units of the current prediction unit: a reference unit sized greater than or equal to the current prediction unit; or a reference unit sized smaller than or equal to the current prediction unit; or a reference unit differently sized within a predetermined range from the current prediction unit; determining a reference frame of the selected reference unit as the most probable reference frame.
  • the third calculating unit is specifically used for selecting from the neighboring reference units of the current prediction unit: a reference unit identically sized with the current prediction unit in horizontal direction and/or vertical direction.
  • the encoding processing module 903 is specifically used for setting a most probable reference frame flag as 1 to indicate that the most probable reference frame is the same as the optimal reference frame; setting the most probable reference frame flag as 0 to indicate that the most probable reference frame is not the same as the optimal reference frame. As shown in Fig.
  • the video decoder may include: a determining module 1001 , for decoding to determine whether a most probable reference frame of a current prediction unit is the same as an optimal reference frame of the current prediction unit; a calculating module 1002, for calculating the most probable reference frame; and a decoding processing module 1003, for determining the most probable reference frame as the optimal reference frame when the most probable reference frame is the same as the optimal reference frame; decoding, when the most probable reference frame is not the same as the optimal reference frame, to obtain residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, and obtaining the optimal reference frame according to the residual values and the most probable reference frame.
  • the calculating module 1002 may include a first calculating unit, for selecting, as the most probable reference frame, a reference frame with the smallest reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, a second calculating unit, for selecting, as the most probable reference frame, a reference frame with a median reference frame serial number from reference frames of neighboring reference units of the current prediction unit; alternatively, a third calculating unit, for selecting the most probable reference frame from reference frames of neighboring reference units of the current prediction unit according to sizes of the current prediction unit and the neighboring reference units of the current prediction unit; alternatively, a fourth calculating unit, for selecting, as the most probable reference frame, a reference frame with the highest use frequency from reference frames of neighboring reference units of the current prediction unit.
  • the third calculating unit is specifically used for determining a size of the current prediction unit according to number of pixels in the current prediction unit; determining sizes of the neighboring reference units of the current prediction unit according to number of pixels in the neighboring reference units of the current prediction unit; wherein the numbers of pixels are proportional to the sizes of the units.
  • the third calculating unit is specifically used for selecting from the neighboring reference units of the current prediction unit: a reference unit sized greater than or equal to the current prediction unit; or a reference unit sized smaller than or equal to the current prediction unit; or a reference unit differently sized within a predetermined range from the current prediction unit; determining a reference frame of the selected reference unit as the most probable reference frame.
  • the third calculating unit is specifically used for selecting from the neighboring reference units of the current prediction unit: a reference unit identically sized with the current prediction unit in horizontal direction and/or vertical direction.
  • the determining module 1001 is specifically used for determining that the most probable reference frame is the same as the optimal reference frame when a most probable reference frame flag obtained by decoding is 1 ; and determining that the most probable reference frame is not the same as the optimal reference frame when the most probable reference frame flag obtained by decoding is 0.
  • reference frame serial number encoding of inter picture prediction encoding performs motion estimation on a current prediction unit to obtain an optimal reference frame of the current prediction unit; calculates a most probable reference frame of the current prediction unit; indicates, if the most probable reference frame is the same as the optimal reference frame, that the most probable reference frame is the same as the optimal reference frame, and terminates reference frame serial number encoding; indicates, if the most probable reference frame is not the same as the optimal reference frame, that the most probable reference frame is not the same as the optimal reference frame, and terminates reference frame serial number encoding after encoding residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, thereby improving reference frame serial number encoding of existing inter picture prediction encoding, and enhancing the inter picture prediction encoding efficiency.
  • Reference frame serial number decoding in inter picture prediction decoding decodes to determine whether a most probable reference frame of a current prediction unit is the same as an optimal reference frame of the current prediction unit; calculates the most probable reference frame; determines the most probable reference frame as the optimal reference frame if the most probable reference frame is the same as the optimal reference frame; decodes, if the most probable reference frame is not the same as the optimal reference frame, to obtain residual values of a serial number of the most probable reference frame and a serial number of the optimal reference frame, and obtains the optimal reference frame according to the residual values and the most probable reference frame, thereby improving reference frame serial number decoding in existing inter picture prediction decoding, and enhancing the inter picture prediction decoding efficiency.
  • the embodiments of the present invention can be provided as a method, system, or computer program product. Therefore, the present invention can adopt the form of hardware-only embodiment, software-only embodiment, or the software-hardware combined embodiment. Moreover, the present invention can adopt the form of computer program product implemented on one or more computer applicable storage media (comprising, but not be limited to, disc storage, CD-ROM and optical memory, etc.) containing the computer applicable program codes therein.
  • the present invention is described with reference to the flowcharts and/or block diagrams of the method, apparatus (system) and computer program product according to the embodiments of the present invention.
  • each process and/or block in the flowchart and/or block diagram, and combination of the flow and/or block in the flowchart and/or block diagram can be implemented by the computer program instructions.
  • the computer program instructions can be provided to processors of the general-purpose computers, special purpose computers, embedded processors or other programmable data processing apparatuses to generate one machine, so as to generate a device for implementing the function specified in one or more flows in the flowchart and/or one or more blocks in the block diagrams by the instructions executed by the processors of the computers or other programmable data processing apparatuses.
  • the computer program instructions can be also stored in the computer readable memory enabling the computer or other programmable data processing apparatuses to operate in a specific way, so that the instructions stored in the computer readable memory generate the manufacturing product containing the instruction device which implements the functions specified in one or more flows in the flowcharts and/or one or more blocks in the block diagrams.
  • the computer program instructions can be also loaded onto the computer or other programmable data processing apparatuses, such that the computer or other programmable apparatuses execute a series of operation steps to generate the processing implemented by the computer, so as to provide the step for implementing the functions specified in one or more flows in the flowcharts and/or one or more blocks in the block diagrams by the instructions executed on the computers or other programmable apparatuses.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Computing Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

