WO2019137732A1 - Codage en mode intra utilisant des statistiques d'utilisation de mode pour générer une liste mpm - Google Patents

Codage en mode intra utilisant des statistiques d'utilisation de mode pour générer une liste mpm Download PDF

Info

Publication number
WO2019137732A1
WO2019137732A1 PCT/EP2018/084770 EP2018084770W WO2019137732A1 WO 2019137732 A1 WO2019137732 A1 WO 2019137732A1 EP 2018084770 W EP2018084770 W EP 2018084770W WO 2019137732 A1 WO2019137732 A1 WO 2019137732A1
Authority
WO
WIPO (PCT)
Prior art keywords
ipm
ipms
priority
modes
video sequence
Prior art date
Application number
PCT/EP2018/084770
Other languages
English (en)
Inventor
Zhi Zhang
Ruoyang YU
Kenneth Andersson
Per Wennersten
Rickard Sjöberg
Original Assignee
Telefonaktiebolaget Lm Ericsson (Publ)
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Telefonaktiebolaget Lm Ericsson (Publ) filed Critical Telefonaktiebolaget Lm Ericsson (Publ)
Publication of WO2019137732A1 publication Critical patent/WO2019137732A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/103—Selection of coding mode or of prediction mode
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/103—Selection of coding mode or of prediction mode
    • H04N19/11—Selection of coding mode or of prediction mode among a plurality of spatial predictive coding modes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/119—Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/46—Embedding additional information in the video signal during the compression process
    • H04N19/463—Embedding additional information in the video signal during the compression process by compressing encoding parameters before transmission
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/593—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques

