WO2019233423A1 - 获取运动矢量的方法和装置 - Google Patents

获取运动矢量的方法和装置 Download PDF

Info

Publication number
WO2019233423A1
WO2019233423A1 PCT/CN2019/090031 CN2019090031W WO2019233423A1 WO 2019233423 A1 WO2019233423 A1 WO 2019233423A1 CN 2019090031 W CN2019090031 W CN 2019090031W WO 2019233423 A1 WO2019233423 A1 WO 2019233423A1
Authority
WO
WIPO (PCT)
Prior art keywords
block
processed
target
sub
motion vector
Prior art date
Application number
PCT/CN2019/090031
Other languages
English (en)
French (fr)
Inventor
陈焕浜
赵寅
杨海涛
陈建乐
Original Assignee
华为技术有限公司
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 to JP2020567611A priority Critical patent/JP7177181B2/ja
Priority to SG11202012078PA priority patent/SG11202012078PA/en
Priority to BR112020024666-2A priority patent/BR112020024666A2/pt
Priority to KR1020237038069A priority patent/KR20230156446A/ko
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to FIEP19814727.4T priority patent/FI3796657T3/fi
Priority to KR1020207037727A priority patent/KR102600071B1/ko
Priority to DK19814727.4T priority patent/DK3796657T3/da
Priority to MX2020013131A priority patent/MX2020013131A/es
Priority to EP19814727.4A priority patent/EP3796657B1/en
Priority to EP24150170.9A priority patent/EP4369711A2/en
Publication of WO2019233423A1 publication Critical patent/WO2019233423A1/zh
Priority to US17/111,376 priority patent/US11463721B2/en
Priority to US17/898,044 priority patent/US11812050B2/en
Priority to JP2022180472A priority patent/JP2023017955A/ja
Priority to US18/485,330 priority patent/US20240121421A1/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102Methods 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
    • H04N19/103Selection of coding mode or of prediction mode
    • H04N19/105Selection 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
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/136Incoming video signal characteristics or properties
    • H04N19/137Motion inside a coding unit, e.g. average field, frame or block difference
    • H04N19/139Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169Methods 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
    • H04N19/17Methods 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
    • H04N19/176Methods 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
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/189Methods 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
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/42Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
    • H04N19/43Hardware specially adapted for motion estimation or compensation
    • H04N19/433Hardware specially adapted for motion estimation or compensation characterised by techniques for memory access
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51Motion estimation or motion compensation
    • H04N19/513Processing of motion vectors
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51Motion estimation or motion compensation
    • H04N19/56Motion estimation with initialisation of the vector search, e.g. estimating a good candidate to initiate a search
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51Motion estimation or motion compensation
    • H04N19/567Motion estimation based on rate distortion criteria