Cette invention porte sur un codage/décodage prédictif inter-image et sur un procédé et un appareil de codage/décodage de numéro d'ordre d'image de référence associés. Le procédé de codage prédictif inter-image comprend les étapes consistant à obtenir des données de mouvement optimal, qui comprennent un vecteur de mouvement optimal et un numéro d'ordre d'image de référence optimal, d'une unité de prédiction courante dans une image en cours de codage ; sélectionner des données de prédiction optimale parmi des données de prédiction des données de mouvement optimal ; obtenir un signal de résidu des données de mouvement optimal et coder le signal de résidu ; coder un numéro d'ordre des données de prédiction optimale. Le procédé de codage du numéro d'ordre d'image de référence de codage prédictif inter-image comprend les étapes consistant à obtenir une image de référence optimale d'une unité de prédiction courante ; calculer une image de référence la plus probable de l'unité de prédiction courante ; indiquer, si l'image de référence la plus probable est la même que l'image de référence optimale, que les deux sont les mêmes, et terminer le codage de numéro d'ordre d'image de référence ; indiquer, si l'image de référence la plus probable n'est pas la même que l'image de référence optimale, que les deux ne sont pas les mêmes, et terminer le codage de numéro d'ordre d'image de référence après codage de valeurs de résidu d'un numéro d'ordre de l'image de référence la plus probable et d'un numéro d'ordre de l'image de référence optimale. Les efficacités de codage/décodage prédictif inter-image peuvent être améliorées.
PCT/CN2011/084466 2011-06-21 2011-12-22 Codage/décodage prédictif inter-image et procédé et appareil de codage/décodage de numéro d'ordre d'image de référence associés WO2012174840A1 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN2011101676234A CN102843554A (zh) 2011-06-21 2011-06-21 帧间图像预测编解码方法及视频编解码器
CN201110167639.5 2011-06-21
CN201110167623.4 2011-06-21
CN201110167639.5A CN102843561B (zh) 2011-06-21 2011-06-21 帧间图像预测编解码的参考帧序号编解码方法及编解码器