Definitions

  • the present application relates to the technical field of video coding and decoding, and more particularly, to a method and an apparatus for acquiring a motion vector.
  • Digital video capabilities can be incorporated into a wide variety of devices, including digital television, digital live broadcast systems, wireless broadcast systems, personal digital assistants (PDAs), laptop or desktop computers, tablet computers, electronics Book readers, digital cameras, digital recording devices, digital media players, video game devices, video game consoles, cellular or satellite radio phones (so-called "smart phones"), video teleconferencing devices, video streaming devices And similar.
  • Digital video devices implement video compression technology, for example, in standards defined by MPEG-2, MPEG-4, ITU-T H.263, ITU-T H.264 / MPEG-4 Part 10 Advanced Video Coding (AVC), Video coding standard H.265 / High Efficiency Video Coding (HEVC) standard and the video compression technology described in the extension of such standards.
  • Video devices can implement such video compression techniques to more efficiently transmit, receive, encode, decode, and / or store digital video information.
  • Video compression techniques perform spatial (intra-image) prediction and / or temporal (inter-image) prediction to reduce or remove redundancy inherent in video sequences.
  • a video slice ie, a video frame or a portion of a video frame
  • image blocks can also be referred to as tree blocks, coding units (CU), and / Or coding nodes.
  • Spatial predictions about reference samples in neighboring blocks in the same image are used to encode image blocks in the to-be-encoded (I) slice of the image.
  • the image blocks in the to-be-intercoded (P or B) slice of the image may use spatial prediction relative to reference samples in neighboring blocks in the same image or temporal prediction relative to reference samples in other reference images.
  • An image may be referred to as a frame, and a reference image may be referred to as a reference frame.
  • various video coding standards including the HEVC standard have proposed predictive coding modes for image blocks, that is, predicting a current block to be coded based on already encoded video data blocks.
  • intra prediction mode the current block is predicted based on one or more previously decoded neighboring blocks in the same image as the current block; in inter prediction mode, the current block is predicted based on already decoded blocks in different images.
  • inter prediction mode the current block is predicted based on already decoded blocks in different images.
  • all pixels in the CU use the same motion information for motion compensation to obtain the predicted values of the pixels in the CU.
  • the pixels in the CU do not necessarily have the same motion characteristics. Therefore, using the same motion information to predict all the pixels in the CU may reduce the accuracy of motion compensation, and thus increase the residual information.
  • an advanced scheme proposes advanced temporal motion vector prediction (advanced temporal vector prediction) technology.
  • the prediction process using ATMVP technology mainly includes:
  • the range of the corresponding subblock determined in the ATMVP technology in the corresponding image is uncertain, and the motion information of the corresponding subblock needs to be frequently read from the corresponding image, which will increase the memory read bandwidth and increase the complexity of the prediction process.
  • the present application provides a method and a device for obtaining a motion vector, so as to reduce the number of memory reads.
  • a method for obtaining a motion vector includes: determining a first value interval and a second value interval; determining a target offset vector and identification information of a target image of a block to be processed,
  • the block contains at least one sub-block to be processed; the corresponding sub-block of the sub-block to be processed is determined in the target image according to the position of the sub-block to be processed and the target offset vector; the motion of the sub-block to be processed is determined according to the motion vector of the corresponding sub-block Vector.
  • the value of the component of the target offset vector in the first direction is within a first value interval
  • the value of the component of the target offset vector in the second direction is within a second value interval.
  • the direction is orthogonal to the second direction.
  • At least one sub-block to be processed may be obtained by dividing the block to be processed again. Further, when the block to be processed is a coding unit CU, the sub-block to be processed is a sub-CU obtained by dividing the CU.
  • the (image) block corresponding to the block to be processed in the target image may be referred to as a corresponding block, and the corresponding block includes at least one corresponding subblock.
  • the target offset vector is used to determine the corresponding block in the target image. Block (location). It should be understood that the target offset vector here is different from the normal motion vector.
  • the target offset vector here is a vector used to determine the corresponding block, and the motion vector is generally used to determine the predicted block of an image block, so as to obtain the Predicted pixel values of an image block
  • the offset vector can also be called a temporal vector, and the target image can also be called a corresponding image.
  • the first value interval and the second value interval may be indicated by using the same syntax element.
  • the first value interval includes a first threshold and a second threshold
  • the second value interval includes a third threshold and a fourth threshold.
  • the first threshold can be obtained by parsing a syntax element.
  • the second threshold value, the third threshold value, and the fourth threshold value have a certain multiple relationship (or other numerical relationship) with the first threshold value. In this way, after the first threshold value is obtained, the second threshold value, the fourth threshold value, and the first threshold value can be obtained. Multiple relationship to obtain the second threshold to the fourth threshold.
  • the identification information of the target image is an image sequence number POC of an image frame where the target image is located.
  • the foregoing POC is used to indicate the actual position of the image in the original image sequence.
  • the POC may also be referred to as a display order or a playback order.
  • the identification information of the target image is index information or an index value of an image frame where the target image is located.
  • the sub-blocks to be processed determined in the target image determined according to the target offset vector are also located in a certain range of the target image, and thus can be By pre-reading the time-domain motion vector field within a certain range of the target image, the memory bandwidth and the number of memory reads are reduced.
  • the above method before determining the motion vector of the sub-block to be processed according to the motion vector of the corresponding sub-block, the above method further includes: obtaining pixels in a preset region in the target image Motion vector, the preset region is a possible occurrence range of a corresponding sub-block of the sub-block to be processed.
  • the memory bandwidth can be reduced.
  • the first value range may be a value interval composed of a first threshold and a second threshold
  • the second value range may be a value interval composed of a third threshold and a fourth threshold.
  • a threshold is smaller than the second threshold
  • the third threshold is smaller than the fourth threshold.
  • the first value interval is (first threshold, second threshold), (first threshold, second threshold), [first threshold, second threshold], and [first threshold, second threshold] Any of them.
  • the second value interval is (third threshold, fourth threshold), (third threshold, fourth threshold), [third threshold, fourth threshold], and [third threshold, fourth threshold] Any of them.
  • the first value interval and the second value interval may be any one of an open interval, a closed interval, and a half-open half-closed interval.
  • the value of the component of the target offset vector in the first direction includes within the first value interval: the target offset vector is in The value of the component in the first direction is larger than the first threshold and smaller than the second threshold.
  • the value of the component of the target offset vector in the first direction includes in the first value interval: the target offset vector is on the first side
  • the value of the upward component is greater than or equal to the first threshold and less than or equal to the second threshold.
  • determining the target offset vector of the block to be processed includes: determining an initial offset vector of the block to be processed; when the component of the initial offset vector in the first direction When the value of is in the first value interval and the value of the component in the second direction is in the second value interval, the initial offset vector is determined as the target offset vector; when the initial offset vector is in the first When the value of the component in one direction and / or the second direction exceeds the first value interval and / or the second value interval, the value of the component of the initial offset vector in the first direction and / or the second direction is taken.
  • the part whose value exceeds the first value interval and / or the second value interval is subjected to clamping processing, and the initial offset vector after the clamping processing is determined as the target offset vector.
  • the values of the components of the clamped initial offset vector in the first direction and the second direction are respectively in the first value interval and the second value interval.
  • the values of the components of the target offset vector in the first direction and the second direction that are finally obtained can be in the first and second directions, respectively.
  • the values of the components of the target offset vector in the first direction and the second direction that are finally obtained can be in the first and second directions, respectively.
  • the initial offset vector when the value of the component of the initial offset vector in the first direction and / or the second direction exceeds the first value interval and / or the second value interval, the initial offset vector is set in the first direction. And / or the component in the second direction whose value exceeds the first value interval and / or the second value interval is clamped, and the initial offset vector after the clamp processing is determined as the target offset vector ,include:
  • the initial offset vector is at the first value interval.
  • the component in the two directions that exceeds the second value interval is subjected to clamping processing, and the initial offset vector after the clamping processing is determined as the target offset vector;
  • the initial offset vector in the first direction When the value of the component of the initial offset vector in the first direction is outside the first value interval and the value of the component in the second direction is within the second value interval, the initial offset vector is in the first direction.
  • the component in one direction that exceeds the first value interval is subjected to clamping processing, and the initial offset vector after the clamping processing is determined as the target offset vector;
  • the initial offset vector is in the first direction.
  • the components in one direction that exceed the first value interval are clamped, and the components in the second direction that the initial offset vector exceeds the second value interval are clamped.
  • the offset vector is determined as the target offset vector.
  • the clamping process when the component value of the initial offset vector in a certain direction exceeds the maximum value that can be obtained in that direction, the clamping process is actually truncating the component value exceeding the maximum value, so that the truncated component value Is less than or equal to the maximum value; when the component value of the initial offset vector in a certain direction is less than the minimum value that can be obtained in that direction, the clamping process is actually to compensate the component value, that is, to increase the component value Size such that the component value after compensation is greater than or equal to the minimum value.
  • the above determining the initial offset vector of the block to be processed includes: determining a motion vector of a preset spatial domain neighboring block of the block to be processed as the initial offset vector. .
  • determining a motion vector of a preset spatial-domain neighboring block of the block to be processed as the initial offset vector includes: determining a motion vector of the first available spatial-domain neighboring block as the initial offset vector according to a preset order. Offset vector.
  • Directly determining the motion vector of the first available spatially adjacent block as the initial offset vector can simplify the process of determining the initial offset vector.
  • the determining the initial offset vector of the block to be processed includes: determining a target spatial block from a preset spatial block adjacent block of the block to be processed, where: The value of the component of the motion vector of the adjacent block in the target space in the first direction is within the first value interval, and the value of the component in the second direction is in the second value interval; Motion vector as the target offset vector.
  • the process of determining the target motion vector can be simplified.
  • determining the first value interval and the second value interval includes: according to the position and size of the block to be processed, and the coding tree unit CTU of the block to be processed.
  • determining the first value interval according to the position and size of the block to be processed, the position and size of the coding tree unit CTU where the block to be processed is located, and a preset extension range of the time domain motion vector field of the CTU includes: The first threshold value and the second threshold value are determined according to the position and size of the block to be processed, the position and size of the coding tree unit CTU where the block to be processed is located, and the extension range of the time domain motion vector field of the preset CTU, where the first threshold value Is the minimum value that can be achieved by the component of the target offset vector in the first direction, and the second threshold is the maximum value that can be achieved by the component of the target offset vector in the first direction; according to the first threshold and the second threshold, the first A range of values.
  • the second value interval is determined according to the position and size of the block to be processed, the position and size of the coding tree unit CTU where the block to be processed is located, and the extension range of the time domain motion vector field of the preset CTU, including:
  • the third threshold and the fourth threshold are determined according to the position and size of the block to be processed, the position and size of the coding tree unit CTU where the block to be processed is located, and the extension range of the time domain motion vector field of the preset CTU, where the third threshold Is the minimum value that the component of the target offset vector can achieve in the second direction, and the fourth threshold is the maximum value that the component of the target offset vector can achieve in the second direction; according to the third threshold and the fourth threshold, the first Two value ranges.
  • the extended range of the time domain motion vector field of the CTU includes the extended range of the time domain motion vector field of the CTU in the first direction and the second direction, respectively.
  • the first direction is a horizontal direction
  • the second direction is a vertical direction
  • the extended range of the time domain motion vector field of the CTU includes the time domain motion vector fields of the CTU being up, down, left, and right, respectively. Extended range in direction.
  • the extended range information of the time domain motion vector field of the CTU is carried in at least one of a sequence parameter set SPS, an image parameter set PPS, and a header field.
  • the extended range of the time domain motion vector field of the CTU is preset.
  • determining a motion vector of a sub-block to be processed according to a motion vector of a corresponding sub-block includes: according to an image sequence number POC of an image frame where the sub-block to be processed is located, The POC of the target reference frame of the image frame where the sub-block is located, the POC of the target image, and the POC of the target reference frame of the target image are processed to scale the motion vector of the corresponding sub-block to obtain the motion vector of the sub-block to be processed.
  • the motion vector of the sub-block is scaled to obtain the motion vector of the sub-block to be processed, including: determining an image sequence number POC of an image frame where the sub-block to be processed and a POC of a target reference frame of the image frame where the sub-block is to be processed
  • the difference is the first difference
  • the difference between the POC of the target image and the POC of the target reference frame of the target image is determined to be the second difference
  • the motion vector of the corresponding subblock is based on the ratio of the first difference and the second difference Scaling is performed to obtain the motion vector of the sub-block to be processed.
  • scaling the motion vector of the corresponding sub-block according to the ratio of the first difference and the second difference to obtain the motion vector of the sub-block to be processed includes: determining a ratio of the first difference and the second difference Is the first ratio; the product of the first ratio and the motion vector of the corresponding sub-block is determined as the motion vector of the sub-block to be processed.
  • the motion vector of the sub-block to be processed is obtained according to the following formula:
  • P1 represents the POC of the image frame where the sub-block to be processed is located
  • P2 represents the POC of the target reference frame of the image frame where the to-be-processed sub-block is located
  • P3 represents the POC of the target image
  • P4 represents the POC of the target reference frame of the target image.
  • MV represents the motion vector of the corresponding sub-block
  • MVs represents the motion vector of the sub-block to be processed.
  • the MV is decomposed into a motion vector MVx in the horizontal direction and a motion vector MVy in the vertical direction, which are respectively calculated according to the above formula to obtain a motion vector MVsx in the horizontal direction and a motion vector MVsy in the vertical direction, respectively.
  • a method for obtaining a motion vector includes: determining a target offset vector and identification information of a target image of a block to be processed, wherein the block to be processed includes at least one subblock to be processed; The position of the block and the target offset vector determine the initial corresponding position point of the sub-block to be processed in the target image; when the initial corresponding position point is within a preset area of the target image, the initial corresponding position point is determined as the target corresponding position Point; when the initial corresponding position point exceeds the preset area, the position closest to the initial corresponding position point in the preset area is determined as the target corresponding position point; the motion vector of the sub-block to be processed is determined according to the motion vector corresponding to the target corresponding position point .
  • the to-be-processed block may be composed of at least one to-be-processed sub-block, and obtaining the motion vector of the to-be-processed block may be decomposed into obtaining the motion vector of each to-be-processed sub-block.
  • the block to be processed may be a CU, and the sub-block to be processed is a sub-CU.
  • the block of the pair of blocks to be processed in the target image may be called a corresponding block, and the corresponding block includes at least one corresponding subblock.
  • the target offset vector is used to determine the position of the corresponding block of the block to be processed in the target image.
  • the target offset vector is different from the normal motion vector.
  • the target offset vector here is a vector used to determine the corresponding block, and the motion vector is generally used to determine the predicted block of an image block to obtain the image. Predicted pixel value of a block
  • the offset vector can also be called a temporal vector, and the target image can also be called a corresponding image.
  • the corresponding position point is relative to the preset area.
  • the corresponding position point also exists in the form of a single point; when the preset area uses the point set as the basic composition In units, the corresponding position points also exist as a point set.
  • the distance between a certain position in the preset area and the corresponding position point is the distance between a single point in the preset area and the corresponding position point, or a set of points in the preset area and the point in the corresponding position point. Set distance.
  • the target corresponding position point located in the preset area is found according to the initial corresponding position point and the preset area of the target image, and then the motion vector of the sub-block to be processed is determined according to a certain range of motion vectors of the target image.
  • the time-domain motion vector field in a certain range of the target image is pre-read to reduce the memory bandwidth and the number of times the memory is read.
  • the above method further includes: obtaining a motion vector of a pixel in a preset region in the target image, where the preset region is the sub-block to be processed The possible range of corresponding subblocks.
  • the memory bandwidth can be reduced.
  • the preset area is based on the position and size of the coding tree unit CTU where the block to be processed is located, the size of the target image, and the time domain motion vector field of the preset CTU.
  • the expansion range is determined.
  • the preset area may be defined by an abscissa value interval and an ordinate value interval.
  • the extended range of the time domain motion vector field of the CTU includes the extended range of the time domain motion vector field of the CTU in the first direction and the second direction, respectively.
  • the first direction is a horizontal direction
  • the second direction is a vertical direction
  • the extended range of the time domain motion vector field of the CTU includes the time domain motion vector fields of the CTU being up, down, left, and right, respectively. Extended range in direction.
  • the extended range information of the time domain motion vector field of the CTU is carried in at least one of a sequence parameter set SPS, an image parameter set PPS, and a header field.
  • the extended range of the time domain motion vector field of the CTU is preset.
  • determining a motion vector of a sub-block to be processed according to a motion vector corresponding to a target corresponding position point includes: according to an image sequence number POC of an image frame where the sub-block to be processed is located The POC of the target reference frame of the image frame where the sub-block to be processed is located, the POC of the target image, and the POC of the target reference frame of the target image, the motion vector corresponding to the target corresponding position point is scaled to obtain the Motion vector.
  • the motion vector corresponding to the target corresponding position point is scaled to obtain the motion vector of the sub-block to be processed, including: determining an image sequence number POC of the image frame where the sub-block to be processed is located and a target reference frame of the image frame where the sub-block is to be processed.
  • the difference between the POC of the target image is the third difference; the difference between the POC of the target image and the POC of the target reference frame of the target image is determined to be the fourth difference; the corresponding sub-groups are compared according to the ratio of the third difference and the fourth difference.
  • the motion vector of the block is scaled to obtain the motion vector of the sub-block to be processed.
  • scaling the motion vector of the corresponding sub-block according to the ratio of the third difference and the fourth difference to obtain the motion vector of the sub-block to be processed includes: determining a ratio of the third difference and the fourth difference Is the second ratio; the product of the second ratio and the motion vector of the corresponding sub-block is determined as the motion vector of the sub-block to be processed.
  • the motion vector of the sub-block to be processed is obtained according to the following formula:
  • P1 represents the POC of the image frame where the sub-block to be processed is located
  • P2 represents the POC of the target reference frame of the image frame where the to-be-processed sub-block is located
  • P3 represents the POC of the target image
  • P4 represents the POC of the target reference frame of the target image.
  • MV represents the motion vector corresponding to the target position and MVs represents the motion vector of the sub-block to be processed.
  • the MV is decomposed into a motion vector MVx in the horizontal direction and a motion vector MVy in the vertical direction, which are respectively calculated according to the above formula to obtain a motion vector MVsx in the horizontal direction and a motion vector MVsy in the vertical direction, respectively.
  • an apparatus for acquiring a motion vector includes a module for executing a method in any one of the foregoing implementation manners of the first aspect or the second aspect.
  • a device for acquiring a motion vector including: a non-volatile memory and a processor coupled to each other, the processor invoking program code stored in the memory to execute the first aspect or the second aspect Some or all of the steps of the method in any of the implementations.
  • a computer-readable storage medium stores program code, where the program code includes an implementation manner for performing any one of the first aspect or the second aspect. Instructions for some or all steps of the method.
  • a computer program product is provided, and when the computer program product runs on a computer, the computer is caused to execute part or all of the steps of the method in any one of the first aspect or the second aspect of the implementation manner. Instructions.
  • FIG. 1 is a schematic diagram of a video encoding process
  • FIG. 2 is a schematic diagram of a video decoding process
  • FIG. 3 is a schematic flowchart of a method for acquiring a motion vector according to an embodiment of the present application
  • FIG. 4 is a schematic diagram of a neighboring block of a block to be processed
  • FIG. 5 is a schematic diagram of a sub-block to be processed included in a current image and a corresponding sub-block in a target image;
  • FIG. 6 is a schematic diagram of a motion vector of a sub-block to be processed and a motion vector of a corresponding sub-block;
  • FIG. 7 is a schematic diagram of an extended range of a time-domain motion vector field of a CTU
  • FIG. 8 is a schematic diagram of an extended range of a time domain motion vector field of a current CTU and a left CTU;
  • FIG. 9 is a schematic flowchart of a method for acquiring a motion vector according to an embodiment of the present application.
  • FIG. 10 is a schematic diagram of determining an initial corresponding position point of a sub-block to be processed
  • FIG. 11 is a schematic diagram of determining a position point corresponding to a target
  • FIG. 12 is a schematic diagram of a motion vector of a sub-block to be processed and a motion vector of a corresponding position point of a target;
  • FIG. 13 is a flowchart of a method for acquiring a motion vector according to an embodiment of the present application.
  • FIG. 14 is a flowchart of a method for acquiring a motion vector according to an embodiment of the present application.
  • 15 is a flowchart of a method for acquiring a motion vector according to an embodiment of the present application.
  • FIG. 16 is a schematic diagram of determining a corresponding sub-block of a sub-block to be processed in the existing scheme
  • 17 is a schematic diagram of determining a corresponding sub-block of a sub-block to be processed according to the present application.
  • FIG. 18 is a process of constructing a candidate information list in an existing fusion mode
  • FIG. 20 is a schematic block diagram of an apparatus for acquiring a motion vector according to an embodiment of the present application.
  • FIG. 21 is a schematic block diagram of an apparatus for acquiring a motion vector according to an embodiment of the present application.
  • 22 is a schematic block diagram of a video encoder according to an embodiment of the present application.
  • FIG. 23 is a schematic block diagram of a video decoder according to an embodiment of the present application.
  • 24 is a schematic block diagram of a video transmission system according to an embodiment of the present application.
  • 25 is a schematic block diagram of a video codec device according to an embodiment of the present application.
  • FIG. 26 is a schematic block diagram of a video codec system according to an embodiment of the present application.
  • FIG. 1 is a schematic diagram of a video encoding process.
  • both intra prediction and inter prediction can be used.
  • intra prediction is used when the current frame Fn is an I frame
  • inter prediction is used when the current frame Fn is a P frame or a B frame.
  • intra prediction is used, the pixel value of the pixel point of the reconstructed area in the current frame Fn can be used to predict the pixel value of the pixel point of the current image block.
  • the reference frame F ' n-1 can be used. The pixel value of the pixel point of the reference block matching the current image block is used to predict the pixel value of the pixel point of the current image block.
  • the difference between the pixel value of the pixel point of the current image block and the pixel value of the pixel point of the prediction block is used to obtain the residual information and Variation, quantization, and entropy coding are performed to obtain a coded code stream.
  • the residual information of the current frame Fn and the prediction information of the current frame Fn are superimposed, and a filtering operation is performed to obtain a reconstructed frame F ' n of the current frame, which is used as a reference frame for subsequent encoding.
  • FIG. 2 is a schematic diagram of a video decoding process.
  • the video decoding process shown in Figure 2 is equivalent to the inverse process of the video decoding process shown in Figure 1.
  • entropy decoding and inverse quantization and inverse transformation are used to obtain residual information, and the current image block usage is determined according to the decoded code stream.
  • Intra prediction or inter prediction If it is intra prediction, use the pixel values of pixels in the reconstructed area in the current frame to construct prediction information according to the intra prediction method; if it is inter prediction, you need to parse out the motion information and use the parsed motion information
  • the reference block is determined in the reconstructed image, and the pixel values of the pixels in the reference block are used as the prediction information. Next, the prediction information and the residual information are superimposed, and the reconstruction information can be obtained after the filtering operation.
  • the method for obtaining a motion vector in this embodiment of the present application may occur in both an encoding process and a decoding process, that is, both the encoding end and the decoding end may execute the method for obtaining a motion vector in the embodiment of the present application. Specifically, the method for obtaining a motion vector in the embodiment of the present application may occur during the inter prediction process shown in FIG. 1 and FIG. 2.
  • HEVC high definition video Coding
  • AMDV advanced motion vector prediction
  • merge mode merge mode
  • FIG. 3 is a schematic flowchart of a method for acquiring a motion vector according to an embodiment of the present application.
  • the method shown in FIG. 3 may be performed by a video codec device, a video codec, a video codec system, and other devices with a video codec function.
  • FIG. 3 is a schematic flowchart of a method for acquiring a motion vector according to an embodiment of the present application.
  • the method shown in FIG. 3 may be performed by a video codec device, a video codec, a video codec system, and other devices with a video codec function.
  • the method shown in FIG. 3 includes steps 110 to 140. Steps 110 to 140 are described in detail below.
  • the first value range is a value interval composed of a first threshold and a second threshold
  • the second value range is a value interval composed of a third threshold and a fourth threshold, where the first The threshold is smaller than the second threshold, and the third threshold is smaller than the fourth threshold.
  • first value interval and the second value interval may be any one of an open interval, a closed interval, and a half-open half-closed interval.
  • the specific expression of the first value interval may be (first threshold, second threshold), (first threshold, second threshold), or [first threshold, second threshold].
  • the specific expression of the second value interval may be (third threshold, fourth threshold), (third threshold, fourth threshold), or [third threshold, fourth threshold].
  • the above-mentioned block to be processed includes at least one sub-block to be processed, the value of the component of the target offset vector in the first direction is within the first value interval, and the value of the component of the target offset vector in the second direction is In the second value interval, the first direction and the second direction have an orthogonal relationship.
  • the (image) block corresponding to the block to be processed in the target image may be referred to as a corresponding block.
  • the corresponding block includes at least one corresponding subblock.
  • the target offset vector is used to determine the corresponding block of the block to be processed in the target image. (s position).
  • the target offset vector is different from a normal motion vector.
  • the target offset vector is a vector used to determine a corresponding block, and the motion vector is generally used to determine a prediction block of an image block, so as to obtain a predicted pixel of the image block.
  • the offset vector can also be called a temporal vector
  • the target image can also be called a corresponding image.
  • the identification information of the target image may specifically be a picture sequence number (POC) of an image frame where the target image is located, or may be index information or an index value of the image frame where the target image is located.
  • POC picture sequence number
  • the foregoing POC is used to indicate the actual position of the image in the original image sequence.
  • the POC may also be referred to as a display order or a playback order.
  • the at least one sub-block to be processed may be obtained by dividing the block to be processed, and the at least one sub-block to be processed may form a block to be processed.
  • the block to be processed is a CU
  • the CU may include at least one sub-CU.
  • the first direction and the second direction may be a horizontal direction and a vertical direction, respectively.
  • the specific form of the component of the target offset vector in the first direction within the first value interval includes the following types:
  • the value of the component of the target offset vector in the first direction is larger than the first threshold and smaller than the second threshold
  • the value of the component of the target offset vector in the first direction is greater than or equal to the first threshold value and less than or equal to the second threshold value.
  • the specific form of the value of the component of the target offset vector in the second direction within the second value interval includes the following:
  • the value of the component of the target offset vector in the second direction is greater than the third threshold and less than or equal to the fourth threshold;
  • the value of the component of the target offset vector in the second direction is greater than or equal to the third threshold and less than or equal to the fourth threshold.
  • various methods may be used to determine the target offset vector. Specifically, it is possible to first determine an initial offset vector (also referred to as a starting offset vector), and then limit the values of the components of the initial offset vector in the first direction and the second direction to the first In the value interval and the second value interval, the final target offset vector is obtained. It is also possible to directly select, from a plurality of motion vectors, motion vectors whose component values in the first direction and the second direction are in the first value interval and the second value interval, respectively, as the target offset vector. The two methods for determining the target offset vector are described in detail below.
  • the first way determine the initial offset vector, and limit the component values of the initial offset vector in the first direction and the second direction to obtain the target offset vector.
  • determining the target offset vector includes steps 1201 to 1203. Steps 1201 to 1203 are described in detail below.
  • a motion vector of a preset spatial domain neighboring block of the block to be processed is determined as the initial offset vector.
  • the motion vector of the first available preset spatial-domain neighboring block obtained may be determined as the initial offset vector in a certain order.
  • a 0 , A 1 , B 0 , B 1, and B 2 are spatial-domain neighboring blocks of a block to be processed
  • T is a temporal-domain neighboring block of the block to be processed
  • C is a block of the block to be processed.
  • the image block in the reference frame is at the same position as the block to be processed.
  • the process of determining the initial offset vector can be simplified.
  • the initial offset vector is Determined as the target offset vector.
  • the initial offset vector may be directly determined as the target offset vector.
  • the values of the components of the clamped initial offset vector obtained in step 1203 in the first direction and the second direction are in the first value interval and the second value interval, respectively.
  • the processing of the initial offset vector includes case 1.
  • case 3 these three cases are described in detail below.
  • the initial offset vector is at the first value interval.
  • the components in the two directions exceeding the second value interval are subjected to clamping processing, and the initial offset vector after the clamping processing is determined as the target offset vector.
  • the first interval is [first threshold, second threshold]
  • the second interval is [third threshold, fourth threshold]
  • the components of the initial offset vector in the first and second directions are first Component and a second component, wherein the first component is smaller than the second threshold and larger than the first threshold, and the second component is larger than the fourth threshold.
  • the second component of the initial offset vector exceeds the value range defined by [the third threshold, the fourth threshold], and the part where the value of the second component exceeds the fourth threshold needs to be clamped (or, it can also be called Is a truncation process), so that the second component after clamping is greater than or equal to the third threshold and less than or equal to the fourth threshold.
  • the initial offset vector in the first direction When the value of the component of the initial offset vector in the first direction is outside the first value interval and the value of the component in the second direction is within the second value interval, the initial offset vector is in the first direction.
  • the part whose component in one direction exceeds the first value interval is subjected to clamping processing, and the initial offset vector after the clamping processing is determined as the target offset vector.
  • the first interval is [first threshold, second threshold]
  • the second interval is [third threshold, fourth threshold]
  • the components of the initial offset vector in the first and second directions are first Component and a second component, wherein the second component is larger than the third threshold and smaller than the fourth threshold, and the first component is smaller than the first threshold.
  • the first component of the initial offset vector exceeds the value range defined by [first threshold, second threshold]
  • the value of the first component needs to be clamped to the value defined by [first threshold, second threshold]
  • the value range is to add the difference between the first threshold and the first component on the basis of the first component, so that the restricted first component is within the first value interval.
  • the initial offset vector is in the first direction.
  • the components in one direction that exceed the first value interval are clamped, and the components in the second direction that the initial offset vector exceeds the second value interval are clamped.
  • the offset vector is determined as the target offset vector.
  • the first interval is [first threshold, second threshold]
  • the second interval is [third threshold, fourth threshold]
  • the components of the initial offset vector in the first and second directions are first And a second component, wherein the first component is larger than a second threshold and the second component is larger than a fourth threshold.
  • the first component of the initial offset vector exceeds the value range defined by [first threshold, second threshold], and the second component of the initial offset vector exceeds the limit defined by [third threshold, fourth threshold]
  • the value range the part where the value of the first component exceeds the second threshold is clamped, and the part where the value of the second component exceeds the fourth threshold is clamped, so that the value of the first component after clamping is A threshold, a second threshold], the value of the second component after clamping is within a range of [a third threshold, a fourth threshold].
  • Second method A motion vector whose component values in a first direction and a second direction of a motion vector in a spatially neighboring block of a block to be processed satisfy a requirement is determined as a target offset vector.
  • determining the target offset vector includes steps 1204 and 1205. Steps 1204 and 1205 are described in detail below.
  • the motion vectors of the spatial-domain neighboring blocks of the block to be processed may be obtained in a certain order, and then the values of the components of the motion vectors of the spatial-domain neighboring blocks in the first direction and the second direction are determined. Whether to be in the first value interval and the second value interval, and determine the value of the component in the first direction and the second direction in the first value interval and the motion vector in the second value interval, respectively Is the target offset vector.
  • the motion vector of the neighboring block available a B 0, A 0 order will find 4, and then determines the motion vector component values satisfies Requirements (the values of the components in the first direction and the second direction are in the first value interval and the second value interval, respectively), and if the motion vector meets the requirements, the motion vector is determined as the target offset vector If the motion vector does not meet the requirements, then continue to find the motion vector of the next available neighboring block in the order of A 1 , B 1 , B 0 , and A 0 , and then determine whether the requirements are met until it is found to meet the requirements. Motion vector.
  • the target offset vector used in step 120 may also be a zero offset vector.
  • the image block in the target image that is at the same position as the block to be processed is the corresponding block of the block to be processed in the target image.
  • the ATMVP technology may not be used, but other technologies may be used to obtain the motion vector of the sub-block to be processed.
  • the corresponding block can be obtained first according to the offset vector, and then determined in the target image to be relative to the block to be processed according to the position of the block to be processed.
  • the corresponding sub-block of the positional relationship (it may also be understood as determining the corresponding sub-block having a relative positional relationship with the sub-block to be processed in the corresponding block).
  • a sub-block in the corresponding block having the same relative position as the sub-block to be processed may be determined as a corresponding sub-block of the sub-block to be processed.
  • the current image includes a block to be processed, and the block to be processed includes 4 sub-blocks.
  • the position of the block to be processed and the offset vector of the block to be processed can be used to obtain the corresponding block in the target image.
  • the corresponding block also contains 4 sub-blocks, and the sub-block to be processed is located at the upper left corner of the block to be processed. Then, when determining the corresponding sub-block of the sub-block to be processed, the sub-block in the upper left corner of the corresponding block Determined as the corresponding sub-block of the sub-block to be processed.
  • the sub-blocks to be processed determined in the target image determined according to the target offset vector are also located in a certain range of the target image, and thus can be By pre-reading the time-domain motion vector field within a certain range of the target image, the memory bandwidth and the number of memory reads are reduced.
  • the method shown in FIG. 3 further includes: obtaining a motion vector of a pixel in a preset area in the target image, where the preset area is a possible occurrence of a corresponding sub-block of the sub-block to be processed. range.
  • the memory bandwidth can be reduced.
  • the motion vector of the sub-block to be processed may be obtained by scaling the motion vector of the corresponding sub-block.
  • determining the motion vector of the sub-block to be processed according to the motion vector of the corresponding sub-block includes: according to the picture sequence number (POC) of the image frame where the sub-block to be processed is located, and the image frame where the sub-block to be processed is located.
  • POC picture sequence number
  • the POC of the target reference frame, the POC of the target image, and the POC of the target reference frame of the target image are used to scale the motion vector of the corresponding sub-block to obtain the motion vector of the sub-block to be processed.
  • the image frame where the sub-block to be processed is located is image frame 1
  • the target reference frame of image frame 1 is image frame 2
  • the image frame where the target image is located is image frame 3
  • the target reference of image frame 3 is The frame is image frame 4 and the motion vector of the corresponding sub-block is MV.
  • the MV can be scaled according to the POC of image frame 1 to image frame 4 to obtain the motion vector of the sub-block to be processed.
  • the scaling of the motion vector to obtain the motion vector of the sub-block to be processed includes: determining the difference between the POC of the image frame where the sub-block to be processed is located and the POC of the target reference frame of the image frame where the sub-block is located is the first The difference; determining the difference between the POC of the target image and the POC of the target reference frame of the target image as the second difference; scaling the motion vector of the corresponding subblock according to the ratio of the first difference and the second difference to Get the motion vector of the sub-block to be processed.
  • the ratio of the first difference to the second difference may be calculated first. Obtain the first ratio, and then determine the product of the first ratio and the motion vector of the corresponding sub-block as the motion vector of the sub-block to be processed.
  • formula (1) can be used to calculate the motion vector of the sub-block to be processed.
  • MV s is the motion vector of the sub-block to be processed
  • MV is the motion vector of the corresponding sub-block
  • P1 to P4 are the POC of image frame 1 to image frame 4, respectively
  • P1-P2 represents the first difference
  • P3-P4 represents Difference.
  • determining the first value interval and the second value interval includes: according to a position and a size of the block to be processed, and a position of a coding tree unit CTU where the block to be processed is located. And the size and the extended range of the time-domain motion vector field of the preset CTU, determine the first value interval and the second value interval.
  • first value interval and the second value interval may be determined by determining the maximum and minimum values that the target offset vector can achieve in the first direction and the second direction. The following uses the method to determine the first value interval and the second value interval in detail.
  • the above determining the first value interval and the second value interval includes the following process:
  • a second value interval is obtained according to the third threshold and the fourth threshold.
  • the first threshold value is the minimum value that the component of the target offset vector can achieve in the first direction
  • the second threshold value is the maximum value that the component of the target offset vector can take in the first direction
  • the third threshold value is the target offset The minimum value that the component of the shift vector in the second direction can take
  • the fourth threshold value is the maximum value that the component of the target offset vector can take in the second direction.
  • the extended range of the time domain motion vector field of the CTU includes the extended range of the time domain motion vector field of the CTU in the first direction and the second direction, respectively.
  • first direction and the second direction are two directions in an orthogonal relationship.
  • the first direction may be a horizontal direction
  • the second direction may be a vertical direction
  • the extended range of the time domain motion vector field of the CTU includes the time domain motion vector fields of the CTU in the up, down, left, and right directions, respectively. On the expansion value.
  • the area enclosed by the dotted line and the rectangle where the CTU is located is the extended range of the CTU, and the extended values of the CTU in the up, down, left, and right directions are U, B, L, and R, respectively.
  • the time domain motion vector field of the current CTU is limited to the time domain motion vector of the CTU at the corresponding position point. field.
  • the time domain motion vector field obtained by the current CTU cannot cross the upper and lower boundaries of the CTU, but it can cross the left and right boundaries of the CTU.
  • the MVF of the current CTU and the left CTU can be expanded left and right, but cannot be expanded upward and downward.
  • the time domain motion vector field of the current CTU and the time domain motion vector field of the left CTU partially overlap. In this case, when processing the current CTU, the time-domain motion vector field that the CTU on the left has been read can be partially reused by sliding the window.
  • the first value interval and the second value interval may be determined according to formula (2) to formula (5).
  • HorMax CTUX + CTUW–CUX–CUW + R (3)
  • VerMin CTUY–CUY–U. The following is the case with the following: (4)
  • VerMax CTUY + CTUH–CUY–CUH + B. (5)
  • CUX and CUY represent the abscissa and ordinate of the upper-left vertex pixel of the block to be processed
  • CUW and CUH represent the width and height of the block to be processed
  • CTUX and CTUY represent the abscissa and ordinate of the CTU where the block to be processed is, respectively.
  • CTUW and CTUH are the width and height of the CTU where the block to be processed is respectively
  • U, B, L, and R represent the expansion values of the CTU in the up, down, left, and right directions
  • HorMin and HorMax represent the first threshold and The second thresholds, VerMin and VerMax represent the third and fourth thresholds, respectively.
  • the extended range information of the time domain motion vector field of the CTU is carried in at least one of a sequence parameter set SPS, an image parameter set PPS, and a header field.
  • SPS, PPS, and header fields may indicate the extended range of the CTU by carrying parameters (U, B, L, R).
  • the extended range of the time domain motion vector field of the CTU is preset.
  • the extended range of the motion vector of the CTU can be directly written into the protocol without carrying it through the SPS, PPS, or header field.
  • FIG. 9 is a schematic flowchart of a method for acquiring a motion vector according to an embodiment of the present application.
  • the method shown in FIG. 9 can be executed by a video codec device, a video codec, a video codec system, and other devices with a video codec function.
  • the method shown in FIG. 9 includes steps 210 to 250. Steps 210 to 250 are described in detail below.
  • the to-be-processed block includes at least one to-be-processed sub-block, and the to-be-processed block is composed of at least one to-be-processed sub-block.
  • the block to be processed is a CU
  • the sub-block to be processed is a sub-CU.
  • the target offset vector in step 210 is consistent with the meaning of the target offset vector mentioned in the method shown in FIG. 3, and will not be discussed in detail here.
  • the identification information of the target image may specifically be a POC of an image frame where the target image is located, or may be index information or an index value of the image frame where the target image is located.
  • the corresponding sub-block of the sub-block to be processed may be determined first, and then a certain set position of the corresponding sub-block is determined as the position of the sub-block to be processed.
  • the initial corresponding position point in the target image For example, after the corresponding position point is determined, the position of the center point or the upper left corner of the corresponding sub-block may be determined as the initial corresponding position point of the sub-block to be processed in the target image.
  • the corresponding sub-block may also be undefined, but the position of the sub-block to be processed and the target offset vector may be directly determined to directly correspond to the sub-block to be processed.
  • the position in the target image is determined as the initial corresponding position point of the sub-block to be processed in the target image. For example, first determine a certain set position of the sub-block to be processed, such as the center point or the upper left corner, and add the position coordinate to the target offset vector to determine the initial corresponding position of the sub-block to be processed in the target image.
  • the current image includes a block to be processed
  • the block to be processed includes a subblock to be processed at the upper left corner position
  • the target image includes a corresponding block
  • the corresponding block includes the corresponding subblock at the upper left corner position
  • the target offset vector of the block to be processed points to the corresponding block
  • the rectangular area within the dashed frame of the target image is the preset area of the target image.
  • the initial corresponding position point of the block to be processed in the target image is located in the target image.
  • the preset area can directly determine the initial corresponding position point as the target corresponding position point.
  • the current image includes a block to be processed
  • the block to be processed includes a subblock to be processed at the upper left corner
  • the target image includes a corresponding block
  • the corresponding block includes the corresponding subblock at the upper left corner.
  • the target offset vector of the block to be processed points to the corresponding block
  • the rectangular area within the dashed frame of the target image is the preset area of the target image.
  • the initial corresponding position point of the subblock to be processed in the target image is located in the target image. Outside the preset area, in this case, the position A closest to the initial corresponding position point in the preset area needs to be determined as the target corresponding position point.
  • the target corresponding position point located in the preset area is found according to the initial corresponding position point and the preset area of the target image, and then the motion vector of the sub-block to be processed is determined according to a certain range of motion vectors of the target image.
  • the time-domain motion vector field in a certain range of the target image is pre-read to reduce the memory bandwidth and the number of times the memory is read.
  • the method shown in FIG. 9 further includes: acquiring a motion vector of a pixel in a preset area in the target image.
  • the motion vectors of the pixels in the preset area of the target image can be obtained in advance. In this way, when the motion vector of the corresponding position point of the target needs to be used, the preset area can be directly read from the The motion vector of the inner pixel points obtains the motion vector corresponding to the target position point, and because only the motion vectors of the pixel points in the preset area of the target image are acquired, unlike the traditional solution, which may require the entire target image to be acquired, the occupation can be reduced Memory bandwidth.
  • the preset area in the above step 230 is based on the position and size of the coding tree unit CTU where the block to be processed is located, the size of the target image, and the extended range of the time domain motion vector field of the preset CTU. definite.
  • the extended range of the time domain motion vector field of the CTU includes the extended range of the time domain motion vector field of the CTU in the first direction and the second direction, respectively.
  • the first direction is a horizontal direction
  • the second direction is a vertical direction
  • the extended range of the time domain motion vector field of the CTU includes the time domain motion vector fields of the CTU being up, down, left, and right, respectively. Extended range in direction.
  • the above-mentioned preset area may be defined in various ways.
  • the above-mentioned preset area may be represented by an abscissa value interval and an ordinate value interval.
  • the preset area may be represented by an abscissa value interval [a minimum coordinate value, an abscissa maximum value] and an ordinate value interval [a minimum coordinate value, a ordinate maximum value].
  • HorMin Min (CTUX-L, 0): (Min. CTUX-L, 0) (6)
  • HorMax Max (CTUX + CTUW + R–1, PicW–1) (7)
  • VerMin Min (CTUY-U, 0): (Min.) (8)
  • VerMax Max (CTUY + CTUH + B–1, PicH–1) (9)
  • CTUX is the abscissa of the CTU where the block to be processed is located
  • CTUY is the ordinate of the CTU where the block is to be processed
  • CTUW and CTUH are the width and height of the CTU where the block to be processed is located
  • PicW and PicH are the target image Width and height
  • U, B, L, R represent the expansion values of the CTU in the up, down, left, and right directions
  • HorMin and HorMax are the abscissa minimum and abscissa maximum, respectively
  • VerMin and VerMax are ordinate respectively. Minimum and ordinate maximum.
  • the abscissa value interval and the ordinate value interval can be obtained, and the preset area of the target image is determined.
  • the extended range information of the time domain motion vector field of the CTU is carried in at least one of an SPS, a PPS, and a header field.
  • the extended range of the time domain motion vector field of the CTU may be preset. In this case, the extended range of the motion vector of the CTU can be directly written into the protocol without carrying it through the SPS, PPS, or header field.
  • the foregoing determining the motion vector of the sub-block to be processed according to the motion vector corresponding to the target corresponding position point includes: according to the POC of the image frame where the sub-block to be processed is located and the image frame where the sub-block to be processed is located.
  • the POC of the target reference frame, the POC of the target image, and the POC of the target reference frame of the target image are used to scale the motion vector corresponding to the target corresponding position point to obtain the motion vector of the sub-block to be processed.
  • the target corresponding position is
  • the difference is a third difference; the difference between the POC of the target image and the POC of the target reference frame of the target image is determined to be the fourth difference; and then the corresponding sub-block is compared according to the ratio of the third difference and the fourth difference. Scaling the motion vector to obtain the motion vector of the sub-block to be processed.
  • scaling the motion vector of the corresponding sub-block according to the ratio of the third difference and the fourth difference to obtain the motion vector of the sub-block to be processed includes: determining a ratio of the third difference and the fourth difference Is the second ratio; the product of the second ratio and the motion vector corresponding to the target corresponding position point is determined as the motion vector of the sub-block to be processed.
  • the image frame where the sub-block to be processed is located is image frame 1
  • the target reference frame of image frame 1 is image frame 2
  • the image frame where the target image is located is image frame 3
  • the target reference of image frame 3 is The frame is image frame 4 and the motion vector of the corresponding position point of the target is MV.
  • the MV can be scaled according to the POC of image frame 1 to image frame 4 to obtain the motion vector of the sub-block to be processed.
  • formula (10) can be used to calculate the motion vector of the sub-block to be processed.
  • MV s is the motion vector of the sub-block to be processed
  • MV is the motion vector of the corresponding position point of the target
  • P1 to P4 are the POC of image frame 1 to image frame 4, respectively
  • P1-P2 represents the third difference
  • P3-P4 Represents the fourth difference.
  • the motion vectors of all the sub-blocks to be processed in the block to be processed can be used to predict the block to be processed to obtain the pixel prediction value of the block to be processed. After obtaining the motion vector of each to-be-processed block, it is also possible to predict the to-be-processed sub-block and obtain the pixel prediction value of each to-be-processed sub-block. After predicting each to-be-processed sub-block, it is also obtained Pixel prediction of the block to be processed.
  • the first embodiment and the second embodiment correspond to the method shown in FIG. 3, and are implemented by limiting the value of the target offset vector so that the corresponding subblock finally found is located in a certain region of the target image, and is implemented Example 3 does not place any restrictions on the value of the target offset vector, but directly restricts the target corresponding position points of the sub-blocks to be processed in the target image to a certain area.
  • the motion vectors of the sub-blocks to be processed are determined according to the motion vectors of the pixels in a certain area of the target image. The first to third embodiments will be described in detail below.
  • the specific process of obtaining a motion vector in Embodiment 1 includes:
  • the constraint range here is equivalent to the first value interval and the second value interval above.
  • constraint range is determined to constrain the component values of the target offset vector of the block to be processed in the first direction and the second direction, so that the component values of the target offset vector in the first direction and the second direction are between Within the constraints.
  • the foregoing constraint range may be determined according to the extended range of the time domain motion vector field of the CTU, the position and size of the block to be processed, and the position and size of the CTU where the block to be processed is located.
  • the extended range of the time domain motion vector field of the CTU can be defined by the parameters RangeL, RangeR, RangeU, and RangeB (equivalent to the parameters L, R, U, and B above).
  • the parameters RangeL, RangeR, RangeU, and RangeB are respectively expressed.
  • the time-domain motion vector field of the CTU can be extended in four directions: left, right, up, and down.
  • RangeL, RangeR, RangeU, RangeB can be 0, 16, 32, and 64 (the unit is pixel).
  • RangeL, RangeR, RangeU, RangeB can be directly determined through a predefined method, or can be carried in a high-level syntax, for example, RangeL, RangeR, RangeU, RangeB can be carried in the SPS, PPS, and header fields. in.
  • the foregoing constraint range may be determined according to formula (11) to formula (14).
  • HorMin CTUX–CUX–RangeL The following are the examples: (11)
  • HorMax CTUX + CTUW–CUX–CUW + RangeR (12)
  • VerMin CTUY–CUY--RangeU, and the following: (13)
  • VerMax CTUY + CTUH–CUY–CUH + RangeB (14)
  • CUX and CUY represent the position coordinates of the upper-left vertex pixel of the block to be processed
  • CUW and CUH represent the width and height of the block to be processed
  • CTUX and CTUY represent the position coordinates of the upper-left pixel of the CTU where the block to be processed is located
  • CTUW and CTUH represent CTU Width and height.
  • HorMin and HorMax represent the minimum and maximum values that the constraint range can take in the horizontal direction
  • VerMin and VerMax represent the minimum and maximum values that the constraint range can take in the vertical direction.
  • the starting offset vector here is equivalent to the initial offset vector described above.
  • steps 310 and 320 may be performed simultaneously or separately.
  • the starting offset vector is used to determine the position of the target image block in the target image of the block to be processed.
  • the following two methods can be used to determine the starting offset vector.
  • the third way determine the motion vector available in the spatially adjacent block of the block to be processed as the starting offset vector.
  • the motion vector of the first available neighboring block can be found in the order of A 1 , B 1 , B 0 , and A 0 in FIG. 4 and used as the block to be processed.
  • the starting offset vector is the motion vector of the first available neighboring block.
  • the fourth method first determine the target image, and then determine the motion vector pointing to the target image in the neighboring block as the starting offset vector.
  • the motion vector of the first available adjacent block can be found in the order of A 1 , B 1 , B 0 , and A 0 in FIG. 4. If the motion vector points to the target image , Then use it as the starting offset vector of the block to be processed. Otherwise, the motion vector is scaled so that it points to the target image, and the scaled motion vector is used as the starting offset vector.
  • step 340 and step 350 are performed, and when the starting offset vector does not exceed the constraint range, step 360 is performed.
  • the starting offset vector exceeds the limited range, it needs to be limited to obtain a limited offset vector so that the limited offset vector is within the constraint range.
  • formulas (15) and (16) can be used to limit the components (x off , y off ) of the starting offset vector to obtain a limited offset vector.
  • the Clip3 function is a clamping function, which is specifically defined as:
  • a zero motion vector may also be directly used as the target offset vector.
  • other technologies may be used to obtain the motion vector of the sub-block to be processed.
  • the restricted offset vector can be determined as the target offset vector.
  • the initial offset vector When the initial offset vector is within the constraint range, there is no need to perform any constraint or restriction operation on the initial offset vector, and the initial offset vector can be directly determined as the target offset vector.
  • the motion vector is scaled to obtain the motion vector of the sub-block to be processed.
  • the predicted pixel value of the sub-block to be processed may be determined according to the motion vector of the sub-block to be processed.
  • the predicted sub-block of the sub-block to be processed may be determined according to the motion vector of the sub-block to be processed, and then the pixel value of the predicted sub-block may be determined as the predicted pixel value of the sub-block to be processed. . In this way, after obtaining the predicted pixel value of each sub-block to be processed, the predicted pixel value of the entire to-be-processed block can be obtained.
  • the specific process of obtaining a motion vector in Embodiment 2 includes:
  • the motion vectors of the first available neighboring blocks can be obtained in the order of A 1 , B 1 , B 0 , and A 0 in FIG. 4.
  • the target image may also be determined first, and then a motion vector of an adjacent block of the block to be processed is obtained in the order of A 1 , B 1 , B 0 , and A 0 in FIG. 4.
  • Step 430 is performed; when the motion vector does not point to the target image, the motion vector is scaled so that the scaled motion vector is directed to the target image, and then the scaled motion vector is used as a motion vector of an adjacent block, and execution is continued.
  • Step 440 is performed
  • step 440 is performed, otherwise step 420 is continued until the component values of the motion vectors of the neighboring blocks are within the constraint range.
  • the motion vector is scaled to obtain the motion vector of the sub-block to be processed.
  • the predicted pixel value of the sub-block to be processed may be determined according to the motion vector of the sub-block to be processed.
  • the specific process of obtaining a motion vector in Embodiment 3 includes:
  • step 510 The process of determining the target offset vector and the identification information of the target image in step 510 is the same as the above step 210.
  • the position coordinates of the initial corresponding position points of the sub-block to be processed in the target image can be determined according to formula (18).
  • (x, y) represents the coordinates of the upper-left vertex of the block to be processed
  • i represents the ith sub-block in the left-to-right direction of the block to be processed
  • j represents the j-th sub-block in the block to be processed from top to bottom
  • M and N are the sizes of the sub-blocks to be processed (where M can represent the width of the sub-blocks to be processed, N (Can represent the height of the sub-block to be processed)
  • (x (i, j) , y (i, j) ) represents the position coordinates of the (i, j) -th corresponding sub-block.
  • step 550 is directly performed.
  • step 540 is performed first and then step 550 is performed.
  • the preset area may be a fixed area in the target image, and the preset area of the corresponding image may be determined according to (19) to (22).
  • HorMin Min (CTUX--RangeL, 0) (19)
  • HorMax Max (CTUX + CTUW + RangeR–1, PicW–1) (20)
  • VerMin Min (CTUY-RangeU, 0) (21)
  • VerMax Max (CTUY + CTUH + RangeB–1, PicH–1) (22)
  • HorMin and HorMax represent the minimum and maximum values that the constraint range can take in the horizontal direction.
  • VerMin and VerMax represent the minimum and maximum values that the constraint range can take in the vertical direction.
  • CTUX and CTUY represent the position coordinates of the top left pixel of the CTU where the block to be processed is located.
  • CTUW and CTUH represent the width and height of the CTU.
  • RangeU, RangeB, RangeL and RangeR represent the extended values of the time domain motion vector field of the CTU in the up, down, left, and right directions, respectively.
  • PicW and PicH are the width and height of the target image, respectively.
  • the initial corresponding position point When the initial corresponding position point is in a preset area of the target image, the initial corresponding position point may be directly determined as the target corresponding position point.
  • a position closest to the initial corresponding position point in the preset area may be determined as the target corresponding position point.
  • the target corresponding position point may be determined according to formula (23) and formula (24).
  • x (i, j) Clip3 (HorMin, HorMax, x (i, j) ) (23)
  • the Clip3 function is a clamping function, and the specific definition is shown in the formula (17) above.
  • the target corresponding position point may be To obtain the motion vector of the sub-block to be processed.
  • the predicted pixel value of the sub-block to be processed may be determined according to the motion vector of the sub-block to be processed.
  • the current image includes the current CTU, and the current CTU includes multiple pending blocks.
  • Each pending block includes multiple pending sub-blocks.
  • Each pending block corresponds to an offset vector.
  • the offset vector can find the corresponding block of the block to be processed, and each sub-block to be processed will also correspond to a corresponding sub-block in the target image. If the region of the sub-block corresponding to the sub-block in the target image to be processed is uncertain, or the position of the sub-block to be processed in the target image is uncertain, then the motion vector of the corresponding sub-block may need to be obtained in advance The time domain motion vector field of the entire corresponding image, thereby increasing the memory bandwidth.
  • the current image includes a current CTU
  • the current CTU includes a plurality of blocks to be processed, each of which includes a plurality of sub-blocks to be processed, and a region within a dotted line in the target image is a preset region of the target image.
  • the present application constrains the range of the offset vector of the block to be processed, so that the corresponding subblock corresponding to the target image in the target image is located in a preset area of the corresponding image, or the corresponding position point of the target image in the target image to be processed Located in a preset area of the corresponding image. In this way, it is only necessary to obtain the motion vector of the target image, instead of acquiring the motion vector of the entire target image. This can reduce the memory bandwidth.
  • the pre-fetching method can also reduce the number of times the memory is read.
  • the method for obtaining a motion vector in the embodiment of the present application may be applied to a fusion mode of inter prediction.
  • Sub coding unit motion vector prediction (Sub-CU based motion vector prediction, SMVP) technology divides the current coding unit into sub coding units of size MxN, and derives the motion information of each sub coding unit, and then uses the information of each sub coding unit. Perform motion compensation on the motion information to obtain the predicted value of the current coding unit.
  • SMVP motion vector prediction
  • ATMVP advanced temporal motion vector prediction
  • ATMVP spatial temporal motion vector prediction
  • STMVP spatial-temporal motion vector prediction
  • the ATMVP technology first determines the corresponding location point reference frame, and then divides the current coding unit into MxN sub-coding units, and obtains the current sub-coding unit corresponding to each sub-coding unit in the corresponding position-point reference frame.
  • the motion information of the pixel at the center point position is scaled and converted to the current motion information of each sub-coding unit.
  • the method for obtaining a motion vector in this embodiment of the present application can be understood as an improved ATMVP technology.
  • the process of constructing the candidate information list shown in FIG. 18 includes:
  • the process of constructing the candidate information list shown in FIG. 19 includes:
  • step 720 of inserting ATMVP motion information and STMVP motion information is added, and other steps are unchanged.
  • the motion vector of the sub-block to be processed obtained by the method for acquiring a motion vector in the embodiment of the present application is a part of the ATMVP motion information inserted in the above step 720.
  • the encoding of the fusion mode specifically includes the following processes:
  • a merge index is written into the code stream and passed to the decoding end.
  • the process of constructing the merge mode candidate motion information list is shown in FIG. 19.
  • the method for obtaining a motion vector in the embodiment of the present application has been described in detail above with reference to FIGS. 3 to 19. It should be understood that the method for obtaining a motion vector in the embodiment of the present application may correspond to the inter-frames shown in FIG. 1 and FIG. 2. It is predicted that the method for obtaining a motion vector in the embodiment of the present application may occur during the inter prediction process shown in FIG. 1 and FIG. 2.
  • the method for obtaining a motion vector in the embodiment of the present application may be specifically performed by an inter prediction module in an encoder or a decoder.
  • the method for obtaining a motion vector in the embodiments of the present application may be implemented in any electronic device or device that may need to encode and / or decode a video image.
  • the apparatus for acquiring a motion vector in the embodiment of the present application is described in detail below with reference to FIGS. 20 and 21.
  • the device shown in 20 corresponds to the method shown in FIG. 3, FIG. 13, and FIG. 14, and can perform each step in the method shown in FIG. 3, FIG. 13, and FIG. 14;
  • the device shown in FIG. 21 and FIG. 9 corresponds to the method shown in FIG. 15, and each step in the method shown in FIGS. 9 and 15 can be performed.
  • duplicate descriptions are appropriately omitted below.
  • FIG. 20 is a schematic block diagram of an apparatus for acquiring a motion vector according to an embodiment of the present application.
  • the apparatus 800 shown in FIG. 20 includes:
  • a determining module 810 the determining module 810 is configured to:
  • the block to be processed includes at least one subblock to be processed, and the value of the component of the target offset vector in the first direction is in the In the first value interval, the value of the component of the target offset vector in the second direction is within the second value interval, and the first direction and the second direction have an orthogonal relationship;
  • the obtaining module 820 is configured to obtain a motion vector of the sub-block to be processed according to a motion vector of the corresponding sub-block.
  • the sub-blocks to be processed determined in the target image determined according to the target offset vector are also located in a certain range of the target image, and thus can be By pre-reading the time-domain motion vector field within a certain range of the target image, the memory bandwidth and the number of memory reads are reduced.
  • the above determining module 810 and obtaining module 820 may correspond to a prediction module (specifically, an inter prediction module) in a codec device.
  • the determining module 810 and obtaining module 820 may be implemented by software or hardware in specific implementation. This can be achieved by a combination of software and hardware.
  • FIG. 21 is a schematic block diagram of an apparatus for acquiring a motion vector according to an embodiment of the present application.
  • the apparatus 900 shown in FIG. 21 includes:
  • a determining module 910 configured to:
  • the obtaining module 920 is configured to obtain a motion vector of the sub-block to be processed according to a motion vector corresponding to the corresponding position point of the target.
  • the target corresponding position point located in the preset area is found according to the initial corresponding position point and the preset area of the target image, and then the motion vector of the sub-block to be processed is determined according to a certain range of motion vectors of the target image.
  • the time-domain motion vector field in a certain range of the target image is pre-read to reduce the memory bandwidth and the number of times the memory is read.
  • the above determination module 910 and acquisition module 920 may correspond to the prediction module in the codec device, and may specifically be an inter prediction module.
  • the determination module 910 and the acquisition module 920 may be implemented by software or hardware, or may be implemented by A combination of software and hardware.
  • the method for obtaining a motion vector in the embodiment of the present application may also be performed by a video encoder or a video decoder.
  • the structures of the video encoder and the video decoder in the embodiments of the present application are described below with reference to FIG. 22 and FIG. 23.
  • FIG. 22 is a schematic block diagram of a video encoder according to an embodiment of the present application.
  • the video encoder 1000 shown in FIG. 22 includes an encoding end prediction module 1001, a transform quantization module 1002, an entropy encoding module 1003, an encoding reconstruction module 1004, and an encoding end filtering module.
  • the video encoder 1000 shown in FIG. 22 may encode a video. Specifically, the video encoder 1000 may perform the video encoding process shown in FIG. 1 to implement video encoding. In addition, the video encoder 1000 may also execute the method for acquiring a motion vector in the embodiment of the present application, and the video encoder 1000 may execute each step of the methods shown in FIG. 3, FIG. 9, and FIGS. 13 to 15.
  • the apparatus for obtaining a motion vector in the embodiment of the present application may also be an encoding-side prediction module 1001 in the video encoder 1000.
  • the apparatus 6000 and the apparatus 7000 shown in FIG. 20 and FIG. 21 are equivalent to those in the video encoder 1000.
  • FIG. 23 is a schematic block diagram of a video decoder according to an embodiment of the present application.
  • the video decoder 2000 shown in FIG. 23 includes an entropy decoding module 2001, an inverse transform and inverse quantization module 2002, a decoding end prediction module 2003, a decoding reconstruction module 2004, and a decoding end filtering module 2005.
  • the video decoder 2000 shown in FIG. 23 can decode the video. Specifically, the video decoder 2000 can perform the video decoding process shown in FIG. 2 to decode the video. In addition, the video decoder 2000 may also execute the method for acquiring a motion vector in the embodiment of the present application, and the video decoder 2000 may perform each step of the method for acquiring a motion vector shown in FIG. 3, FIG. 9, and FIGS.
  • the apparatus for acquiring a motion vector in the embodiment of the present application may also be a decoding-end prediction module 2003 in the video decoder 2000.
  • the apparatus 6000 and the apparatus 7000 shown in FIG. 20 and FIG. 21 are equivalent to those in the video decoder 2000.
  • Decoding end prediction module 2003 Decoding end prediction module 2003.
  • the application scenarios of the method for acquiring a motion vector according to the embodiments of the present application are described below with reference to FIGS. 24 to 26.
  • the method for acquiring a motion vector according to the embodiments of the present application may be implemented by the video transmission system and the codec shown in FIG. 24 to FIG. 26. And codec system to perform.
  • FIG. 24 is a schematic block diagram of a video transmission system according to an embodiment of the present application.
  • the video transmission system includes an acquisition module 3001, an encoding module 3002, a sending module 3003, a network transmission 3004, a receiving module 3005, a decoding module 3006, a rendering module 3007, and a display module 208.
  • each module in the video transmission system is as follows:
  • the acquisition module 3001 includes a camera or a camera group, which is used to collect video images and perform pre-encoding processing on the collected video images to convert the optical signal into a digital video sequence;
  • the encoding module 3002 is configured to encode a video sequence to obtain a code stream
  • the sending module 3003 is configured to send the coded code stream out
  • the receiving module 3005 is configured to receive a code stream sent by the sending module 3003.
  • the network 3004 is configured to transmit the code stream sent by the sending module 3003 to the receiving module 3005;
  • the decoding module 3006 is configured to decode a code stream received by the receiving module 3005 and reconstruct a video sequence
  • the rendering module 3007 is configured to render the reconstructed video sequence decoded by the decoding module 3006 to improve the display effect of the video.
  • the video transmission system shown in FIG. 24 can execute the method for obtaining a motion vector in the embodiment of the present application. Specifically, both the encoding module 3001 and the decoding module 3006 in the video transmission system shown in FIG. 24 can perform the acquisition in the embodiment of the present application. Method of motion vector.
  • the codec device and the codec system composed of the codec device will be described in detail below with reference to FIG. 25 and FIG. 26. It should be understood that the codec device and the codec system shown in FIG. 25 and FIG. 26 can execute the method for acquiring a motion vector in the embodiment of the present application.
  • FIG. 25 is a schematic diagram of a video codec device according to an embodiment of the present application.
  • the video codec device 50 may be a device specifically used for encoding and / or decoding a video image, or may be an electronic device with a video codec function. Further, the codec device 50 may be a mobile communication system Terminal or user equipment.
  • the codec device 50 may include the following modules or units: a controller 56, a codec 54, a radio interface 52, an antenna 44, a smart card 46, a card reader 48, a keypad 34, a memory 58, an infrared port 42, and a display 32.
  • the codec device 50 may also include a microphone or any suitable audio input module.
  • the audio input module may be a digital or analog signal input.
  • the codec device 50 may also include an audio output. Module, the audio output module can be headphones, speakers, or analog audio or digital audio output connections.
  • the codec device 50 may also include a battery, which may be a solar cell, a fuel cell, or the like.
  • the codec device 50 may further include an infrared port for short-range sight communication with other devices.
  • the codec device 50 may also use any suitable short-range communication method to communicate with other devices, such as Bluetooth wireless connection, USB / Firewire wired connection.
  • the memory 58 may store data in the form of images and audio data, and may also store instructions for execution on the controller 56.
  • the codec 54 may implement the encoding and decoding of audio and / or video data or the auxiliary encoding and decoding of audio and / or video data under the control of the controller 56.
  • the smart card 46 and the card reader 48 can provide user information, as well as network authentication and authentication information for authorized users.
  • Specific implementation forms of the smart card 46 and the card reader 48 may be an integrated circuit card (UICC) and a UICC reader.
  • UICC integrated circuit card
  • the radio interface circuit 52 may generate a wireless communication signal, and the wireless communication signal may be a communication signal generated during a cellular communication network, a wireless communication system, or a wireless local area network communication.
  • the antenna 44 is used to transmit the radio frequency signal generated by the radio interface circuit 52 to other devices (the number of devices may be one or more), and may also be used to transmit other signals (the number of devices may be one or more) Each) receive radio frequency signals.
  • the codec device 50 may receive the video image data to be processed from another device before transmitting and / or storing. In other embodiments of the present application, the codec device 50 may receive images through a wireless or wired connection and encode / decode the received images.
  • FIG. 26 is a schematic block diagram of a video codec system 7000 according to an embodiment of the present application.
  • the video codec system 7000 includes a source device 4000 and a destination device 5000.
  • the source device 4000 generates encoded video data.
  • the source device 4000 can also be referred to as a video encoding device or a video encoding device.
  • the destination device 5000 can decode the encoded video data generated by the source device 4000.
  • the destination device 5000 can also be called a video decoding device or a video decoding device.
  • the specific implementation form of the source device 4000 and the destination device 5000 may be any one of the following devices: desktop computer, mobile computing device, notebook (eg, laptop) computer, tablet computer, set-top box, smart phone, handheld, TV, camera, display, digital media player, video game console, car computer, or other similar devices.
  • the destination device 5000 can receive the encoded video data from the source device 4000 via the channel 6000.
  • Channel 6000 may include one or more media and / or devices capable of moving the encoded video data from the source device 4000 to the destination device 5000.
  • the channel 6000 may include one or more communication media enabling the source device 4000 to directly transmit the encoded video data to the destination device 5000 in real time.
  • the source device 4000 may be based on a communication standard ( For example, a wireless communication protocol) is used to modulate the encoded video data, and the modulated video data may be transmitted to the destination device 5000.
  • the one or more communication media may include wireless and / or wired communication media, such as a radio frequency (RF) frequency spectrum or one or more physical transmission lines.
  • RF radio frequency
  • the one or more communication media described above may form part of a packet-based network (eg, a local area network, a wide area network, or a global network (eg, the Internet)).
  • the one or more communication media may include a router, a switch, a base station, or other devices that implement communication from the source device 4000 to the destination device 5000.
  • channel 6000 may include a storage medium that stores encoded video data generated by source device 4000.
  • the destination device 5000 can access the storage medium via disk access or card access.
  • Storage media can include a variety of locally accessible data storage media, such as Blu-ray discs, high-density digital video discs (DVDs), compact discs (read-only memories, CD-ROMs), flash memory , Or other suitable digital storage media for storing encoded video data.
  • the channel 6000 may include a file server or another intermediate storage device that stores the encoded video data generated by the source device 4000.
  • the destination device 5000 may access the encoded video data stored at a file server or other intermediate storage device via streaming or downloading.
  • the file server may be a server type capable of storing encoded video data and transmitting the encoded video data to the destination device 5000.
  • the file server may include a global wide area network (Web) server (for example, for a website), a file transfer protocol (FTP) server, a network attached storage (NAS) device, and a local disk. driver.
  • Web global wide area network
  • FTP file transfer protocol
  • NAS network attached storage
  • the destination device 5000 can access the encoded video data via a standard data connection (eg, an Internet connection).
  • data connection types include wireless channels, wired connections (e.g., cable modems, etc.), or a combination of both, suitable for accessing encoded video data stored on a file server.
  • the transmission of the encoded video data from the file server can be streaming, downloading, or a combination of the two.
  • the method for obtaining a motion vector of the present application is not limited to a wireless application scenario.
  • the method for obtaining a motion vector of the present application can be applied to video codecs that support a variety of multimedia applications such as: aerial television broadcasting, cable television transmission, Satellite TV transmission, streaming video transmission (eg, via the Internet), encoding of video data stored on a data storage medium, decoding of video data stored on a data storage medium, or other applications.
  • video codec system 7000 may be configured to support one-way or two-way video transmission to support applications such as video streaming, video playback, video broadcasting, and / or video telephony.
  • the source device 4000 includes a video source 4001, a video encoder 4002, and an output interface 4003.
  • the output interface 4003 may include a modulator / demodulator (modem) and / or a transmitter.
  • Video source 4001 may include a video capture device (e.g., a video camera), a video archive containing previously captured video data, a video input interface to receive video data from a video content provider, and / or a computer to generate video data Graphics system, or a combination of the above video data sources.
  • Video encoder 4002 may encode video data from video source 4001.
  • the source device 4000 directly transmits the encoded video data to the destination device 5000 via the output interface 4003.
  • the encoded video data may also be stored on a storage medium or file server for later access by the destination device 5000 for decoding and / or playback.
  • the destination device 5000 includes an input interface 5003, a video decoder 5002, and a display device 5001.
  • the input interface 5003 includes a receiver and / or a modem.
  • the input interface 5003 can receive the encoded video data through the channel 6000.
  • the display device 5001 may be integrated with the destination device 5000 or may be external to the destination device 5000. Generally, the display device 5001 displays decoded video data.
  • the display device 5001 may include various display devices, such as a liquid crystal display, a plasma display, an organic light emitting diode display, or other types of display devices.
  • the video encoder 4002 and the video decoder 5002 can operate according to a video compression standard (for example, a high-efficiency video codec H.265 standard), and can conform to a high-efficiency video coding (HEVC) test model (HM ).
  • a video compression standard for example, a high-efficiency video codec H.265 standard
  • HEVC high-efficiency video coding
  • HM high-efficiency video coding test model
  • the disclosed systems, devices, and methods may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objective of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each of the units may exist separately physically, or two or more units may be integrated into one unit.
  • the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially a part that contributes to the existing technology or a part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.
  • the aforementioned storage media include: U disks, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks, and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)
  • Physics & Mathematics (AREA)
  • Discrete Mathematics (AREA)
  • General Physics & Mathematics (AREA)