Publications (1)

Publication Number Publication Date
WO2012174840A1 true WO2012174840A1 (fr) 2012-12-27

Family

ID=47422014

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/084466 WO2012174840A1 (fr) 2011-06-21 2011-12-22 Codage/décodage prédictif inter-image et procédé et appareil de codage/décodage de numéro d'ordre d'image de référence associés

Country Status (1)

Country Link
WO (1) WO2012174840A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1801945A (zh) * 2005-01-07 2006-07-12 株式会社东芝 编码视频序列变换装置和编码视频序列变换方法
CN101459842A (zh) * 2008-12-17 2009-06-17 浙江大学 一种空间降采样解码方法和装置
CN101682778A (zh) * 2007-06-08 2010-03-24 三星电子株式会社 使用基于对象边缘的分区进行编码和解码的方法和设备
CN101766030A (zh) * 2007-07-31 2010-06-30 三星电子株式会社 使用加权预测的视频编码和解码方法以及设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1801945A (zh) * 2005-01-07 2006-07-12 株式会社东芝 编码视频序列变换装置和编码视频序列变换方法
CN101682778A (zh) * 2007-06-08 2010-03-24 三星电子株式会社 使用基于对象边缘的分区进行编码和解码的方法和设备
CN101766030A (zh) * 2007-07-31 2010-06-30 三星电子株式会社 使用加权预测的视频编码和解码方法以及设备
CN101459842A (zh) * 2008-12-17 2009-06-17 浙江大学 一种空间降采样解码方法和装置

Similar Documents

Publication Publication Date Title
US20200162754A1 (en) Method of Motion Vector Derivation for Video Coding
JP7368554B2 (ja) Ｄｍｖｒのためのブロックサイズ制限
CN110169061B (zh) 编解码电子装置与方法
EP3130147B1 (fr) Procédés de prédiction et de décodage de vecteur de bloc pour codage à mode de copie de bloc intra
CN116915983B (zh) 用于视频解码的方法、计算设备、存储介质和程序产品
US20190058896A1 (en) Method and apparatus of video coding with affine motion compensation
CN103069800B (zh) 用于编码视频的方法和设备、以及解码方法和设备
WO2016078599A1 (fr) Procédé de commande de résolution de vecteur de mouvement et de vecteur de bloc
CN112400319A (zh) 具有多重假设之合并候选者
US9749657B2 (en) Buffer compression for motion vector competition
EP3646596A1 (fr) Procédé et appareil de codage ou de décodage de données vidéo en mode fruc avec des accès mémoire réduits
US10298951B2 (en) Method and apparatus of motion vector prediction
CN113853795A (zh) 发信运动向量差值推导
CN114009033A (zh) 用于用信号通知对称运动矢量差模式的方法和装置
CN115104309A (zh) 对merge模式的改进
CN111095932A (zh) 使用帧速率上转换工具用于改进的压缩/解压缩的方法和装置
CN113302935A (zh) 视频编码/解码系统中的端运动细化
EP4162682A1 (fr) Mode de partition géométrique comportant un affinement de vecteur de mouvement
WO2015008339A1 (fr) Dispositif de codage d'image vidéo, procédé de codage d'image vidéo, dispositif de décodage d'image vidéo et procédé de décodage d'image vidéo
WO2012174840A1 (fr) Codage/décodage prédictif inter-image et procédé et appareil de codage/décodage de numéro d'ordre d'image de référence associés
Chen et al. Motion hooks for the multiview extension of HEVC
KR20140097997A (ko) 움직임 정보 부호화/복호화 장치 및 움직임 정보를 부호화/복호화하는 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11868102

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11868102

Country of ref document: EP

Kind code of ref document: A1