Abstract

本申请提供了一种获取运动矢量的方法和装置。该方法包括:确定第一取值区间和第二取值区间;确定待处理块的目标偏移矢量和目标图像的标识信息,其中,待处理块包含至少一个待处理子块,目标偏移矢量在第一方向上的分量的取值在第一取值区间内,目标偏移矢量在第二方向上的分量的取值在第二取值区间内,第一方向和第二方向呈正交关系;根据待处理子块的位置和目标偏移矢量,在目标图像中确定待处理子块的对应子块;根据对应子块的运动矢量确定待处理子块的运动矢量。本申请通过限制目标偏移矢量的范围,能够在获取对应子块的运动矢量时,减少内存读取次数。

Description

获取运动矢量的方法和装置
本申请要求于2018年06月04日提交中国专利局、申请号为201810564560.8、申请名称为“获取运动矢量的方法和装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及视频编解码技术领域,并且更具体地,涉及一种获取运动矢量的方法和装置。
背景技术
数字视频能力可并入到多种多样的装置中,包含数字电视、数字直播系统、无线广播系统、个人数字助理(personal digital assistant,PDA)、膝上型或桌上型计算机、平板计算机、电子图书阅读器、数码相机、数字记录装置、数字媒体播放器、视频游戏装置、视频游戏控制台、蜂窝式或卫星无线电电话(所谓的“智能电话”)、视频电话会议装置、视频流式传输装置及其类似者。数字视频装置实施视频压缩技术,例如,在由MPEG-2、MPEG-4、ITU-T H.263、ITU-T H.264/MPEG-4第10部分高级视频编码(AVC)定义的标准、视频编码标准H.265/高效视频编码(high efficiency video coding,HEVC)标准以及此类标准的扩展中所描述的视频压缩技术。视频装置可通过实施此类视频压缩技术来更有效率地发射、接收、编码、解码和/或存储数字视频信息。
视频压缩技术执行空间(图像内)预测和/或时间(图像间)预测以减少或去除视频序列中固有的冗余。对于基于块的视频编码,视频条带(即,视频帧或视频帧的一部分)可分割成若干图像块,所述图像块也可被称作树块、编码单元(coding unit,CU)和/或编码节点。使用关于同一图像中的相邻块中的参考样本的空间预测来编码图像的待帧内编码(I)条带中的图像块。图像的待帧间编码(P或B)条带中的图像块可使用相对于同一图像中的相邻块中的参考样本的空间预测或相对于其它参考图像中的参考样本的时间预测。图像可被称作帧,且参考图像可被称作参考帧。
其中,包含HEVC标准在内的各种视频编码标准提出了用于图像块的预测性编码模式,即基于已经编码的视频数据块来预测当前待编码的块。在帧内预测模式中,基于与当前块在相同的图像中的一或多个先前经解码相邻块来预测当前块;在帧间预测模式中,基于不同图像中的已经解码块来预测当前块。
在HEVC的帧间预测中,CU中的所有像素采用相同的运动信息进行运动补偿,以得到CU中的像素的预测值。但是CU内的像素并不一定都具有相同的运动特性,因此,采用相同的运动信息对CU内所有的像素进行预测可能会降低运动补偿的准确性,进而增加了残差信息。
为了解决上述问题,现有方案提出了高级时域运动矢量预测(advanced temporal motion  vector prediction,ATMVP)技术。
利用ATMVP技术对进行预测的过程主要包括:
(1)确定当前待处理块的偏移矢量;
(2)根据当前待处理中的待处理子块的位置以及上述偏移矢量,在对应图像中确定待处理子块的对应子块;
(3)根据对应子块的运动矢量,确定当前待处理子块的运动矢量;
(4)根据待处理子块的运动矢量对待处理子块进行运动补偿预测,得到待处理子块的预测像素值。
但是ATMVP技术中确定的对应子块的在对应图像中的范围不定,需要从对应图像中频繁读取对应子块的运动信息,进而会增加内存读取带宽,增加预测过程的复杂度。
发明内容
本申请提供一种获取运动矢量的方法和装置,以减少内存读取次数。
第一方面,提供一种获取运动矢量的方法,该方法包括:确定第一取值区间和第二取值区间;确定待处理块的目标偏移矢量和目标图像的标识信息,该待处理子块包含至少一个待处理子块;根据待处理子块的位置和目标偏移矢量,在目标图像中确定待处理子块的对应子块;根据对应子块的运动矢量确定待处理子块的运动矢量。
其中,上述目标偏移矢量在第一方向上的分量的取值在第一取值区间内,上述目标偏移矢量在第二方向上的分量的取值在第二取值区间内,第一方向和第二方向呈正交关系。
应理解,可以通过对待处理块进行再次划分来得到至少一个待处理子块。进一步地,当待处理块为编码单元CU时,待处理子块就是对CU进行划分后得到的子CU。
此外,上述待处理块在目标图像中所对应的(图像)块可以称为对应块,对应块中包含至少一个对应子块,上述目标偏移矢量用于确定待处理块在目标图像中的对应块(的位置)。应理解,这里的目标偏移矢量与正常的运动矢量不同,这里的目标偏移矢量是用于确定对应块的矢量,而运动矢量的一般用于确定某个图像块的预测块,从而得到该图像块的预测像素值另外,偏移矢量还可以称为时间矢量(temporal vector),目标图像也可以称为对应图像。
上述第一取值区间和第二取值区间可以采用同一语法元素来指示。
具体地,上述第一取值区间包含第一阈值和第二阈值,上述第二取值区间包含第三阈值和第四阈值,通过解析语法元素可以获取第一阈值。第二阈值、第三阈值以及第四阈值与第一阈值有一定的倍数关系(或者其它数值关系),这样当获取到了第一阈值之后,就可以根据第二阈值至第四阈值与第一阈值的倍数关系来获取第二阈值至第四阈值了。
可选地,上述目标图像的标识信息为目标图像所在的图像帧的图像顺序序号POC。
应理解,上述POC用于指示图像在原图像序列中的实际位置,为了与编码/解码顺序相区别,也可以将POC称为显示顺序或者播放顺序。
可选地,上述目标图像的标识信息为目标图像所在的图像帧的索引信息或者索引值。
本申请中,由于目标偏移矢量的分量值在一定的取值范围内,因此,根据目标偏移矢量确定的在目标图像中确定的待处理子块也位于目标图像的一定范围内,进而可以通过预读取目标图像的一定范围内的时域运动矢量场的方式来减少内存带宽和内存的读取次数。
结合第一方面,在第一方面的某些实现方式中,在根据对应子块的运动矢量确定待处理子块的运动矢量之前,上述方法还包括:获取目标图像中预设区域中的像素点的运动矢量,该预设区域为待处理子块的对应子块的可能出现的范围。
本申请中,由于对应子块在目标图像中出现的范围是确定的,因此,通过预先获取目标图像中一定区域内的运动矢量,能够减少从目标图像中读取运动矢量的次数,也就是减少了内存的读取次数,另外,由于只获取目标图像中一定范围内的运动矢量,可以减少内存带宽。
应理解,上述第一取值范围可以是由第一阈值和第二阈值组成的取值区间,上述第二取值范围可以是由第三阈值和第四阈值组成的取值区间,其中,第一阈值小于第二阈值,第三阈值小于第四阈值。
可选地,上述第一取值区间为(第一阈值,第二阈值)、(第一阈值,第二阈值]、[第一阈值,第二阈值)以及[第一阈值,第二阈值]中的任意一种。
可选地,上述第二取值区间为(第三阈值,第四阈值)、(第三阈值,第四阈值]、[第三阈值,第四阈值)以及[第三阈值,第四阈值]中的任意一种。
也就是说,上述第一取值区间和第二取值区间可以是开区间、闭区间和半开半闭区间中的任一种。
应理解,当上述第一区间为(第一阈值,第二阈值)时,上述目标偏移矢量在第一方向上的分量的取值在第一取值区间内包括:上述目标偏移矢量在第一方向上的分量的取值大于第一阈值,且小于第二阈值。
当上述第一区间为[第一阈值,第二阈值]时,上述目标偏移矢量在第一方向上的分量的取值在第一取值区间内包括:上述目标偏移矢量在第一方向上的分量的取值大于或者等于第一阈值,且小于或者等于第二阈值。
结合第一方面,在第一方面的某些实现方式中,确定待处理块的目标偏移矢量,包括:确定待处理块的初始偏移矢量;当初始偏移矢量在第一方向上的分量的取值在第一取值区间内,且在第二方向上的分量的取值在第二取值区间内时,将初始偏移矢量确定为目标偏移矢量;当初始偏移矢量在第一方向和/或第二方向上的分量的取值超出第一取值区间和/或第二取值区间时,将初始偏移矢量在第一方向和/或第二方向上的分量的取值超出第一取值区间和/或第二取值区间的部分进行钳位处理,并将钳位处理后的初始偏移矢量确定为目标偏移矢量。
其中,钳位后的初始偏移矢量在第一方向和第二方向上的分量的取值分别在第一取值区间内和第二取值区间内。
通过对初始偏移矢量在第一方向和第二方向上的分量的取值进行限制,能够使得最终得到的目标偏移矢量在第一方向和第二方向上的分量的取值分别在第一取值区间内和第二取值区间内。
可选地,当初始偏移矢量在第一方向和/或第二方向上的分量的取值超出第一取值区间和/或第二取值区间时,将初始偏移矢量在第一方向和/或第二方向上的分量的取值超出第一取值区间和/或第二取值区间的部分进行钳位处理,并将钳位处理后的初始偏移矢量确定为目标偏移矢量,包括:
当初始偏移矢量在第一方向上的分量的取值在第一取值区间内,且在第二方向上的分 量的取值在第二取值区间外时,对初始偏移矢量在第二方向上的分量超出第二取值区间的部分进行钳位处理,将钳位处理后的初始偏移矢量确定为目标偏移矢量;
当初始偏移矢量在第一方向上的分量的取值在第一取值区间外,且在第二方向上的分量的取值在第二取值区间内时,对初始偏移矢量在第一方向上的分量超出第一取值区间的部分进行钳位处理,将钳位处理后的初始偏移矢量确定为目标偏移矢量;
当初始偏移矢量在第一方向上的分量的取值在第一取值区间外,且在第二方向上的分量的取值在第二取值区间外时,对初始偏移矢量在第一方向上的分量超出第一取值区间的部分进行钳位处理,对初始偏移矢量在第二方向上的分量超出第二取值区间的部分进行钳位处理,将钳位处理后的初始偏移矢量确定为目标偏移矢量。
应理解,当初始偏移矢量在某个方向上的分量值超过该方向上能够取得的最大值时,钳位处理实际上就是对分量值超出该最大值部分进行截断,使得截断后的分量值小于或者等于该最大值;当初始偏移矢量在某个方向上的分量值小于该方向上能够取得的最小值时,钳位处理实际上就是对分量值进行补偿,也就是增加该分量值的大小,使得补偿后的分量值的大于或者等于该最小值。
结合第一方面,在第一方面的某些实现方式中,上述确定待处理块的初始偏移矢量,包括:将待处理块的预设空域相邻块的一个运动矢量确定为初始偏移矢量。
可选地,将待处理块的预设空域相邻块的一个运动矢量确定为初始偏移矢量,包括:按照预设顺序,将第一个可得的空域相邻块的运动矢量确定为初始偏移矢量。
直接将第一个可得的空域相邻块的运动矢量确定为初始偏移矢量,能够简化确定初始偏移矢量的过程。
结合第一方面,在第一方面的某些实现方式中,上述确定待处理块的初始偏移矢量,包括:从待处理块的预设空域相邻块中确定目标空域相邻块,其中,目标空域相邻块的运动矢量在第一方向上的分量的取值在第一取值区间内,在第二方向上的分量的取值在第二取值区间内;将目标空域相邻块的运动矢量作为目标偏移矢量。
通过直接将空域相邻块中分量值位于第一取值区间和第二取值区间内的运动矢量确定为目标偏移矢量,能够简化确定目标运动矢量的过程。
结合第一方面,在第一方面的某些实现方式中,确定第一取值区间和第二取值区间,包括:根据待处理块的位置和尺寸、待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围,确定第一取值区间和第二取值区间。
可选地,根据待处理块的位置和尺寸、待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围,确定第一取值区间,包括:根据待处理块的位置和尺寸、待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围确定第一阈值和第二阈值,其中,第一阈值为目标偏移矢量在第一方向上的分量能够取得的最小值,第二阈值为目标偏移矢量在第一方向上的分量能够取得的最大值;根据第一阈值和第二阈值,得到第一取值区间。
可选地,根据待处理块的位置和尺寸、待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围,确定第二取值区间,包括:根据待处理块的位置和尺寸、待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围确定第三阈值和第四阈值,其中,第三阈值为目标偏移矢量在第 二方向上的分量能够取得的最小值,第四阈值为目标偏移矢量在第二方向上的分量能够取得的最大值;根据第三阈值和第四阈值,得到第二取值区间。
结合第一方面,在第一方面的某些实现方式中,CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在第一方向和第二方向上的扩展范围。
可选地,上述第一方向为水平方向,上述第二方向为竖直方向。
可选地,当上述第一方向为水平方向,上述第二方向为竖直方向时,CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在上、下、左、右方向上的扩展范围。
结合第一方面,在第一方面的某些实现方式中,CTU的时域运动矢量场的扩展范围信息携带于序列参数集SPS、图像参数集PPS以及头字段中的至少一种。
可选地,上述CTU的时域运动矢量场的扩展范围是预先设定好的。
结合第一方面,在第一方面的某些实现方式中,根据对应子块的运动矢量确定待处理子块的运动矢量,包括:根据待处理子块所在的图像帧的图像顺序序号POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图像的目标参考帧的POC,对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
可选地,根据待处理子块所在的图像帧的图像顺序序号POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图像的目标参考帧的POC,对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量,包括:确定待处理子块所在的图像帧的图像顺序序号POC与待处理子块所在的图像帧的目标参考帧的POC的差值为第一差值;确定目标图像的POC与目标图像的目标参考帧的POC的差值为第二差值;根据第一差值和第二差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
可选地,根据第一差值和第二差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量,包括:确定第一差值和第二差值的比值为第一比值;将第一比值与对应子块的运动矢量的乘积确定为待处理子块的运动矢量。
结合第一方面,在第一方面的某些实现方式中,待处理子块的运动矢量,根据以下公式获得:
Figure PCTCN2019090031-appb-000001
其中,P1表示待处理子块所在的图像帧的POC,P2表示待处理子块所在的图像帧的目标参考帧的POC,P3表示目标图像的POC,P4表示目标图像的目标参考帧的POC,MV表示对应子块的运动矢量,MVs表示待处理子块的运动矢量。
可选地,MV分解为水平方向的运动矢量MVx和竖直方向的运动矢量MVy,分别按照上述公式计算,分别获得水平方向的运动矢量MVsx和竖直方向的运动矢量MVsy。
第二方面,提供一种获取运动矢量的方法,该方法包括:确定待处理块的目标偏移矢量和目标图像的标识信息,其中,待处理块包含至少一个待处理子块;根据待处理子块的位置和目标偏移矢量,确定出待处理子块在目标图像中的初始对应位置点;当初始对应位置点位于目标图像的预设区域内时,将初始对应位置点确定为目标对应位置点;当初始对应位置点超出预设区域时,将预设区域内距离初始对应位置点最近的位置确定为目标对应位置点;根据目标对应位置点对应的运动矢量确定待处理子块的运动矢量。
应理解,待处理块可以由至少一个待处理子块组成,获取待处理块的运动矢量可以分 解成获取每个待处理子块的运动矢量。另外,上述待处理块可以是CU,待处理子块为子CU。
上述待处理块在目标图像中对的块可以称为对应块,对应块中包含至少一个对应子块,上述目标偏移矢量用于确定待处理块在目标图像中的对应块(的位置)。
应理解,上述目标偏移矢量与正常的运动矢量不同,这里的目标偏移矢量是用于确定对应块的矢量,而运动矢量的一般用于确定某个图像块的预测块,从而得到该图像块的预测像素值另外,偏移矢量还可以称为时间矢量(temporal vector),目标图像也可以称为对应图像。
应理解,对应位置点是相对于预设区域而言的,当预设区域以单点为基本组成单位时,对应位置点也以单点的形式存在;当预设区域以点集为基本组成单位时,对应位置点也以点集的形式存在。对应的,预设区域内的某一位置和对应位置点的距离,即为预设区域内某一单点和对应位置点的距离,或者,预设区域内某一点集和对应位置点的点集的距离。
本申请中,根据初始对应位置点以及目标图像的预设区域找到位于预设区域的目标对应位置点,进而根据目标图像的一定范围的运动矢量来确定待处理子块的运动矢量,进而可以通过预读取目标图像的一定范围内的时域运动矢量场的方式来减少内存带宽和内存的读取次数。
可选地,在根据对应子块的运动矢量确定待处理子块的运动矢量之前,上述方法还包括:获取目标图像中预设区域中的像素点的运动矢量,预设区域为待处理子块的对应子块的可能出现的范围。
本申请中,由于对应子块在目标图像中出现的范围是确定的,因此,通过预先获取目标图像中一定区域内的运动矢量,能够减少从目标图像中读取运动矢量的次数,也就是减少了内存的读取次数,另外,由于只获取目标图像中一定范围内的运动矢量,可以减少内存带宽。
结合第二方面,在第二方面的某些实现方式中,预设区域是根据待处理块所在的编码树单元CTU的位置和尺寸、目标图像的尺寸以及预设的CTU的时域运动矢量场的扩展范围确定的。
可选地,上述预设区域可以用横坐标取值区间和纵坐标取值区间来限定。
结合第二方面,在第二方面的某些实现方式中,CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在第一方向和第二方向上的扩展范围。
可选地,上述第一方向为水平方向,上述第二方向为竖直方向。
可选地,当上述第一方向为水平方向,上述第二方向为竖直方向时,CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在上、下、左、右方向上的扩展范围。
结合第二方面,在第二方面的某些实现方式中,CTU的时域运动矢量场的扩展范围信息携带于序列参数集SPS、图像参数集PPS以及头字段中的至少一种。
可选地,上述CTU的时域运动矢量场的扩展范围是预先设定好的。
结合第二方面,在第二方面的某些实现方式中,根据目标对应位置点对应的运动矢量确定待处理子块的运动矢量,包括:根据待处理子块所在的图像帧的图像顺序序号POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图象的目标参考帧的POC,对目标对应位置点对应的运动矢量进行缩放,以得到待处理子块的运动矢量。
可选地,根据待处理子块所在的图像帧的图像顺序序号POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图象的目标参考帧的POC,对目标对应位置点对应的运动矢量进行缩放,以得到待处理子块的运动矢量,包括:确定待处理子块所在的图像帧的图像顺序序号POC与待处理子块所在的图像帧的目标参考帧的POC的差值为第三差值;确定目标图像的POC与目标图象的目标参考帧的POC的差值为第四差值;根据第三差值和第四差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
可选地,根据第三差值和第四差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量,包括:确定第三差值和第四差值的比值为第二比值;将第二比值与对应子块的运动矢量的乘积确定为待处理子块的运动矢量。
结合第二方面,在第二方面的某些实现方式中,待处理子块的运动矢量,根据以下公式获得:
Figure PCTCN2019090031-appb-000002
其中,P1表示待处理子块所在的图像帧的POC,P2表示待处理子块所在的图像帧的目标参考帧的POC,P3表示目标图像的POC,P4表示目标图像的目标参考帧的POC,MV表示目标对应位置点对应的运动矢量,MVs表示待处理子块的运动矢量。
可选地,MV分解为水平方向的运动矢量MVx和竖直方向的运动矢量MVy,分别按照上述公式计算,分别获得水平方向的运动矢量MVsx和竖直方向的运动矢量MVsy。
第三方面,提供一种获取运动矢量的装置,该装置包含用于执行上述第一方面或者第二方面中的任意一种实现方式中的方法的模块。
第四方面,提供一种获取运动矢量的装置,包括:相互耦合的非易失性存储器和处理器,所述处理器调用存储在所述存储器中的程序代码以执行第一方面或者第二方面中的任意一种实现方式中的方法的部分或全部步骤。
第五方面,提供一种计算机可读存储介质,所述计算机可读存储介质存储了程序代码,其中,所述程序代码包括用于执行第一方面或者第二方面中的任意一种实现方式中的方法的部分或全部步骤的指令。
第六方面,提供一种计算机程序产品,当所述计算机程序产品在计算机上运行时,使得所述计算机执行第一方面或者第二方面中的任意一种实现方式中的方法的部分或全部步骤的指令。
附图说明
图1是视频编码过程的示意图;
图2视频解码过程的示意图;
图3是本申请实施例的获取运动矢量的方法示意性流程图;
图4是待处理块的相邻块的示意图;
图5是当前图像中包含的待处理子块和目标图像中的对应子块的示意图;
图6待处理子块的运动矢量和对应子块的运动矢量的示意图;
图7是CTU的时域运动矢量场的扩展范围的示意图;
图8是当前CTU和左边CTU的时域运动矢量场的扩展范围的示意图;
图9是本申请实施例的获取运动矢量的方法示意性流程图;
图10是确定待处理子块的初始对应位置点的示意图;
图11是确定目标对应位置点的示意图;
图12是待处理子块的运动矢量和目标对应位置点的运动矢量的示意图;
图13是本申请实施例的获取运动矢量的方法的流程图;
图14是本申请实施例的获取运动矢量的方法的流程图;
图15是本申请实施例的获取运动矢量的方法的流程图;
图16是现有方案确定待处理子块的对应子块的示意图;
图17是本申请确定待处理子块的对应子块的示意图;
图18是现有融合模式构建候选信息列表的过程;
图19是应用SMVP技术后的融合模式构建候选信息列表的过程;
图20是本申请实施例的获取运动矢量的装置的示意性框图;
图21是本申请实施例的获取运动矢量的装置的示意性框图;
图22是本申请实施例的视频编码器的示意性框图;
图23是本申请实施例的视频解码器的示意性框图;
图24是本申请实施例的视频传输系统的示意性框图;
图25是本申请实施例的视频编解码装置的示意性框图;
图26是本申请实施例的视频编解码系统的示意性框图。
具体实施方式
下面将结合附图,对本申请中的技术方案进行描述。
为了更好地理解本申请实施例的获取运动矢量的方法发生的过程,下面先结合图1和图2对视频编解码的整个过程进行简单的介绍。
图1是视频编码过程的示意图。
如图1所示,在对当前帧Fn中的当前图像块进行预测时,既可以采用帧内预测也可以采用帧间预测,具体地,可以根据当前帧Fn的类型,选择采用帧内编码还是帧间编码,例如,当前帧Fn为I帧时采用帧内预测,当前帧Fn为P帧或者B帧时采用帧间预测。当采用帧内预测时可以采用当前帧Fn中已经重建区域的像素点的像素值对当前图像块的像素点的像素值进行预测,当采用帧间预测时可以采用参考帧F’ n-1中与当前图像块匹配的参考块的像素点的像素值对当前图像块的像素点的像素值进行预测。
在根据帧间预测或者帧内预测得到当前图像块的预测块之后,将当前图像块的像素点的像素值与预测块的像素点的像素值进行做差,得到残差信息,对残差信息进行变化、量化以及熵编码,得到编码码流。另外,在编码过程中还要对当前帧Fn的残差信息与当前帧Fn的预测信息进行叠加,并进行滤波操作,得到当前帧的重建帧F’ n,并将其作为后续编码的参考帧。
图2是视频解码过程的示意图。
图2所示的视频解码过程相当于图1所示的视频解码过程的逆过程,在解码时,利用熵解码以及反量化和反变换得到残差信息,并根据解码码流确定当前图像块使用帧内预测 还是帧间预测。如果是帧内预测,则利用当前帧中已重建区域内像素点的像素值按照帧内预测方法构建预测信息;如果是帧间预测,则需要解析出运动信息,并使用所解析出的运动信息在已重建的图像中确定参考块,并将参考块内像素点的像素值作为预测信息,接下来,再将预测信息与残差信息进行叠加,并经过滤波操作便可以得到重建信息。
本申请实施例的获取运动矢量的方法既可以发生在编码过程也可以发生解码过程,也就是说,编码端和解码端均可以执行本申请实施例的获取运动矢量的方法。具体地,本申请实施例的获取运动矢量的方法可以发生在图1和图2中所示的帧间预测过程中。
在HEVC中,有两种帧间预测模式,分别是先进的运动矢量预测(advanced motion vector prediction,AMVP)模式和融合(merge)模式。其中,本申请实施例的获取运动矢量的方法可以应用在帧间预测的融合模式中。
图3是本申请实施例的获取运动矢量的方法示意性流程图。图3所示的方法可以由视频编解码装置、视频编解码器、视频编解码系统以及其它具有视频编解码功能的设备来执行。
图3是本申请实施例的获取运动矢量的方法示意性流程图。图3所示的方法可以由视频编解码装置、视频编解码器、视频编解码系统以及其它具有视频编解码功能的设备来执行。
图3所示的方法包括步骤110至步骤140,下面对步骤110至步骤140进行详细的介绍。
110、确定第一取值区间和第二取值区间。
可选地,上述第一取值范围是由第一阈值和第二阈值组成的取值区间,上述第二取值范围是由第三阈值和第四阈值组成的取值区间,其中,第一阈值小于第二阈值,第三阈值小于第四阈值。
应理解,上述第一取值区间和第二取值区间可以是开区间、闭区间和半开半闭区间中的任一种。
上述第一取值区间的具体表示形式可以是(第一阈值,第二阈值),也可以是(第一阈值,第二阈值],也可以是[第一阈值,第二阈值]。
上述第二取值区间的具体表示形式可以是(第三阈值,第四阈值),也可以是(第三阈值,第四阈值],也可以是[第三阈值,第四阈值]。
120、确定待处理块的目标偏移矢量和目标图像的标识信息。
其中,上述待处理块包含至少一个待处理子块,目标偏移矢量在第一方向上的分量的取值在第一取值区间内,目标偏移矢量在第二方向上的分量的取值在第二取值区间内,第一方向和第二方向呈正交关系。
上述待处理块在目标图像中所对应的(图像)块可以称为对应块,该对应块中包含至少一个对应子块,上述目标偏移矢量用于确定待处理块在目标图像中的对应块(的位置)。上述目标偏移矢量与正常的运动矢量不同,上述目标偏移矢量是用于确定对应块的矢量,而运动矢量的一般用于确定某个图像块的预测块,从而得到该图像块的预测像素值另外,偏移矢量还可以称为时间矢量(temporal vector),目标图像也可以称为对应图像。
上述目标图像的标识信息具体可以是目标图像所在的图像帧的图像顺序序号(picture order count,POC),也可以是目标图像所在的图像帧的索引信息或者索引值。
应理解,上述POC用于指示图像在原图像序列中的实际位置,为了与编码/解码顺序相区别,也可以将POC称为显示顺序或者播放顺序。
应理解,上述至少一个待处理子块可以是对待处理块进行划分得到的,该至少一个待处理子块可以组成待处理块。当待处理块为CU时,CU可以包含至少一个子CU。另外,上述第一方向和第二方向可以分别是水平方向和竖直方向。
当第一取值区间的形式不同时,目标偏移矢量在第一方向上的分量的取值在第一取值区间内的具体形式包含以下几种:
(1)当第一取值区间为(第一阈值,第二阈值)时,目标偏移矢量在第一方向上的分量的取值大于第一阈值且小于第二阈值;
(2)当第一取值区间为(第一阈值,第二阈值]时,目标偏移矢量在第一方向上的分量的取值大于第一阈值且小于或者等于第二阈值;
(3)当第一取值区间为[第一阈值,第二阈值]时,目标偏移矢量在第一方向上的分量的取值大于或者等于第一阈值且小于或者等于第二阈值。
类似地,当第二取值区间的形式不同时,目标偏移矢量在第二方向上的分量的取值在第二取值区间内的具体形式包含以下几种:
(4)当第二取值区间为(第三阈值,第四阈值)时,目标偏移矢量在第二方向上的分量的取值大于第三阈值且小于第四阈值;
(5)当第二取值区间为(第三阈值,第四阈值]时,目标偏移矢量在第二方向上的分量的取值大于第三阈值且小于或者等于第四阈值;
(6)当第二取值区间为[第三阈值,第四阈值]时,目标偏移矢量在第二方向上的分量的取值大于或者等于第三阈值且小于或者等于第四阈值。
在上述步骤120中,可以采用多种方式来确定目标偏移矢量。具体地,既可以先确定一个初始偏移矢量(也可以称为起始偏移矢量),然后将初始偏移矢量在第一方向和第二方向上的分量的取值分别限制到第一取值区间内和第二取值区间内,得到最终的目标偏移矢量。也可以直接从多个运动矢量中选择出在第一方向和第二方向上的分量的取值分别在第一取值区间内和第二取值区间内的运动矢量作为目标偏移矢量。下面对这两种确定目标偏移矢量的方式进行详细介绍。
第一种方式:确定初始偏移矢量,对初始偏移矢量在第一方向和第二方向上的分量值进行限制,得到目标偏移矢量。
具体地,在第一种方式中,确定目标偏移矢量包括步骤1201至步骤1203,下面对步骤1201至步骤1203进行详细介绍。
1201、确定待处理块的初始偏移矢量。
可选地,将待处理块的预设空域相邻块的一个运动矢量确定为该初始偏移矢量。
具体地,可以按照一定的顺序,将获取的第一个可得的预设空域相邻块的运动矢量确定为该初始偏移矢量。
例如,如图4所示,A 0、A 1、B 0、B 1和B 2是待处理块的空域相邻块,T为待处理块的时域相邻块,C是待处理块的参照帧中与待处理块处于相同位置的图像块。在确定待处理块的初始偏移矢量时,可以按照A 1、B 1、B 0、A 0的顺序,将找到的第一个可得的相邻块的运动矢量确定为初始偏移矢量。
通过直接将第一个可得的空域相邻块的运动矢量确定为初始偏移矢量,能够简化确定初始偏移矢量的过程。
1202、当初始偏移矢量在第一方向上的分量的取值在第一取值区间内,且在第二方向上的分量的取值在第二取值区间内时,将初始偏移矢量确定为目标偏移矢量。
例如,当第一取值区间为(第一阈值,第二阈值),第二取值区间为(第三阈值,第四阈值)时,如果初始偏移矢量在第一方向上的分量的取值大于第一阈值且小于第二阈值,并且初始偏移矢量在第二方向上的分量的取值大于第三阈值且小于第四阈值时,可以直接将初始偏移矢量确定为目标偏移矢量。
1203、当初始偏移矢量在第一方向和/或第二方向上的分量的取值超出第一取值区间和/或第二取值区间时,将初始偏移矢量在第一方向和/或第二方向上的分量的取值超出第一取值区间和/或第二取值区间的部分进行钳位处理,并将钳位处理后的初始偏移矢量确定为目标偏移矢量。
其中,步骤1203中得到的钳位后的初始偏移矢量在第一方向和第二方向上的分量的取值分别在第一取值区间内和第二取值区间内。
具体地,当初始偏移矢量在第一方向和/或第二方向上的分量的取值超出第一取值区间和/或第二取值区间时,对初始偏移矢量的处理包含情况1至情况3,下面对这三种情况进行详细的描述。
情况1:
当初始偏移矢量在第一方向上的分量的取值在第一取值区间内,且在第二方向上的分量的取值在第二取值区间外时,对初始偏移矢量在第二方向上的分量超出第二取值区间的部分进行钳位处理,将钳位处理后的初始偏移矢量确定为目标偏移矢量。
例如,第一取值区间为[第一阈值,第二阈值],第二区间为[第三阈值,第四阈值],初始偏移矢量在第一方向和第二方向上的分量为第一分量和第二分量,其中,第一分量小于第二阈值且大于第一阈值,第二分量大于第四阈值。那么,初始偏移矢量的第二分量超出[第三阈值,第四阈值]所限定的取值范围,需要将第二分量的数值超出第四阈值的部分进行钳位处理(或者,也可以称为截断处理),使得钳位后的第二分量大于或者等于第三阈值且小于或者等于第四阈值。
情况2:
当初始偏移矢量在第一方向上的分量的取值在第一取值区间外,且在第二方向上的分量的取值在第二取值区间内时,对初始偏移矢量在第一方向上的分量超出第一取值区间的部分进行钳位处理,将钳位处理后的初始偏移矢量确定为目标偏移矢量。
例如,第一取值区间为[第一阈值,第二阈值],第二区间为[第三阈值,第四阈值],初始偏移矢量在第一方向和第二方向上的分量为第一分量和第二分量,其中,第二分量大于第三阈值且小于第四阈值,第一分量小于第一阈值。那么,初始偏移矢量的第一分量超出[第一阈值,第二阈值]所限定的取值范围,需要将第一分量的取值钳位到[第一阈值,第二阈值]所限定的取值范围,也就是要在第一分量的基础上加上第一阈值与第一分量的差值,使得限制后的第一分量位于第一取值区间内。
情况3:
当初始偏移矢量在第一方向上的分量的取值在第一取值区间外,且在第二方向上的分 量的取值在第二取值区间外时,对初始偏移矢量在第一方向上的分量超出第一取值区间的部分进行钳位处理,对初始偏移矢量在第二方向上的分量超出第二取值区间的部分进行钳位处理,将钳位处理后的初始偏移矢量确定为目标偏移矢量。
例如,第一取值区间为[第一阈值,第二阈值],第二区间为[第三阈值,第四阈值],初始偏移矢量在第一方向和第二方向上的分量为第一分量和第二分量,其中,第一分量大于第二阈值,第二分量大于第四阈值。那么,初始偏移矢量的第一分量超出了[第一阈值,第二阈值]所限定的取值范围,初始偏移矢量的第二分量超出了[第三阈值,第四阈值]所限定的取值范围,需要将第一分量的数值超出第二阈值的部分进行钳位,将第二分量的数值超出第四阈值的部分进行钳位,使得钳位后的第一分量的数值在[第一阈值,第二阈值]的范围内,钳位后的第二分量的数值在[第三阈值,第四阈值]的范围内。
第二种方式:从待处理块的空域相邻块中的运动矢量在第一方向和第二方向上的分量值满足要求的运动矢量确定为目标偏移矢量。
具体地,在第二种方式中,确定目标偏移矢量包括步骤1204和步骤1205,下面对步骤1204和步骤1205进行详细介绍。
1204、从待处理块的预设空域相邻块中确定目标空域相邻块,其中,目标空域相邻块的运动矢量在第一方向上的分量的取值在第一取值区间内,在第二方向上的分量的取值在第二取值区间内;
具体地,在步骤1204中,可以按照一定的顺序获取待处理块的空域相邻块的运动矢量,然后确定这些空域相邻块的运动矢量在第一方向和第二方向上的分量的取值是否分别在第一取值区间和第二取值区间,并将在第一方向和第二方向上的分量的取值分别在第一取值区间内和第二取值区间内的运动矢量确定为目标偏移矢量。
1205、将目标空域相邻块的运动矢量作为目标偏移矢量。
例如,如图4所示,可以按照A 1、B 1、B 0、A 0的顺序,将找到的第一个可得的相邻块的运动矢量,然后确定该运动矢量的分量值是否满足要求(在第一方向和第二方向上的分量的取值分别在第一取值区间内和第二取值区间内),如果该运动矢量满足要求就把该运动矢量确定为目标偏移矢量,如果该运动矢量不满足要求,那么就按照A 1、B 1、B 0、A 0的顺序继续寻下一个可得的相邻块的运动矢量,然后再确定是否满足要求,直到找到满足要求的运动矢量。
应理解,步骤120中采用的目标偏移矢量还可以是零偏移矢量,此时,目标图像中与待处理块处于相同位置的图像块为待处理块在目标图像中的对应块。
另外,在寻找不到满足要求的目标偏移矢量时,也可以不采用ATMVP技术,而是采用其它技术来获取待处理子块的运动矢量。
130、根据待处理子块的位置和目标偏移矢量,在目标图像中确定待处理子块的对应子块。
由于目标偏移矢量指向待处理块在目标图像中的对应块,因此,可以先根据偏移矢量得到对应块,然后再根据待处理子块的位置在目标图像中确定与待处理子块具有相对位置关系的对应子块(也可以理解为在对应块中确定与待处理子块具有相对位置关系的对应子块)。
在对应块中确定与待处理子块具有相对位置关系的对应子块时,可以将对应块中与待 处理子块中相对位置相同的子块,确定为待处理子块的对应子块。
例如,如图5所示,当前图像中包含待处理块,待处理块包含4个子块,通过待处理块的位置以及待处理块的偏移矢量可以得到目标图像中的与待处理块相对应的对应块,对应块也包含4个子块,待处理子块位于待处理块的左上角位置,那么,在确定待处理子块的对应子块时,可以将对应块中位于左上角的子块确定为待处理子块的对应子块。
140、根据对应子块的运动矢量确定待处理子块的运动矢量。
本申请中,由于目标偏移矢量的分量值在一定的取值范围内,因此,根据目标偏移矢量确定的在目标图像中确定的待处理子块也位于目标图像的一定范围内,进而可以通过预读取目标图像的一定范围内的时域运动矢量场的方式来减少内存带宽和内存的读取次数。
可选地,在步骤140之前,图3所示的方法还包括:获取目标图像中预设区域中的像素点的运动矢量,该预设区域为待处理子块的对应子块的可能出现的范围。
本申请中,由于对应子块在目标图像中出现的范围是确定的,因此,通过预先获取目标图像中一定区域内的运动矢量,能够减少从目标图像中读取运动矢量的次数,也就是减少了内存的读取次数,另外,由于只获取目标图像中一定范围内的运动矢量,可以减少内存带宽。
在得到了对应子块的运动矢量之后,可以通过对对应子块的运动矢量进行缩放来得到待处理子块的运动矢量。
具体地,根据对应子块的运动矢量确定待处理子块的运动矢量,包括:根据待处理子块所在的图像帧的图像顺序序号(picture order count,POC)、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图像的目标参考帧的POC,对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
例如,如图6所示,待处理子块所在的图像帧为图像帧1,图像帧1的目标参考帧为图像帧2,目标图像所在的图像帧为图像帧3,图像帧3的目标参考帧为图像帧4,对应子块的运动矢量为MV,那么,可以根据图像帧1至图像帧4的POC对MV进行缩放,以得到待处理子块的运动矢量。
可选地,根据待处理子块所在的图像帧的POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图像的目标参考帧的POC,对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量,包括:将待处理子块所在的图像帧的POC与待处理子块所在的图像帧的目标参考帧的POC的差值确定为第一差值;将目标图像的POC与目标图像的目标参考帧的POC的差值确定为第二差值;根据第一差值和第二差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
在根据第一差值和第二差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量时,具体可以先计算第一差值与第二差值的比值,得到第一比值,然后将第一比值与对应子块的运动矢量的乘积确定为待处理子块的运动矢量。
具体地,仍以图6为例,可以采用公式(1)来计算待处理子块的运动矢量。
Figure PCTCN2019090031-appb-000003
其中,MV s为待处理子块的运动矢量,MV为对应子块的运动矢量,P1至P4分别是图像帧1至图像帧4的POC,P1-P2表示第一差值,P3-P4表示差值。
可选地,作为一个实施例,确定第一取值区间和所述第二取值区间,包括:根据所述待处理块的位置和尺寸、所述待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围,确定所述第一取值区间和所述第二取值区间。
应理解,可以通过确定目标偏移矢量在第一方向和第二方向上所能够取得的最大值和最小值的方式来确定第一取值区间和第二取值区间。下面对采用这种方式确定第一取值区间和第二取值区间进行详细的说明。
具体地,上述确定第一取值区间和第二取值区间包含以下过程:
(1)根据待处理块的位置和尺寸、待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围确定第一阈值和第二阈值;
(2)根据待处理块的位置和尺寸、待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围确定第三阈值和第四阈值;
(3)根据第一阈值和第二阈值,得到第一取值区间;
(4)根据第三阈值和第四阈值,得到第二取值区间。
其中,第一阈值为目标偏移矢量在第一方向上的分量能够取得的最小值,第二阈值为目标偏移矢量在第一方向上的分量能够取得的最大值;第三阈值为目标偏移矢量在第二方向上的分量能够取得的最小值,第四阈值为目标偏移矢量在第二方向上的分量能够取得的最大值。
可选地,作为一个实施例,上述CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在第一方向和第二方向上的扩展范围。
其中,上述第一方向和第二方向呈正交关系的两个方向。
具体地,上述第一方向可以为水平方向,上述第二方向可以为竖直方向。
应理解,当上述第一方向为水平方向,上述第二方向为竖直方向时,CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在上、下、左、右方向上的扩展值。
例如,如图7所示,虚线与CTU所在的矩形所围成的区域为CTU的扩展范围,CTU在上、下、左、右方向上的扩展值分别为U、B、L、R。
当上述U、B、L、R均为0时表示CTU在各个方向上的扩展值均为0,此时,当前CTU的时域运动矢量场被限制在对应位置点的CTU的时域运动矢量场。
当U和B为0,L和R不为0时,当前CTU所获取的时域运动矢量场不能跨越CTU的上、下边界,但可以跨越CTU的左、右边界。如图8所示,当前CTU以及左边CTU的MVF可以向左、右扩展,但是不能向上、下扩展,当前CTU的时域运动矢量场和左边CTU的时域运动矢量场有部分重叠。在这种情况下,在处理当前CTU时,可以通过滑动窗口的方式,部分重用其左边CTU已经读取的时域运动矢量场。
具体地,可以根据公式(2)至公式(5)来确定第一取值区间和第二取值区间。
HorMin=CTUX–CUX–L                                       (2)
HorMax=CTUX+CTUW–CUX–CUW+R                         (3)
VerMin=CTUY–CUY–U                                       (4)
VerMax=CTUY+CTUH–CUY–CUH+B                         (5)
其中,CUX和CUY分别表示待处理块的左上顶点像素的横坐标和纵坐标,CUW和CUH表示待处理块的宽和高,CTUX和CTUY分别表示待处理块所在的CTU的横坐标和 纵坐标,CTUW和CTUH分别为待处理块所在的CTU的宽和高,U、B、L、R分别表示CTU在上、下、左、右方向上的扩展值,HorMin和HorMax分别表示第一阈值和第二阈值,VerMin和VerMax分别表示第三阈值和第四阈值。
可选的,上述CTU的时域运动矢量场的扩展范围信息携带于序列参数集SPS、图像参数集PPS以及头字段中的至少一种。
应理解,SPS、PPS以及头字段可以通过携带的参数(U、B、L、R)来指示CTU的扩展范围。
可选地,上述CTU的时域运动矢量场的扩展范围为预先设定好的。在这种情况下,可以直接将CTU的运动矢量的扩展的范围直接写入到协议中,而无需再通过SPS、PPS或者头字段来携带。
图9是本申请实施例的获取运动矢量的方法示意性流程图。图9所示的方法可以由视频编解码装置、视频编解码器、视频编解码系统以及其它具有视频编解码功能的设备来执行。
图9所示的方法包括步骤210至步骤250,下面对步骤210至步骤250进行详细的介绍。
210、确定待处理块的目标偏移矢量和目标图像的标识信息。
其中,上述待处理块包含至少一个待处理子块,该待处理块由至少一个待处理子块组成。在获取待处理块的运动矢量时实际上是要获取该待处理块中的每个待处理子块的运动矢量。当上述待处理块为CU时,待处理子块就为子CU。
步骤210中的目标偏移矢量与图3所示的方法中提及的目标偏移矢量的含义一致,这里不再详细论述。
上述目标图像的标识信息具体可以是目标图像所在的图像帧的POC,也可以是目标图像所在的图像帧的索引信息或者索引值。
220、根据待处理子块的位置和偏移矢量,确定出待处理子块在目标图像中的初始对应位置点。
具体地,在确定待处理子块在目标图像中的初始对应位置点时,可以先确定待处理子块的对应子块,然后将对应子块的某个设定位置确定为待处理子块在目标图像中的初始对应位置点。例如,在确定了对应位置点后,可以将对应子块的中心点或者左上角的位置确定为待处理子块在目标图像中的初始对应位置点。
其中,确定对应子块的方式可以参见图3所示的方法中步骤130下方的相关段落的描述。
另外,在确定待处理子块在目标图像中的初始对应位置点时,也可以不确定对应子块,而是直接按照待处理子块的位置以及目标偏移矢量,直接确定待处理子块对应在目标图像中的位置确定为待处理子块在目标图像中的初始对应位置点。例如,先确定待处理子块的某个设定位置,如中心点或者左上角,将该位置坐标加上目标偏移矢量,即可确定待处理子块在目标图像中的初始对应位置点。
230、当初始对应位置点位于目标图像的预设区域内时,将初始对应位置点确定为目标对应位置点。
例如,如图10所示,当前图像中包括待处理块,待处理块中包括位于左上角位置的 待处理子块,目标图像包括对应块,对应块中包括位于左上角位置的对应子块,待处理块的目标偏移矢量指向对应块,目标图像虚线框内的矩形区域为目标图像的预设区域,由图10可知,待处理子块在目标图像中的初始对应位置点位于目标图像的预设区域,可以将初始对应位置点直接确定为目标对应位置点。
240、当初始对应位置点超出预设区域时,将预设区域内距离初始对应位置点最近的位置确定为目标对应位置点。
例如,如图11所示,当前图像中包括待处理块,待处理块中包括位于左上角位置的待处理子块,目标图像包括对应块,对应块中包括位于左上角位置的对应子块,待处理块的目标偏移矢量指向对应块,目标图像虚线框内的矩形区域为目标图像的预设区域,由图11可知,待处理子块在目标图像中的初始对应位置点位于目标图像的预设区域之外,在这种情况下,需要将预设区域中距初始对应位置点最近的位置A确定为目标对应位置点。
250、根据目标对应位置点对应的运动矢量确定待处理子块的运动矢量。
本申请中,根据初始对应位置点以及目标图像的预设区域找到位于预设区域的目标对应位置点,进而根据目标图像的一定范围的运动矢量来确定待处理子块的运动矢量,进而可以通过预读取目标图像的一定范围内的时域运动矢量场的方式来减少内存带宽和内存的读取次数。
可选地,在步骤250之前,图9所示的方法还包括:获取目标图像中预设区域中的像素点的运动矢量。
应理解,在步骤250中根据目标对应位置点的运动矢量确定待处理子块的运动矢量之前,需要先获取目标对应位置点对应的运动矢量。为了减少内存读取的次数以及占用的内存带宽,可以预先获取目标图像的预设区域内像素点的运动矢量,这样需要使用目标对应位置点的运动矢量时可以直接从已经读取的预设区域内像素点的运动矢量获取目标对应位置点对应的运动矢量,并且,由于只获取目标图像预设区域内的像素点的运动矢量,而不像传统方案那样可能需要获取整个目标图像,能够减少占用的内存带宽。
可选地,作为一个实施例,上述步骤230中的预设区域根据待处理块所在的编码树单元CTU的位置和尺寸、目标图像的尺寸以及预设的CTU的时域运动矢量场的扩展范围确定的。
上述CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在第一方向和第二方向上的扩展范围。
可选地,上述第一方向为水平方向,上述第二方向为竖直方向。
可选地,当上述第一方向为水平方向,上述第二方向为竖直方向时,CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在上、下、左、右方向上的扩展范围。
上述预设区域可以用多种方式进行限定,例如,可选地,上述预设区域可以用横坐标取值区间和纵坐标取值区间来表示。
具体地,上述预设区域可以用横坐标取值区间[横坐标最小值,横坐标最大值]和纵坐标取值区间[纵坐标最小值,纵坐标最大值]来表示。
其中,横坐标取值和纵坐标取值区间中的取值可以采用公式(6)至公式(9)进行计算。
HorMin=Min(CTUX–L,0)                                 (6)
HorMax=Max(CTUX+CTUW+R–1,PicW–1)                 (7)
VerMin=Min(CTUY–U,0)                                  (8)
VerMax=Max(CTUY+CTUH+B–1,PicH–1)                  (9)
其中,CTUX为待处理块所在的CTU的横坐标,CTUY为待处理块所在的CTU的纵坐标,CTUW和CTUH分别为待处理块所在的CTU的宽和高,PicW和PicH分别为目标图像的宽和高,U、B、L、R分别表示CTU在上、下、左、右方向上的扩展值,HorMin和HorMax分别为横坐标最小值和横坐标最大值,VerMin和VerMax分别为纵坐标最小值和纵坐标最大值。
在根据公式(6)至公式(9)计算得到了横纵坐标的最小值和最大值之后就可以得到横坐标取值区间和纵坐标取值区间了,也就确定了目标图像的预设区域了。
可选地,上述CTU的时域运动矢量场的扩展范围信息携带在SPS、PPS以及头字段中的至少一种。
另外,上述CTU的时域运动矢量场的扩展范围也可以是预先设定好的。在这种情况下,可以直接将CTU的运动矢量的扩展的范围直接写入到协议中,而无需再通过SPS、PPS或者头字段来携带。
可选地,作为一个实施例,上述根据目标对应位置点对应的运动矢量确定待处理子块的运动矢量,包括:根据待处理子块所在的图像帧的POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图象的目标参考帧的POC,对目标对应位置点对应的运动矢量进行缩放,以得到待处理子块的运动矢量。
具体地,在根据待处理子块所在的图像帧的POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图象的目标参考帧的POC,对目标对应位置点对应的运动矢量进行缩放,得到待处理子块的运动矢量时,可以先确定待处理子块所在的图像帧的图像顺序序号POC与待处理子块所在的图像帧的目标参考帧的POC的差值为第三差值;确定目标图像的POC与目标图象的目标参考帧的POC的差值为第四差值;然后再根据第三差值和第四差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
可选地,根据第三差值和第四差值的比值对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量,包括:确定第三差值和第四差值的比值为第二比值;将第二比值与目标对应位置点对应的运动矢量的乘积确定为待处理子块的运动矢量。
例如,如图12所示,待处理子块所在的图像帧为图像帧1,图像帧1的目标参考帧为图像帧2,目标图像所在的图像帧为图像帧3,图像帧3的目标参考帧为图像帧4,目标对应位置点的运动矢量为MV,那么,可以根据图像帧1至图像帧4的POC对MV进行缩放,以得到待处理子块的运动矢量。
具体地,可以采用公式(10)来计算待处理子块的运动矢量。
Figure PCTCN2019090031-appb-000004
其中,MV s为待处理子块的运动矢量,MV为目标对应位置点的运动矢量,P1至P4分别是图像帧1至图像帧4的POC,P1-P2表示第三差值,P3-P4表示第四差值。
为了实现对待处理块的预测,可以在得到待处理块中全部待处理子块的运动矢量之后 再对待处理块进行预测,得到待处理块的像素预测值。也可以在得到每个待处理块的运动矢量之后,就对待处理子块进行预测,得到每个待处理子块的像素预测值,在对每个待处理子块进行预测之后,也就得到了待处理块的像素预测值。
上文结合图3至图12对本申请实施例的获取运动矢量的方法进行了介绍,为了更好地理解本申请的技术方案,下面结合具体的实例对本申请实施例的获取运动矢量的方法进行详细介绍。
其中,实施例一和实施例二与图3所示的方法是对应的,都是通过对目标偏移矢量的取值的限制使得最终找到的对应子块位于目标图像的一定区域中,而实施例三则不对目标偏移矢量的取值进行任何限制,而是将待处理子块在目标图像中的目标对应位置点直接限制在一定的区域内。最终实施例一至实施例三均实现了根据目标图像的一定区域内的像素的运动矢量来确定待处理子块的运动矢量。下面分别对实施例一至实施例三进行详细介绍。
实施例一:
如图13所示,实施例一中获取运动矢量的具体过程包括:
310、确定约束范围。
这里的约束范围相当于上文中的第一取值区间和第二取值区间。
应理解,确定约束范围是为了对待处理块的目标偏移矢量的在第一方向和第二方向上的分量值进行约束,使得目标偏移矢量在第一方向和第二方向上的分量值在约束范围内。
可选地,上述约束范围可以根据CTU的时域运动矢量场的扩展范围、待处理块的位置和尺寸,待处理块所在的CTU的位置和尺寸来确定。
其中,CTU的时域运动矢量场的扩展范围可以通过参数RangeL、RangeR、RangeU、RangeB(相当于上文中的参数L、R、U、B)来定义,参数RangeL,RangeR,RangeU,RangeB分别表示CTU的时域运动矢量场能够分别向左、右、上、下这四个方向上的扩展值。
RangeL,RangeR,RangeU,RangeB所取的数值可以是0、16、32和64等(单位为像素)。
应理解,上述参数RangeL,RangeR,RangeU,RangeB可以直接通过预定义的方式来确定,或者,也可以携带在高层语法中,例如,RangeL,RangeR,RangeU,RangeB可以携带在SPS、PPS以及头字段中。
具体地,可以根据公式(11)至公式(14)来确定上述约束范围。
HorMin=CTUX–CUX–RangeL                                     (11)
HorMax=CTUX+CTUW–CUX–CUW+RangeR                       (12)
VerMin=CTUY–CUY–RangeU                                     (13)
VerMax=CTUY+CTUH–CUY–CUH+RangeB                        (14)
其中,CUX和CUY表示待处理块左上顶点像素的位置坐标,CUW和CUH表示待处理块的宽和高,CTUX和CTUY表示待处理块所在CTU的左上顶点像素的位置坐标,CTUW和CTUH表示CTU的宽和高。HorMin和HorMax表示约束范围在水平方向能够取的最小值和最大值,VerMin和VerMax表示约束范围在竖直方向能够取的最小值和最大值。
320、确定待处理块的起始偏移矢量。
这里的起始偏移矢量相当于上文中的初始偏移矢量。
应理解,上述步骤310和步骤320既可以同时进行,也可以分别先后进行。
起始偏移矢量用于确定待处理块在目标图像中的目标图像块的位置,可以采用以下两种方式来确定起始偏移矢量。
第三种方式:将待处理块的空域相邻块中可用的运动矢量确定为起始偏移矢量。
具体地,在第三种方式中,可以按照图4中A 1,B 1,B 0,A 0的顺序,找到第一个可得的相邻块的运动矢量,并将其作为待处理块的起始偏移矢量。
第四种方式:先确定目标图像,然后再将相邻块中指向目标图像的运动矢量确定为起始偏移矢量。
具体地,在第四种方式中,可以按照图4中A 1,B 1,B 0,A 0的顺序,找到第一个可得的相邻块的运动矢量,若该运动矢量指向目标图像,则将其作为待处理块的起始偏移矢量。否则,对该运动矢量进行缩放,使得其指向目标图像,并将缩放后的运动矢量作为起始偏移矢量。
330、确定起始偏移矢量是否超出约束范围。
应理解,当起始偏移矢量在第一方向或者第二方向中至少一个方向上的分量值超出约束范围时就确定该起始偏移矢量超出约束范围;当起始偏移矢量在第一方向和第二方向上的分量值均未超出约束范围时,确定起始偏移矢量未超出约束范围。
当起始偏移矢量超出约束范围时,执行步骤340和步骤350,当起始偏移矢量未超出约束范围时,执行步骤360。
340、根据约束范围确定受限的偏移矢量;
当起始偏移矢量超出限定范围时,需要对其进行限定,得到受限的偏移矢量,使得受限后的偏移矢量在约束范围内。
具体地,当起始偏移矢量超出约束范围时,可以采用公式(15)和(16)对起始偏移矢量的分量(x off,y off)进行限制,得到受限的偏移矢量。
x off=Clip3(HorMin,HorMax,x off)                    (15)
y off=Clip3(VerMin,VerMax,y off)                     (16)
其中,Clip3函数为钳位函数,具体定义为:
Figure PCTCN2019090031-appb-000005
应理解,除了根据约束范围确定目标偏移矢量之外,还可以直接采用零运动矢量作为目标偏移矢量。或者也可以不采用ATMVP技术,而是采用其它技术来获取待处理子块的运动矢量。
350、将受限的偏移矢量确定为目标偏移矢量。
由于受限后的偏移矢量在约束范围内,因此,可以该受限后的偏移矢量确定为目标偏移矢量。
360、将初始偏移矢量确定为目标偏移矢量。
当初始偏移矢量在约束范围内时,无需再对初始偏移矢量进行任何约束或者限制操作,而可以直接将该初始偏移矢量确定为目标偏移矢量。
370、根据待处理子块的位置和目标偏移矢量,在目标图像中确定待处理子块的对应 子块。
380、根据对应子块的运动矢量确定待处理子块的运动矢量。
具体地,可以根据待处理子块所在的图像帧的POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图像的目标参考帧的POC,对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
在得到了待处理子块的运动矢量之后就可以根据待处理子块的运动矢量确定待处理子块的预测像素值。
在确定待处理子块的预测像素值时,可以根据待处理子块的运动矢量确定待处理子块的预测子块,然后将该预测子块的像素值确定为待处理子块的预测像素值。这样在得到了每个待处理子块的预测像素值之后,就可以得到整个待处理块的预测像素值。
实施例二:
如图14所示,实施例二中获取运动矢量的具体过程包括:
410、确定约束范围。
这里的约束范围的含义以及确定方式与实施例一种的约束范围相同,这里不再重复描述。
420、获取待处理块的相邻块的一个运动矢量。
具体地,可以按照图4中A 1,B 1,B 0,A 0的顺序,获取第一个可得的相邻块的运动矢量。
另外,也可以先确定目标图像,然后再按照图4中A 1,B 1,B 0,A 0的顺序,获取待处理块的相邻块的一个运动矢量,当该运动矢量指向目标图像时,执行步骤430;当该运动矢量未指向目标图像,对该运动矢量进行缩放,使得缩放后的运动矢量指向目标图像,然后再将缩放后的运动矢量作为相邻块的一个运动矢量,继续执行步骤440。
430、确定相邻块的运动矢量是否在约束范围内;
当相邻块的运动矢量的分量值在约束范围内时,执行步骤440,否则继续执行步骤420,直到相邻块的运动矢量的分量值在约束范围内。
应理解,当相邻块的运动矢量在第一方向和第二方向上的分量值均在约束范围内时,确定相邻块的运动矢量在约束范围内;当相邻块的运动矢量在第一方向和/或第二方向上的分量值超出约束范围时,确定相邻块的运动矢量不在约束范围内(相邻块的运动矢量超出约束范围)。
440、将相邻块的运动矢量确定为目标偏移矢量。
450、根据待处理子块的位置和目标偏移矢量,在目标图像中确定待处理子块的对应子块。
460、根据对应子块的运动矢量确定待处理子块的运动矢量。
具体地,可以根据待处理子块所在的图像帧的POC、待处理子块所在的图像帧的目标参考帧的POC、目标图像的POC以及目标图像的目标参考帧的POC,对对应子块的运动矢量进行缩放,以得到待处理子块的运动矢量。
在得到了待处理子块的运动矢量之后就可以根据待处理子块的运动矢量确定待处理子块的预测像素值。
实施例三:
如图15所示,实施例三中获取运动矢量的具体过程包括:
510、确定待处理块的目标偏移矢量和目标图像的标识信息。
步骤510中确定目标偏移矢量和目标图像的标识信息的过程与上述步骤210相同。
520、根据待处理子块的位置和目标偏移矢量,确定出待处理子块在目标图像中的初始对应位置点。
具体地,可以根据公式(18)来确定待处理子块在目标图像中的初始对应位置点的位置坐标。
Figure PCTCN2019090031-appb-000006
其中,(x,y)表示待处理块的左上顶点的坐标,i表示待处理块中从左到右方向的第i个子块,j表示待处理块中从上到下方向的第j个子块,(x off,y off)表示偏移运动矢量在第一方向和第二方向上的分量值,M和N为待处理子块的尺寸(其中,M可以表示待处理子块的宽,N可以表示待处理子块的高),(x (i,j),y (i,j))表示第(i,j)个对应子块的位置坐标。
530、确定初始对应位置点是否在对应图像的预设区域内。
当初始对应位置点在对应图像的预设区域内时,直接执行步骤550,当初始对应位置点超出对应图像的预设区域内时,先执行步骤540再执行步骤550。
上述预设区域可以是目标图像中的一个固定区域,可以根据(19)至(22)来确定对应图像的预设区域。
HorMin=Min(CTUX–RangeL,0)                                      (19)
HorMax=Max(CTUX+CTUW+RangeR–1,PicW–1)                    (20)
VerMin=Min(CTUY–RangeU,0)                                      (21)
VerMax=Max(CTUY+CTUH+RangeB–1,PicH–1)                     (22)
其中,HorMin和HorMax表示约束范围在水平方向能够取的最小值和最大值,
VerMin和VerMax表示约束范围在竖直方向能够取的最小值和最大值,CTUX和CTUY表示待处理块所在CTU的左上顶点像素的位置坐标,CTUW和CTUH表示CTU的宽和高,RangeU、RangeB、RangeL、RangeR分别表示CTU的时域运动矢量场在上、下、左、右方向上的扩展值,PicW和PicH分别为目标图像的宽和高。
540、将初始对应位置点确定为目标对应位置点。
当初始对应位置点在目标图像的预设区域时,可以直接将该初始对应位置点确定为目标对应位置点。
550、将预设区域内距离初始对应位置点最近的位置确定为目标对应位置点。
具体地,当初始对应位置点超出目标图像的预设区域时,可以将预设区域内距离初始对应位置点最近的位置确定为目标对应位置点。具体地,当初始对应位置点超出目标图像的预设区域时,可以根据公式(23)和公式(24)来确定目标对应位置点。
x (i,j)=Clip3(HorMin,HorMax,x (i,j))                                    (23)
y (i,j)=Clip3(VerMin,VerMax,_y (i,j))                                    (24)
其中,Clip3函数为钳位函数,具体定义如上文中的公式(17)所示。
560、根据目标对应位置点对应的运动矢量确定待处理子块的运动矢量。
具体地,可以根据待处理子块所在的图像帧的POC,待处理子块所在的图像帧的目标参考帧的POC,目标图像的POC以及目标图像的目标参考帧的POC,对目标对应位置点的运动矢量进行缩放,来获取待处理子块的运动矢量。
在得到了待处理子块的运动矢量之后就可以根据待处理子块的运动矢量确定待处理子块的预测像素值。
为了更好地理解本申请带来的有益效果,下面结合图16和图17对本申请实例的获取运动矢量的方法的有益效果进行详细的说明。
如图16所示,当前图像中包括当前CTU,当前CTU包括多个待处理块,每个待处理块包含多个待处理子块,每个待处理块对应一个偏移矢量,根据该对应的偏移矢量可以找到该待处理块的对应块,每个待处理子块也会在目标图像中对应一个对应子块。如果待处理子对应在目标图像中的对应子块所在的区域不确定,或者,待处理子块对应在目标图像中的位置不确定,那么,在获取对应子块的运动矢量时可能需要预先获取整个对应图像的时域运动矢量场,从而增加了内存带宽。
如图17所示,当前图像中包括当前CTU,当前CTU包括多个待处理块,每个待处理块包含多个待处理子块,目标图像中虚线内的区域为目标图像的预设区域。本申请通过对待处理块的偏移矢量的范围进行约束,使得待处理子对应在目标图像中的对应子块位于对应图像的预设区域,或者使得待处理子对应在目标图像中的对应位置点位于对应图像的预设区域。这样就只需要获取目标图像的运动矢量即可,而不必获取整个目标图像的运动矢量,这样可以减少内存带宽,另外,通过预取的方式还可以降低内存的读取次数。
本申请实施例的获取运动矢量的方法可以应用在帧间预测的融合模式中。
子编码单元运动矢量预测(Sub-CU based motion vector prediction,SMVP)技术将当前编码单元划分为大小为MxN的子编码单元,并推导出每个子编码单元的运动信息,然后利用各个子编码单元的运动信息进行运动补偿,得到当前编码单元的预测值。
当SMVP技术应用到融合模式时,是在原有融合模式的基础上增加了两种候选运动信息,分别为高级时域运动矢量预测(advanced temporal motion vector prediction,ATMVP)和空域时域运动矢量预测(spatial-temporal motion vector prediction,STMVP)。
其中,ATMVP技术是先确定对应位置点参考帧(collocated reference picture),然后将当前编码单元划分为MxN的子编码单元,获取当前每一个子编码单元在对应位置点参考帧中对应的子编码单元中心点位置像素的运动信息,并将其进行缩放,转换为当前每一个子编码单元的运动信息。本申请实施例的获取运动矢量的方法可以理解为一种改进后的ATMVP技术。
下面以图4所示的待处理块的相邻块为基础,结合图18和图19分别对现有的融合模式以及应用SMVP技术后的融合模式构建候选信息列表的过程进行说明。
现有融合模式构建候选信息列表的过程如图18所示。
图18所示的构建候选信息列表的过程包括:
610、插入A 1,B 1,B 0,A 2位置的运动信息;
620、插入B 2位置的运动信息;
630、插入时域位置(T或C)的运动信息;
640、填充候选运动信息列表。
应用SMVP技术后的融合模式构建候选信息列表的过程如图19所示。
图19所示的构建候选信息列表的过程包括:
710、插入A 1,B 1,B 0,A 2位置的运动信息;
720、插入ATMVP运动信息和STMVP运动信息;
730、插入B 2位置的运动信息;
740、插入时域位置(T或C)的运动信息;
750、填充候选运动信息列表。
其中,图19与18相比,多了插入ATMVP运动信息和STMVP运动信息的步骤720,其它步骤不变。
应理解,本申请实施例的获取运动矢量的方法获取的待处理子块的运动矢量就是属于上述步骤720中插入的ATMVP运动信息的一部分。
具体地,以图4所示的待处理块的相邻块为基础,在将SMVP技术应用到融合模式时,融合模式的编码时具体包含以下过程:
1)按照顺序获得空域相邻块A 1,B 1,B 0,A 0位置的运动信息,检查可用性及剔除重复项,插入候选运动信息列表;
2)获得ATMVP和STMVP运动信息,检查可用性及剔除重复项,插入候选运动信息列表;
3)当候选运动信息列表的长度小于6时,获得B 2位置的运动信息,检查可用性及剔除重复项,插入候选运动信息列表;
4)获得相邻已编码帧中T位置对应块的运动信息(若不存在,则获得C位置对应块的运动信息),进行缩放,插入候选运动信息列表;
5)若候选运动信息列表长度小于7,则进行填充,得到长度为7的候选运动信息列表;
6)遍历候选运动信息列表中的每个候选运动信息,进行运动补偿和重建,得到重建值,再根据率失真优化(rate distortion optimization,RDO)方法决策出率失真代价(rate distortion cost,RD cost)最小的候选运动信息,得到合并索引(index);
7)根据候选运动信息列表的长度,将合并索引写入码流,传递给解码端。
其中,合并模式候选运动信息列表的构建过程如图19所示。
上文结合图3至图19对本申请实施例的获取运动矢量的方法进行了详细的描述,应理解,本申请实施例的获取运动矢量的方法可以对应于图1和图2所示的帧间预测,本申请实施例的获取运动矢量的方法可以发生在图1和图2所示的帧间预测过程中。本申请实施例的获取运动矢量的方法可以具体由编码器或者解码器中的帧间预测模块来执行。另外,本申请实施例的获取运动矢量的方法可以在可能需要对视频图像进行编码和/或解码的任何电子设备或者装置内实施。
下面结合图20和图21对本申请实施例的获取运动矢量的装置进行详细的描述。其中,20所示的装置与图3、图13和图14所示的方法相对应,能够执行图3、图13和图14所示的方法中的各个步骤;图21所示的装置与图9和图15所示的方法相对应,能够执行图9和图15所示的方法中的各个步骤。为了简洁,下面适当省略重复的描述。
图20是本申请实施例的获取运动矢量的装置的示意性框图。图20所示的装置800包括:
确定模块810,该确定模块810用于:
确定第一取值区间和第二取值区间;
确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块,所述目标偏移矢量在第一方向上的分量的取值在所述第一取值区间内,所述目标偏移矢量在第二方向上的分量的取值在所述第二取值区间内,所述第一方向和所述第二方向呈正交关系;
根据所述待处理子块的位置和所述目标偏移矢量,在所述目标图像中确定所述待处理子块的对应子块;
获取模块820,用于根据所述对应子块的运动矢量获取所述待处理子块的运动矢量。
本申请中,由于目标偏移矢量的分量值在一定的取值范围内,因此,根据目标偏移矢量确定的在目标图像中确定的待处理子块也位于目标图像的一定范围内,进而可以通过预读取目标图像的一定范围内的时域运动矢量场的方式来减少内存带宽和内存的读取次数。
上述确定模块810和获取模块820可以对应于编解码装置中的预测模块(具体可以是帧间预测模块),该确定模块810和获取模块820在具体实现时,可以通过软件或硬件来实现,也可以通过软件和硬件相结合的方式来实现。
图21是本申请实施例的获取运动矢量的装置的示意性框图。图21所示的装置900包括:
确定模块910,该确定模块用于:
确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块;
根据所述待处理子块的位置和所述目标偏移矢量,确定出所述待处理子块在所述目标图像中的初始对应位置点;
当所述初始对应位置点位于所述目标图像的预设区域内时,将所述初始对应位置点确定为目标对应位置点;
当所述初始对应位置点超出所述预设区域时,将所述预设区域内距离所述初始对应位置点最近的位置确定为目标对应位置点;
获取模块920,用于根据所述目标对应位置点对应的运动矢量获取所述待处理子块的运动矢量。
本申请中,根据初始对应位置点以及目标图像的预设区域找到位于预设区域的目标对应位置点,进而根据目标图像的一定范围的运动矢量来确定待处理子块的运动矢量,进而可以通过预读取目标图像的一定范围内的时域运动矢量场的方式来减少内存带宽和内存的读取次数。
上述确定模块910和获取模块920可以对应于编解码装置中的预测模块,具体可以是帧间预测模块,确定模块910和获取模块920在具体实现时,可以通过软件或硬件来实现,也可以通过软件和硬件相结合的方式来实现。
本申请实施例的获取运动矢量的方法还可以由视频编码器或者视频解码器来执行,下 面结合图22和图23对本申请实施例的视频编码器和视频解码器的结构进行描述。
图22是本申请实施例的视频编码器的示意性框图。图22所示的视频编码器1000包括:编码端预测模块1001、变换量化模块1002、熵编码模块1003、编码重建模块1004和编码端滤波模块。
图22所示的视频编码器1000可以对视频进行编码,具体地,视频编码器1000可以执行图1所示的视频编码过程,实现对视频的编码。另外,视频编码器1000还可以执行本申请实施例的获取运动矢量的方法,视频编码器1000可以执行图3、图9以及图13至图15所示的方法的各个步骤。
本申请实施例中的获取运动矢量的装置还可以是视频编码器1000中的编码端预测模块1001,具体地,图20和图21所示的装置6000和装置7000相当于视频编码器1000中的编码端预测模块1001。
图23是本申请实施例的视频解码器的示意性框图。图23所示的视频解码器2000包括:熵解码模块2001、反变换反量化模块2002、解码端预测模块2003、解码重建模块2004和解码端滤波模块2005。
图23所示的视频解码器2000可以对视频进行解码,具体地,视频解码器2000可以执行图2所示的视频解码过程,实现对视频的解码。另外,视频解码器2000还可以执行本申请实施例的获取运动矢量的方法,视频解码器2000可以执行图3、图9以及图13至图15所示的获取运动矢量的方法的各个步骤。
本申请实施例中的获取运动矢量的装置还可以是视频解码器2000中的解码端预测模块2003,具体地,图20和图21所示的装置6000和装置7000相当于视频解码器2000中的解码端预测模块2003。
下面结合图24至图26对本申请实施例的获取运动矢量的方法的应用场景进行介绍,本申请实施例的获取运动矢量的方法可以由图24至图26所示的视频传输系统、编解码装置以及编解码系统来执行。
图24是本申请实施例的视频传输系统的示意性框图。
如图24所示,视频传输系统包括采集模块3001、编码模块3002、发送模块3003、网络传输3004、接收模块3005、解码模块3006、渲染模块3007和显示模块208组成。
其中,视频传输系统中各个模块的具体作用如下:
采集模块3001包含摄像头或者摄像头组,用于采集视频图像,并对采集到的视频图像进行编码前的处理,将光信号转化为数字化的视频序列;
编码模块3002用于对视频序列进行编码,得到码流;
发送模块3003用于将编码得到的码流发送出去;
接收模块3005用于接收发送模块3003发送的码流;
网络3004用于将发送模块3003发送的码流传输到接收模块3005;
解码模块3006用于对接收模块3005接收的码流进行解码,重建视频序列;
渲染模块3007用于对解码模块3006解码得到的重建视频序列进行渲染,以提高视频的显示效果。
图24所示的视频传输系统可以执行本申请实施例的获取运动矢量的方法,具体地,图24所示的视频传输系统中的编码模块3001和解码模块3006都可以执行本申请实施例 的获取运动矢量的方法。
下面结合图25和图26对编解码装置和编解码装置组成的编解码系统进行详细的介绍。应理解,图25和图26中所示的编解码装置和编解码系统能够执行本申请实施例的获取运动矢量的方法。
图25是本申请实施例的视频编解码装置的示意性图。该视频编解码装置50可以是专门用于对视频图像进行编码和/或解码的装置,也可以是具有视频编解码功能的电子设备,进一步地,该编解码装置50可以是无线通信系统的移动终端或者用户设备。
编解码装置50可以包括下列模块或者单元:控制器56、编解码器54、无线电接口52、天线44、智能卡46、读卡器48、小键盘34、存储器58、红外线端口42、显示器32。除了图25中所示的模块和单元之外,编解码装置50还可以包括麦克风或者任何适当的音频输入模块,该音频输入模块可以是数字或者模拟信号输入,编解码装置50还可以包括音频输出模块,该音频输出模块可以是耳机、扬声器或者模拟音频或者数字音频输出连接。编解码装置50也可以包括电池,该电池可以是太阳能电池、燃料电池等等。编解码装置50还可以包括用于与其它设备进行近程视线通信的红外线端口,该编解码装置50还可以采用包括任何适当的近程通信方式与其它设备进行通信,例如,蓝牙无线连接、USB/火线有线连接。
存储器58可以存储形式为图像的数据和音频的数据,也可以存储用于在控制器56上执行的指令。
编解码器54可以实现对音频和/或视频数据的编码和解码或者在控制器56的控制下实现对音频和/或视频数据的辅助编码和辅助解码。
智能卡46和读卡器48可以提供用户信息,也可以提供网络认证和授权用户的认证信息。智能卡46和读卡器48的具体实现形式可以是集成电路卡(universal Integrated circuit card,UICC)和UICC读取器。
无线电接口电路52可以生成无线通信信号,该无线通信信号可以是在进行蜂窝通信网络、无线通信系统或者无线局域网通信产生的通信信号。
天线44用于向其它装置(装置的数目可以为一个也可以为多个)发送在无线电接口电路52生成的射频信号,并且还可以用于从其它装置(装置的数目可以为一个也可以为多个)接收射频信号。
在本申请的一些实施例中,编解码装置50可以在传输和/或存储之前从另一设备接收待处理的视频图像数据。在本申请的另一些实施例中,编解码装置50可以通过无线或者有线连接接收图像并对接收到的图像进行编码/解码。
图26是本申请实施例的视频编解码系统7000的示意性框图。
如图26所示,视频编解码系统7000包含源装置4000及目的地装置5000。源装置4000产生经过编码后的视频数据,源装置4000也可以被称为视频编码装置或视频编码设备,目的地装置5000可以对源装置4000产生的经过编码后的视频数据进行解码,目的地装置5000也可以被称为视频解码装置或视频解码设备。
源装置4000和目的地装置5000的具体实现形式可以是如下设备中的任意一种:台式计算机、移动计算装置、笔记本(例如,膝上型)计算机、平板计算机、机顶盒、智能电话、手持机、电视、相机、显示装置、数字媒体播放器、视频游戏控制台、车载计算机, 或者其它类似的设备。
目的地装置5000可以经由信道6000接收来自源装置4000编码后的视频数据。信道6000可包括能够将编码后的视频数据从源装置4000移动到目的地装置5000的一个或多个媒体及/或装置。在一个实例中,信道6000可以包括使源装置4000能够实时地将编码后的视频数据直接发射到目的地装置5000的一个或多个通信媒体,在此实例中,源装置4000可以根据通信标准(例如,无线通信协议)来调制编码后的视频数据,并且可以将调制后的视频数据发射到目的地装置5000。上述一个或多个通信媒体可以包含无线及/或有线通信媒体,例如射频(radio frequency,RF)频谱或一根或多根物理传输线。上述一个或多个通信媒体可以形成基于包的网络(例如,局域网、广域网或全球网络(例如,因特网))的部分。上述一个或多个通信媒体可以包含路由器、交换器、基站,或者实现从源装置4000到目的地装置5000的通信的其它设备。
在另一实例中,信道6000可包含存储由源装置4000产生的编码后的视频数据的存储媒体。在此实例中,目的地装置5000可经由磁盘存取或卡存取来存取存储媒体。存储媒体可包含多种本地存取式数据存储媒体,例如蓝光光盘、高密度数字视频光盘(digital video disc,DVD)、只读光盘(compact disc read-only memory,CD-ROM)、快闪存储器,或用于存储经编码视频数据的其它合适数字存储媒体。
在另一实例中,信道6000可包含文件服务器或存储由源装置4000产生的编码后的视频数据的另一中间存储装置。在此实例中,目的地装置5000可经由流式传输或下载来存取存储于文件服务器或其它中间存储装置处的编码后的视频数据。文件服务器可以是能够存储编码后的视频数据且将所述编码后的视频数据发射到目的地装置5000的服务器类型。例如,文件服务器可以包含全球广域网(world wide web,Web)服务器(例如,用于网站)、文件传送协议(file transfer protocol,FTP)服务器、网络附加存储(network attached storage,NAS)装置以及本地磁盘驱动器。
目的地装置5000可经由标准数据连接(例如,因特网连接)来存取编码后的视频数据。数据连接的实例类型包含适合于存取存储于文件服务器上的编码后的视频数据的无线信道、有线连接(例如,缆线调制解调器等),或两者的组合。编码后的视频数据从文件服务器的发射可为流式传输、下载传输或两者的组合。
本申请的获取运动矢量的方法不限于无线应用场景,示例性的,本申请的获取运动矢量的方法可以应用于支持以下应用等多种多媒体应用的视频编解码:空中电视广播、有线电视发射、卫星电视发射、流式传输视频发射(例如,经由因特网)、存储于数据存储媒体上的视频数据的编码、存储于数据存储媒体上的视频数据的解码,或其它应用。在一些实例中,视频编解码系统7000可经配置以支持单向或双向视频发射,以支持例如视频流式传输、视频播放、视频广播及/或视频电话等应用。
在图26中,源装置4000包含视频源4001、视频编码器4002及输出接口4003。在一些实例中,输出接口4003可包含调制器/解调器(调制解调器)及/或发射器。视频源4001可包含视频俘获装置(例如,视频相机)、含有先前俘获的视频数据的视频存档、用以从视频内容提供者接收视频数据的视频输入接口,及/或用于产生视频数据的计算机图形系统,或上述视频数据源的组合。
视频编码器4002可编码来自视频源4001的视频数据。在一些实例中,源装置4000 经由输出接口4003将编码后的视频数据直接发射到目的地装置5000。编码后的视频数据还可存储于存储媒体或文件服务器上以供目的地装置5000稍后存取以用于解码及/或播放。
在图26的实例中,目的地装置5000包含输入接口5003、视频解码器5002及显示装置5001。在一些实例中,输入接口5003包含接收器及/或调制解调器。输入接口5003可经由信道6000接收编码后的视频数据。显示装置5001可与目的地装置5000整合或可在目的地装置5000外部。一般来说,显示装置5001显示解码后的视频数据。显示装置5001可包括多种显示装置,例如液晶显示器、等离子体显示器、有机发光二极管显示器或其它类型的显示装置。
视频编码器4002及视频解码器5002可根据视频压缩标准(例如,高效率视频编解码H.265标准))而操作,并且可以可遵照高效视频编码(high efficiency video coding,HEVC)测试模型(HM)。H.265标准的文本描述ITU-TH.265(V3)(04/2015)于2015年4月29号发布,可从http://handle.itu.int/11.1002/7000/12455下载,所述文件的全部内容以引用的方式并入本文中。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统、装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中,应该理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(read-only memory,ROM)、随机存取存储器(random access memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以所述权利要求的保护范围为准。

Claims (49)

  1. 一种获取运动矢量的方法,其特征在于,包括:
    确定第一取值区间和第二取值区间;
    确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块,所述目标偏移矢量在第一方向上的分量的取值在所述第一取值区间内,所述目标偏移矢量在第二方向上的分量的取值在所述第二取值区间内,所述第一方向和所述第二方向呈正交关系;
    根据所述待处理子块的位置和所述目标偏移矢量,在所述目标图像中确定所述待处理子块的对应子块;
    根据所述对应子块的运动矢量获取所述待处理子块的运动矢量。
  2. 如权利要求1所述的方法,其特征在于,所述确定待处理块的目标偏移矢量,包括:
    确定所述待处理块的初始偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间内,且在所述第二方向上的分量的取值在第二取值区间内时,将所述初始偏移矢量确定为所述目标偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间内,且在所述第二方向上的分量的取值在第二取值区间外时,对所述初始偏移矢量在所述第二方向上的分量超出所述第二取值区间的部分进行钳位处理,将所述钳位处理后的初始偏移矢量确定为所述目标偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间外,且在所述第二方向上的分量的取值在第二取值区间内时,对所述初始偏移矢量在所述第一方向上的分量超出所述第一取值区间的部分进行钳位处理,将所述钳位处理后的初始偏移矢量确定为所述目标偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间外,且在所述第二方向上的分量的取值在第二取值区间外时,对所述初始偏移矢量在所述第一方向上的分量超出所述第一取值区间的部分进行钳位处理,对所述初始偏移矢量在所述第二方向上的分量超出所述第二取值区间的部分进行钳位处理,将所述钳位处理后的初始偏移矢量确定为所述目标偏移矢量。
  3. 如权利要求2所述的方法,其特征在于,所述确定所述待处理块的初始偏移矢量,包括:
    将所述待处理块的预设空域相邻块的一个运动矢量确定为所述初始偏移矢量。
  4. 如权利要求1所述的方法,其特征在于,所述确定待处理块的目标偏移矢量,包括:
    从所述待处理块的预设空域相邻块中确定目标空域相邻块,其中,所述目标空域相邻块的运动矢量在第一方向上的分量的取值在第一取值区间内,在第二方向上的分量的取值在第二取值区间内;
    将所述目标空域相邻块的运动矢量作为所述目标偏移矢量。
  5. 如权利要求1-4中任一项所述的方法,其特征在于,所述确定第一取值区间和所述第二取值区间,包括:
    根据所述待处理块的位置和尺寸、所述待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围,确定所述第一取值区间和所述第二取值区间。
  6. 如权利要求5所述的方法,其特征在于,所述CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在所述第一方向和所述第二方向上的扩展范围。
  7. 如权利要求5或6所述的方法,其特征在于,所述CTU的时域运动矢量场的扩展范围信息携带于序列参数集SPS、图像参数集PPS以及头字段中的至少一种。
  8. 如权利要求1-7中任一项所述的方法,其特征在于,所述根据所述对应子块的运动矢量获取所述待处理子块的运动矢量,包括:
    根据所述待处理子块所在的图像帧的图像顺序序号POC、所述待处理子块所在的图像帧的目标参考帧的POC、所述目标图像的POC以及所述目标图像的目标参考帧的POC,对所述对应子块的运动矢量进行缩放,以得到所述待处理子块的运动矢量。
  9. 根据权利要求8所述的方法,其特征在于,所述待处理子块的运动矢量,根据以下公式获得:
    Figure PCTCN2019090031-appb-100001
    其中,P1表示所述待处理子块所在的图像帧的POC,P2表示所述待处理子块所在的图像帧的目标参考帧的POC,P3表示所述目标图像的POC,P4表示所述目标图像的目标参考帧的POC,MV表示所述对应子块的运动矢量,MVs表示所述待处理子块的运动矢量。
  10. 一种获取运动矢量的方法,其特征在于,包括:
    确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块;
    根据所述待处理子块的位置和所述目标偏移矢量,确定出所述待处理子块在所述目标图像中的初始对应位置点;
    当所述初始对应位置点位于所述目标图像的预设区域内时,将所述初始对应位置点确定为目标对应位置点;
    当所述初始对应位置点超出所述预设区域时,将所述预设区域内距离所述初始对应位置点最近的位置确定为目标对应位置点;
    根据所述目标对应位置点对应的运动矢量获取所述待处理子块的运动矢量。
  11. 如权利要求10所述的方法,其特征在于,所述预设区域采用横坐标取值区间和纵坐标取值区间表示,所述横坐标取值区间包括横坐标最小值和横坐标最大值,所述纵坐标取值区间包括纵坐标最小值和纵坐标最大值,
    当所述初始对应位置点位于所述目标图像的预设区域内时,将所述初始对应位置点确定为目标对应位置点,当所述初始对应位置点超出所述预设区域时,将所述预设区域内距离所述初始对应位置点最近的位置确定为目标对应位置点,包括:
    根据以下公式确定所述目标对应位置点:
    x (i,j)=Clip3(HorMin,HorMax,x (i,j));
    y (i,j)=Clip3(VerMin,VerMax,y (i,j));
    其中,HorMin为所述横坐标最小值,HorMax为所述横坐标最大值,VerMin为所述纵坐标最小值,VerMax为所述纵坐标最大值,Clip3函数为钳位函数,Clip3函数的定义如下:
    Figure PCTCN2019090031-appb-100002
  12. 如权利要10或11所述的方法,其特征在于,所述根据所述待处理子块的位置和所述目标偏移矢量,确定出所述待处理子块在所述目标图像中的初始对应位置点,包括:
    根据所述目标偏移矢量对所述待处理子块的位置坐标进行偏移,得到所述待处理子块在所述目标图像中的初始对应位置点。
  13. 如权利要求12所述的方法,其特征在于,所述根据所述目标偏移矢量对所述待处理子块的位置坐标进行偏移,得到所述待处理子块在所述目标图像中的初始对应位置点,包括:
    据以下公式对所述待处理子块的位置坐标进行偏移,得到所述待处理子块在目标图像中的初始对应位置点的位置坐标;
    Figure PCTCN2019090031-appb-100003
    其中,(x,y)表示所述待处理块的位置坐标,i表示所述待处理块中第一方向上的第i个子块,j表示所述待处理块中第二方向上的第j个子块,(xoff,yoff)表示所述目标偏移运动矢量在所述第一方向和所述第二方向上的分量值,M和N分别为所述待处理子块的宽和高,(x (i,j),y (i,j))表示所述待处理块中的第(i,j)个子块的位置坐标。
  14. 如权利要求13所述的方法,其特征在于,(x,y)表示所述待处理块的左上顶点的位置坐标,(x (i,j),y (i,j))表示所述待处理块中的第(i,j)个子块的中心点或者左上顶点的位置坐标。
  15. 权利要求10-14中任一项所述的方法,其特征在于,所述预设区域是根据所述待处理块所在的编码树单元CTU的位置和尺寸、所述目标图像的尺寸以及预设的CTU的时域运动矢量场的扩展范围确定的。
  16. 如权利要求15所述的方法,其特征在于,所述CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在所述第一方向和所述第二方向上的扩展范围。
  17. 如权利要求15或16所述的方法,其特征在于,所述预设区域是根据以下公式计算得到的:
    HorMin=Min(CTUX–RangeL,0)
    HorMax=Max(CTUX+CTUW+RangeR–1,PicW–1)
    VerMin=Min(CTUY–RangeU,0)
    VerMax=Max(CTUY+CTUH+RangeB–1,PicH–1)
    其中,HorMin和HorMax分别为所述预设范围在第一方向上能够取得的最小值和最大值,VerMin和VerMax分别为所述预设范围在第二方向上能够取得的最小值和最大值,CTUX和CTUY表示所述待处理块所在CTU的位置坐标,CTUW和CTUH表示所述待处理块所在CTU的宽和高,RangeU、RangeB、RangeL、RangeR分别表示所述待处理块所在CTU的时域运动矢量场在上、下、左、右方向上的扩展值,PicW和PicH分别为所述目标图像的宽和高。
  18. 如权利要求15-17中任一项所述的方法,其特征在于,所述CTU的时域运动矢量场的扩展范围信息携带于序列参数集SPS、图像参数集PPS以及头字段中的至少一种。
  19. 如权利要求10-18中任一项所述的方法,其特征在于,所述根据所述目标对应位置点对应的运动矢量获取所述待处理子块的运动矢量,包括:
    根据所述待处理子块所在的图像帧的图像顺序序号POC、所述待处理子块所在的图像帧的目标参考帧的POC、所述目标图像的POC以及所述目标图像的目标参考帧的POC,对所述目标对应位置点对应的运动矢量进行缩放,以得到所述待处理子块的运动矢量。
  20. 根据权利要求19所述的方法,其特征在于,所述待处理子块的运动矢量,根据以下公式获得:
    Figure PCTCN2019090031-appb-100004
    其中,P1表示所述待处理子块所在的图像帧的POC,P2表示所述待处理子块所在的图像帧的目标参考帧的POC,P3表示所述目标图像的POC,P4表示所述目标图像的目标参考帧的POC,MV表示所述目标对应位置点对应的运动矢量,MVs表示所述待处理子块的运动矢量。
  21. 如权利要求10-20中任一项所述的方法,其特征在于,所述确定待处理块的目标偏移矢量,包括:
    将所述待处理块的空域相邻块的运动矢量确定为所述目标偏移矢量。
  22. 一种获取运动矢量的装置,其特征在于,包括:
    确定模块,用于确定第一取值区间和第二取值区间;
    所述确定模块还用于确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块,所述目标偏移矢量在第一方向上的分量的取值在所述第一取值区间内,所述目标偏移矢量在第二方向上的分量的取值在所述第二取值区间内,所述第一方向和所述第二方向呈正交关系;
    所述确定模块还用于根据所述待处理子块的位置和所述目标偏移矢量,在所述目标图像中确定所述待处理子块的对应子块;
    获取模块,用于根据所述对应子块的运动矢量获取所述待处理子块的运动矢量。
  23. 如权利要求22所述的装置,其特征在于,所述确定模块用于:
    确定所述待处理块的初始偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间内,且在所述第二方向上的分量的取值在第二取值区间内时,将所述初始偏移矢量确定为所述目标偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间内,且在所述第二方向上的分量的取值在第二取值区间外时,对所述初始偏移矢量在所述第二方向上的分量超出所述第二取值区间的部分进行钳位处理,将所述钳位处理后的初始偏移矢量确定为所述目标偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间外,且在所述第二方向上的分量的取值在第二取值区间内时,对所述初始偏移矢量在所述第一方向上的分量超出所述第一取值区间的部分进行钳位处理,将所述钳位处理后的初始偏移矢量确定为所述目标偏移矢量;
    当所述初始偏移矢量在所述第一方向上的分量的取值在所述第一取值区间外,且在所述第二方向上的分量的取值在第二取值区间外时,对所述初始偏移矢量在所述第一方向上的分量超出所述第一取值区间的部分进行钳位处理,对所述初始偏移矢量在所述第二方向上的分量超出所述第二取值区间的部分进行钳位处理,将所述钳位处理后的初始偏移矢量确定为所述目标偏移矢量。
  24. 如权利要求23所述的装置,其特征在于,所述确定模块用于:
    将所述待处理块的预设空域相邻块的一个运动矢量确定为所述初始偏移矢量。
  25. 如权利要求22所述的装置,其特征在于,所述确定模块用于:
    从所述待处理块的预设空域相邻块中确定目标空域相邻块,其中,所述目标空域相邻块的运动矢量在第一方向上的分量的取值在第一取值区间内,在第二方向上的分量的取值在第二取值区间内;
    将所述目标空域相邻块的运动矢量作为所述目标偏移矢量。
  26. 如权利要求22-25中任一项所述的装置,其特征在于,所述确定模块用于:
    根据所述待处理块的位置和尺寸、所述待处理块所在的编码树单元CTU的位置和尺寸以及预设的CTU的时域运动矢量场的扩展范围,确定所述第一取值区间和所述第二取值区间。
  27. 如权利要求26所述的装置,其特征在于,所述CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在所述第一方向和所述第二方向上的扩展范围。
  28. 如权利要求26或27所述的装置,其特征在于,所述CTU的时域运动矢量场的扩展范围信息携带于序列参数集SPS、图像参数集PPS以及头字段中的至少一种。
  29. 如权利要求22-28中任一项所述的装置,其特征在于,所述获取模块用于:
    根据所述待处理子块所在的图像帧的图像顺序序号POC、所述待处理子块所在的图像帧的目标参考帧的POC、所述目标图像的POC以及所述目标图像的目标参考帧的POC,对所述对应子块的运动矢量进行缩放,以得到所述待处理子块的运动矢量。
  30. 根据权利要求29所述的装置,其特征在于,所述待处理子块的运动矢量,根据以下公式获得:
    Figure PCTCN2019090031-appb-100005
    其中,P1表示所述待处理子块所在的图像帧的POC,P2表示所述待处理子块所在的图像帧的目标参考帧的POC,P3表示所述目标图像的POC,P4表示所述目标图像的目标参考帧的POC,MV表示所述对应子块的运动矢量,MVs表示所述待处理子块的运动矢量。
  31. 一种获取运动矢量的装置,其特征在于,包括:
    确定模块,用于确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块;
    所述确定模块还用于:
    根据所述待处理子块的位置和所述目标偏移矢量,确定出所述待处理子块在所述目标图像中的初始对应位置点;
    当所述初始对应位置点位于所述目标图像的预设区域内时,将所述初始对应位置点确定为目标对应位置点;
    当所述初始对应位置点超出所述预设区域时,将所述预设区域内距离所述初始对应位置点最近的位置确定为目标对应位置点;
    获取模块,用于根据所述目标对应位置点对应的运动矢量获取所述待处理子块的运动矢量。
  32. 如权利要求31所述的装置,其特征在于,所述预设区域采用横坐标取值区间和纵坐标取值区间表示,所述横坐标取值区间包括横坐标最小值和横坐标最大值,所述纵坐标取值区间包括纵坐标最小值和纵坐标最大值,所述确定模块用于:
    根据以下公式确定所述目标对应位置点:
    x (i,j)=Clip3(HorMin,HorMax,x (i,j));
    y (i,j)=Clip3(VerMin,VerMax,y (i,j));
    其中,HorMin为所述横坐标最小值,HorMax为所述横坐标最大值,VerMin为所述纵坐标最小值,VerMax为所述纵坐标最大值,Clip3函数为钳位函数,Clip3函数的定义如下:
    Figure PCTCN2019090031-appb-100006
  33. 如权利要31或32所述的装置,其特征在于,所述确定模块用于:
    根据所述目标偏移矢量对所述待处理子块的位置坐标进行偏移,得到所述待处理子块在所述目标图像中的初始对应位置点。
  34. 如权利要求33所述的装置,其特征在于,所述确定模块用于:
    据以下公式对所述待处理子块的位置坐标进行偏移,得到所述待处理子块在目标图像中的初始对应位置点的位置坐标;
    Figure PCTCN2019090031-appb-100007
    其中,(x,y)表示所述待处理块的位置坐标,i表示所述待处理块中第一方向上的第i 个子块,j表示待处理块中第二方向上的第j个子块,(xoff,yoff)表示所述目标偏移运动矢量在所述第一方向和所述第二方向上的分量值,M和N分别为所述待处理子块的宽和高,(x (i,j),y (i,j))表示所述待处理块中的第(i,j)个子块的位置坐标。
  35. 如权利要求34所述的装置,其特征在于,(x,y)表示所述待处理块的左上顶点的位置坐标,(x (i,j),y (i,j))表示所述待处理块中的第(i,j)个子块的中心点或者左上顶点的位置坐标。
  36. 权利要求31-35中任一项所述的装置,其特征在于,所述预设区域是根据所述待处理块所在的编码树单元CTU的位置和尺寸、所述目标图像的尺寸以及预设的CTU的时域运动矢量场的扩展范围确定的。
  37. 如权利要求36所述的装置,其特征在于,所述CTU的时域运动矢量场的扩展范围包括CTU的时域运动矢量场分别在所述第一方向和所述第二方向上的扩展范围。
  38. 如权利要求36或37所述的装置,其特征在于,所述预设区域是根据以下公式计算得到的:
    HorMin=Min(CTUX–RangeL,0)
    HorMax=Max(CTUX+CTUW+RangeR–1,PicW–1)
    VerMin=Min(CTUY–RangeU,0)
    VerMax=Max(CTUY+CTUH+RangeB–1,PicH–1)
    其中,HorMin和HorMax分别为所述预设范围在第一方向上能够取得的最小值和最大值,VerMin和VerMax分别为所述预设范围在第二方向上能够取得的最小值和最大值,CTUX和CTUY表示所述待处理块所在CTU的位置坐标,CTUW和CTUH表示所述待处理块所在CTU的宽和高,RangeU、RangeB、RangeL、RangeR分别表示所述待处理块所在CTU的时域运动矢量场在上、下、左、右方向上的扩展值,PicW和PicH分别为所述目标图像的宽和高。
  39. 如权利要求36-38中任一项所述的装置,其特征在于,所述CTU的时域运动矢量场的扩展范围信息携带于序列参数集SPS、图像参数集PPS以及头字段中的至少一种。
  40. 如权利要求31-39中任一项所述的装置,其特征在于,所述获取模块用于:
    根据所述待处理子块所在的图像帧的图像顺序序号POC、所述待处理子块所在的图像帧的目标参考帧的POC、所述目标图像的POC以及所述目标图像的目标参考帧的POC,对所述目标对应位置点对应的运动矢量进行缩放,以得到所述待处理子块的运动矢量。
  41. 根据权利要求40所述的装置,其特征在于,所述待处理子块的运动矢量,根据以下公式获得:
    Figure PCTCN2019090031-appb-100008
    其中,P1表示所述待处理子块所在的图像帧的POC,P2表示所述待处理子块所在的图像帧的目标参考帧的POC,P3表示所述目标图像的POC,P4表示所述目标图像的目标参考帧的POC,MV表示所述目标对应位置点对应的运动矢量,MVs表示所述待处理子块的运动矢量。
  42. 如权利要求31-41中任一项所述的装置,其特征在于,所述确定模块用于:
    将所述待处理块的空域相邻块的运动矢量确定为所述目标偏移矢量。
  43. 一种解码装置,其特征在于,包括:
    存储器,用于存储码流形式的视频数据;
    视频解码器,用于解码码流形式的视频数据,其中所述解码过程包括:
    确定第一取值区间和第二取值区间;
    确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块,所述目标偏移矢量在第一方向上的分量的取值在所述第一取值区间内,所述目标偏移矢量在第二方向上的分量的取值在所述第二取值区间内,所述第一方向和所述第二方向呈正交关系;
    根据所述待处理子块的位置和所述目标偏移矢量,在所述目标图像中确定所述待处理子块的对应子块;
    根据所述对应子块的运动矢量获取所述待处理子块的运动矢量。
  44. 一种编码装置,其特征在于,包括:
    存储器,用于存储视频数据,所述视频数据包括一个或多个图像块;
    视频编码器,用于编码所述视频数据,其中所述编码过程包括:
    确定第一取值区间和第二取值区间;
    确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块,所述目标偏移矢量在第一方向上的分量的取值在所述第一取值区间内,所述目标偏移矢量在第二方向上的分量的取值在所述第二取值区间内,所述第一方向和所述第二方向呈正交关系;
    根据所述待处理子块的位置和所述目标偏移矢量,在所述目标图像中确定所述待处理子块的对应子块;
    根据所述对应子块的运动矢量获取所述待处理子块的运动矢量。
  45. 一种解码装置,其特征在于,包括:
    存储器,用于存储码流形式的视频数据;
    视频解码器,用于解码码流形式的视频数据,其中所述解码过程包括:
    确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块;
    根据所述待处理子块的位置和所述目标偏移矢量,确定出所述待处理子块在所述目标图像中的初始对应位置点;
    当所述初始对应位置点位于所述目标图像的预设区域内时,将所述初始对应位置点确定为目标对应位置点;
    当所述初始对应位置点超出所述预设区域时,将所述预设区域内距离所述初始对应位置点最近的位置确定为目标对应位置点;
    根据所述目标对应位置点对应的运动矢量获取所述待处理子块的运动矢量。
  46. 一种编码装置,其特征在于,包括:
    存储器,用于存储视频数据,所述视频数据包括一个或多个图像块;
    视频编码器,用于编码所述视频数据,其中所述编码过程包括:
    确定待处理块的目标偏移矢量和目标图像的标识信息,其中,所述待处理块包含至少一个待处理子块;
    根据所述待处理子块的位置和所述目标偏移矢量,确定出所述待处理子块在所述目标图像中的初始对应位置点;
    当所述初始对应位置点位于所述目标图像的预设区域内时,将所述初始对应位置点确定为目标对应位置点;
    当所述初始对应位置点超出所述预设区域时,将所述预设区域内距离所述初始对应位置点最近的位置确定为目标对应位置点;
    根据所述目标对应位置点对应的运动矢量获取所述待处理子块的运动矢量。
  47. 一种解码装置,其特征在于,所述解码装置包括:存储器和处理器;
    其中,所述存储器用于存储程序代码;所述处理器用于调用所述程序代码,以执行如权利要求1至9或者10至21中任一项所述的方法。
  48. 一种编码装置,其特征在于,所述编码装置包括:存储器和处理器;
    其中,所述存储器用于存储程序代码;所述处理器用于调用所述程序代码,以执行如权利要求1至9或者10至21中任一项所述的方法。
  49. 一种计算机可读存储介质,其特征在于,包括程序代码,所述程序代码包括用于执行如权利要求1至9或者10至21中任一项所述的方法。
PCT/CN2019/090031 2018-06-04 2019-06-04 获取运动矢量的方法和装置 WO2019233423A1 (zh)

Priority Applications (14)

Application Number Priority Date Filing Date Title
KR1020207037727A KR102600071B1 (ko) 2018-06-04 2019-06-04 모션 벡터 획득 방법 및 장치
BR112020024666-2A BR112020024666A2 (pt) 2018-06-04 2019-06-04 método e aparelho de obtenção de vetor de movimento
KR1020237038069A KR20230156446A (ko) 2018-06-04 2019-06-04 모션 벡터 획득 방법 및 장치
MX2020013131A MX2020013131A (es) 2018-06-04 2019-06-04 Método de obtención de vector de movimiento y aparato.
FIEP19814727.4T FI3796657T3 (fi) 2018-06-04 2019-06-04 Liikevektorin hankintamenetelmä ja -laite
SG11202012078PA SG11202012078PA (en) 2018-06-04 2019-06-04 Motion vector obtaining method and apparatus
DK19814727.4T DK3796657T3 (da) 2018-06-04 2019-06-04 Fremgangsmåde og indretning til opnåelse af en bevægelsesvektor
JP2020567611A JP7177181B2 (ja) 2018-06-04 2019-06-04 動きベクトル取得方法および装置
EP19814727.4A EP3796657B1 (en) 2018-06-04 2019-06-04 Motion vector acquisition method and device
EP24150170.9A EP4369711A2 (en) 2018-06-04 2019-06-04 Motion vector obtaining method and apparatus
US17/111,376 US11463721B2 (en) 2018-06-04 2020-12-03 Motion vector obtaining method and apparatus
US17/898,044 US11812050B2 (en) 2018-06-04 2022-08-29 Motion vector obtaining method and apparatus
JP2022180472A JP2023017955A (ja) 2018-06-04 2022-11-10 動きベクトル取得方法および装置
US18/485,330 US20240121421A1 (en) 2018-06-04 2023-10-12 Motion vector obtaining method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810564560.8A CN110557641B (zh) 2018-06-04 2018-06-04 获取运动矢量的方法和装置
CN201810564560.8 2018-06-04

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/111,376 Continuation US11463721B2 (en) 2018-06-04 2020-12-03 Motion vector obtaining method and apparatus

Publications (1)

Publication Number Publication Date
WO2019233423A1 true WO2019233423A1 (zh) 2019-12-12

Family

ID=68736012

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/090031 WO2019233423A1 (zh) 2018-06-04 2019-06-04 获取运动矢量的方法和装置

Country Status (12)

Country Link
US (3) US11463721B2 (zh)
EP (2) EP3796657B1 (zh)
JP (2) JP7177181B2 (zh)
KR (2) KR20230156446A (zh)
CN (6) CN117956182A (zh)
BR (1) BR112020024666A2 (zh)
DK (1) DK3796657T3 (zh)
FI (1) FI3796657T3 (zh)
MX (1) MX2020013131A (zh)
PT (1) PT3796657T (zh)
SG (1) SG11202012078PA (zh)
WO (1) WO2019233423A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111050182B (zh) * 2019-12-27 2022-02-18 浙江大华技术股份有限公司 运动矢量的预测方法、视频编码方法及相关设备、装置
CN113706573B (zh) * 2020-05-08 2024-06-11 杭州海康威视数字技术股份有限公司 一种运动物体的检测方法、装置及存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160021386A1 (en) * 2014-06-17 2016-01-21 Imagination Technologies Limited Biasing Selection of Motion Estimation Vectors
CN105939475A (zh) * 2016-06-06 2016-09-14 中国矿业大学 一种高质量边信息生成方法
CN108076347A (zh) * 2016-11-15 2018-05-25 阿里巴巴集团控股有限公司 一种编码起始点的获取方法以及装置

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101291433A (zh) * 2007-04-19 2008-10-22 上海中科计算技术研究所 视频编码技术中的模块运动矢量匹配评估方法
KR20120088488A (ko) * 2011-01-31 2012-08-08 한국전자통신연구원 시간적 움직임 벡터 저장 방법 및 그 장치
CN103716629B (zh) * 2012-09-29 2017-02-22 华为技术有限公司 用于图像处理的方法、装置、编码器和解码器
JP2016526853A (ja) * 2013-07-12 2016-09-05 サムスン エレクトロニクス カンパニー リミテッド 深さ基盤ディスパリティベクトルを利用するインターレイヤビデオ復号化方法及びその装置、並びに深さ基盤ディスパリティベクトルを利用するインターレイヤビデオ符号化方法及びその装置
JP2015106747A (ja) * 2013-11-28 2015-06-08 富士通株式会社 動画像符号化装置、動画像符号化方法及び動画像符号化用コンピュータプログラム
JP2015173404A (ja) * 2014-03-12 2015-10-01 富士通株式会社 動画像符号化装置、動画像符号化方法及び動画像符号化用コンピュータプログラム
WO2016008157A1 (en) * 2014-07-18 2016-01-21 Mediatek Singapore Pte. Ltd. Methods for motion compensation using high order motion model
US11477477B2 (en) * 2015-01-26 2022-10-18 Qualcomm Incorporated Sub-prediction unit based advanced temporal motion vector prediction
CN104918057B (zh) * 2015-05-08 2018-07-13 上海交通大学 一种利用邻域运动信息的运动矢量后处理系统
US10200713B2 (en) * 2015-05-11 2019-02-05 Qualcomm Incorporated Search region determination for inter coding within a particular picture of video data
US10306258B2 (en) * 2016-01-29 2019-05-28 Google Llc Last frame motion vector partitioning
US10560718B2 (en) * 2016-05-13 2020-02-11 Qualcomm Incorporated Merge candidates for motion vector prediction for video coding
WO2018210315A1 (en) * 2017-05-18 2018-11-22 Mediatek Inc. Method and apparatus of motion vector constraint for video coding
US10602180B2 (en) * 2017-06-13 2020-03-24 Qualcomm Incorporated Motion vector prediction
WO2019004283A1 (ja) * 2017-06-28 2019-01-03 シャープ株式会社 動画像符号化装置及び動画像復号装置
CN117579828A (zh) * 2018-01-09 2024-02-20 夏普株式会社 运动矢量推导装置、运动图像解码装置以及运动图像编码装置
JP7260562B2 (ja) * 2018-01-29 2023-04-18 ヴィド スケール インコーポレイテッド 複雑性の低いフレームレートアップ変換
WO2019190224A1 (ko) * 2018-03-30 2019-10-03 한국전자통신연구원 영상 부호화/복호화 방법, 장치 및 비트스트림을 저장한 기록 매체
US11381834B2 (en) * 2018-04-02 2022-07-05 Hfi Innovation Inc. Video processing methods and apparatuses for sub-block motion compensation in video coding systems

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160021386A1 (en) * 2014-06-17 2016-01-21 Imagination Technologies Limited Biasing Selection of Motion Estimation Vectors
CN105939475A (zh) * 2016-06-06 2016-09-14 中国矿业大学 一种高质量边信息生成方法
CN108076347A (zh) * 2016-11-15 2018-05-25 阿里巴巴集团控股有限公司 一种编码起始点的获取方法以及装置

Also Published As

Publication number Publication date
JP7177181B2 (ja) 2022-11-22
SG11202012078PA (en) 2021-01-28
EP3796657A1 (en) 2021-03-24
EP3796657B1 (en) 2024-03-06
CN110557641A (zh) 2019-12-10
CN112040243B (zh) 2021-06-29
CN117956182A (zh) 2024-04-30
JP2023017955A (ja) 2023-02-07
BR112020024666A2 (pt) 2021-03-02
MX2020013131A (es) 2021-03-25
KR20210015931A (ko) 2021-02-10
FI3796657T3 (fi) 2024-05-20
KR102600071B1 (ko) 2023-11-07
US20210092433A1 (en) 2021-03-25
CN115550664A (zh) 2022-12-30
PT3796657T (pt) 2024-05-20
JP2021526338A (ja) 2021-09-30
EP4369711A2 (en) 2024-05-15
CN110557641B (zh) 2024-01-16
CN115550664B (zh) 2023-08-22
US11463721B2 (en) 2022-10-04
EP3796657A4 (en) 2021-06-23
US20230070282A1 (en) 2023-03-09
CN117956183A (zh) 2024-04-30
CN112040243A (zh) 2020-12-04
DK3796657T3 (da) 2024-05-21
US11812050B2 (en) 2023-11-07
US20240121421A1 (en) 2024-04-11
KR20230156446A (ko) 2023-11-14
CN118075486A (zh) 2024-05-24

Similar Documents

Publication Publication Date Title
RU2715010C1 (ru) Способ и устройство кодирования и декодирования изображения с использованием межкадрового прогнозирования
US20190158866A1 (en) Motion vector prediction method and apparatus for encoding or decoding video
JP2018050308A (ja) ビデオコーディングのための動きベクトル予測の実行
CN106998470B (zh) 解码方法、编码方法、解码设备和编码设备
AU2023200956B2 (en) Video data inter prediction method and apparatus
US11812050B2 (en) Motion vector obtaining method and apparatus
CN109996080B (zh) 图像的预测方法、装置及编解码器
US11412210B2 (en) Inter prediction method and apparatus for video coding
CN112640468A (zh) 应用于双向帧间预测中的参考帧获取方法及装置
KR102407912B1 (ko) 양방향 인트라 예측 시그널링
CN111050164B (zh) 一种编解码的方法和装置
CN110677645B (zh) 一种图像预测方法及装置
CN110971899B (zh) 一种确定运动信息的方法、帧间预测方法及装置
US20210185323A1 (en) Inter prediction method and apparatus, video encoder, and video decoder

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: 19814727

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020567611

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112020024666

Country of ref document: BR

ENP Entry into the national phase

Ref document number: 2019814727

Country of ref document: EP

Effective date: 20201215

ENP Entry into the national phase

Ref document number: 20207037727

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 112020024666

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20201202