Definitions

  • a video sequence consists of a series of pictures, and each picture consists of one or more components (e.g., three components). Each component can be described as a two- dimensional rectangular array of sample values. It is common that a picture in a video sequence consists of three components: one luma component (Y) where the sample values are luma values and two chroma components (Cb) and (Cr), where the sample values are chroma values. It is common that the dimensions of the chroma components are smaller than the luma components by a factor of two in each dimension. For example, the size of the luma component of a high-definition (HD) picture would be 1920x1080 and the chroma components would each have the size 960x540. Components are sometimes referred to as color
  • a block is one two-dimensional array of samples.
  • each component e.g., Y, Cb, and Cr
  • the coded video bitstream is a series of blocks.
  • CU High Efficiency Video Coding
  • CTU Coding Tree Unit
  • CUs are squares, i.e. have a size of NxN luma samples, where N can have a value of 8, 16, 32 and 64.
  • CUs can have other sizes and additionally be rectangular, i.e. have a size of NxM luma samples where N is different to M.
  • sample predictions There are two types of sample predictions: intra prediction and inter prediction.
  • Intra prediction predicts blocks or unit based on spatial extrapolation of samples from previously decoded blocks or units of the same (current) picture. Inter prediction predicts blocks or units by using samples from previous pictures.
  • Intra directional prediction is utilized in HE VC and JEM.
  • HE VC there are 33 directional modes and 35 modes in total.
  • JEM there are 65 directional modes and 67 modes in total.
  • the modes are numbered using indices and mode index 0 represents the planar prediction mode.
  • Mode index 1 is used for the DC prediction mode.
  • the angular prediction mode indices range from 2 to 34 for HEVC and from 2 to 66 for JEM.
  • the 67 intra prediction modes are subdivided into the most probable modes (MPM) set, the selected modes set, and the non-selected modes set.
  • the MPM set contains 6 modes, the selected modes set contains 16 modes, and the non-selected modes set contains 45 modes.
  • the 6 modes in the MPM set are first populated by the intra modes of 5 neighboring blocks or units as illustrated in FIG. 6 plus the planar and DC modes.
  • the MPM set is populated in the order of left (L), above (A), planar, DC, below left (BL), above right (AR) and above left (AL). If the MPM set is not fully populated, for instance because some neighboring blocks or units are not available for prediction, derived modes are added. These added modes are derived by adding +1 or -1 to the angular modes that are already in MPM set. If the MPM set is still not fully populated, the default modes vertical, horizontal, diagonal from below left and diagonal from top left are added until the MPM is full. If a mode is already in the MPM set, that mode is not added again, i.e. modes can be included once into the MPM set.
  • the selected modes set is generated by sub-sampling the 61 non-MPM modes using an interval of 4. Only unique modes can be included in selected modes set.
  • the non-selected modes set includes 45 modes which are not included in MPM set and selected modes set. The order of the 45 modes are from the smallest index to the largest index.
  • the modes in the selected modes set is coded with a 4-bit fixed length code.
  • the modes in the non-selected modes set is coded with a truncated binary code.
  • the intra mode code words used for signaling are shown in Table 1 :
  • MPM intra prediction modes As the coding probability is equally spread from horizontal intra prediction modes to vertical intra prediction modes, the spatial correlation is not utilized.
  • the non-selected modes set contains 45 modes. These 45 modes are arranged in order from small to large mode index.
  • the horizontal direction indices range from 2 to 34 and the vertical direction indices ranges from 34 to 66.
  • the first 19 modes are coded with 7 binaries and the rest are coded with 8 binaries. This means that if the mode to use is a vertical intra prediction mode from non- selected modes set, it is more likely coded with 8 bits. When the intra prediction mode is likely to be vertical, it is not efficient to code the intra modes in the non-selected modes set.
  • the historical intra prediction mode (IPM) information is used to generate the MPM set, selected modes set (SM set), and non-selected modes set (N-SM set). For example, based on the historical IPM information, the most frequently occurring IPMs can be determined and one or more of these may become part of the MPM set, which requires the fewest coding bits. The less frequently occurring intra prediction modes will be placed in the non-selected modes set, which require more coding bits.
  • An advantage the embodiments disclosed herein is that they improve the coding efficiency of coding the IPMs. This is done by, for example, improving the generation of the MPM set, the selected modes set, and the non-selected modes set.
  • a method for encoding or decoding a video sequence is provided.
  • the method includes for each intra prediction mode (IPM) included in a certain set of IPMs, assigning a priority value to the IPM.
  • the method also includes, deriving an intra prediction characteristic based on the priority values assigned to the IPMs, and using the derived intra prediction characteristic in a video sequence encoding process and/or a video sequence decoding process.
  • the derived intra prediction characteristic is represented by an ordered sequence of the IPMs ordered in accordance with the assigned priority values.
  • the priority values assigned to the IPMs are derived from statistics regarding the usage of IPMs that were used to encode blocks or units within an area of a picture of the video sequence.
  • the priority value is equal to, or based on, the number of determined occurrences of the IPM in an area of a picture of the video sequence.
  • the method also includes selecting an IPM to encode or decode a block or unit of an area of a picture of the video sequence; and after selecting the IPM to encode or decode the block or unit: assigning a priority value to the selected IPM; and adding the IPM to the ordered sequence of IPMs based on the priority value assigned to the selected IPM.
  • the ordered sequence of IPMs includes a particular IPM
  • the method further comprises: selecting the particular IPM to encode or decode an intra block or unit within an area of a picture of the video sequence, and after selecting the particular IPM to encode or decode the intra block or unit, modifying (e.g., increasing by a value of one) the priority value assigned to the particular IPM.
  • the method also includes, after modifying the priority value assigned to the particular IPM, updating the ordered sequence of the IPMs.
  • the set of IPMs includes a particular IPM
  • the method further comprises determining that the particular IPM has been applied on a particular intra block or unit, and the priority value assigned to the particular IPM is a function of the size of the particular intra block or unit.
  • the method further includes one of: 1) a video encoder transmitting information for identifying an ordinal position within the ordered sequence, thereby identifying the IPM that is positioned at the identified ordinal position, and 2) a video decoder receiving information for identifying an ordinal position within the ordered sequence, thereby identifying the IPM that is positioned at the identified ordinal position.
  • the method further includes: determining a set of most probable modes (MPMs), the set of MPMs comprising M IPMs; determining that M is less than T, wherein T is a predetermined value; and selecting a first particular one of the IPMs included in the set of IPMs based at least in part on the position of the first particular IPM within the ordered sequence of IPMs; and adding the selected first particular IPM to the set of MPMs.
  • MPMs most probable modes
  • the method also includes: incrementing M by a value of 1 ; determining whether M is less than T; and as a result of determining that M is less than T, selecting a second particular one of the IPMs from the set of IPMs based at least in part on the position of said second particular IPM within the ordered sequence of IPMs and adding the selected second particular IPM to the set of MPMs.
  • selecting the first particular IPM from the set of IPMs based at least in part on the position of the first particular IPM within the ordered sequence of IPMs comprises: based on the priority values assigned to the IPMs, identifying an IPM having the highest priority; and after identifying the IPM, selecting the identified IPM, wherein the identified IPM is the first particular IPM. In some embodiments, selecting the first particular IPM from the set of IPMs based at least in part on the position of the first particular IPM within the ordered sequence of IPMs further comprises determining that the identified IPM is not already included in the set of MPMs.
  • the method also includes after adding the first particular
  • IPM to the set of MPMs, creating a selected set of IPMs based on (e.g., based at least in part on) the IPMs included in the set of MPMs.
  • the method also includes after creating the selected set of
  • IPMs creating a non-selected set of IPMs based on (e.g., based at least in part on) the IPMs included in the set of MPMs and the IPMs included in the selected set of IPMs.
  • an encoding or decoding apparatus wherein the apparatus is adapted to: 1) for each intra prediction mode (IPM) included in a certain set of IPMs, assign a priority value to the IPM, 2) create an ordered sequence of the IPMs included in said set in accordance with the assigned priority values, and 3) use the ordered sequence of the IPMs in one of a video sequence encoding process and a video sequence decoding process.
  • IPM intra prediction mode
  • FIG. 1 illustrates an example priority list.
  • FIG. 2 is a flow chart illustrating a process according to some embodiments.
  • FIG. 3 illustrate the updating of a priority list.
  • FIG. 4 is a block diagram of an encoding/decoding apparatus according to some embodiments.
  • FIG. 5 is a diagram showing functional units of an encoding/decoding apparatus according to some embodiments.
  • FIG. 6 illustrates neighboring blocks.
  • a video encoding/decoding process identifies an intra prediction pattern in an area inside of a picture of a video sequence.
  • the intra prediction pattern is used to generate an MPM set, a selected modes set, and a non- selected modes set when coding an intra block or unit.
  • the intra prediction pattern could be represented by a priority list of intra prediction modes (IPMs).
  • IPMs intra prediction modes
  • the list could contain one single prediction mode or several prediction modes.
  • the method can be used on both the encoder side and the decoder side.
  • the intra prediction pattern is represented by a certain set of IPMs, such as, for example, a priority list of IPMs.
  • the priority list of IPMs is derived by collecting the statistics of the IPMs from already coded intra blocks or units in the area of the picture.
  • the priority list contains a list of paired elements (e.g., a list of“2 -tuples”) where each 2-tuple (or“tuple” for short) consists of one index value representing an IPM and a priority value assigned to the IPM. An IPM with higher priority may be assigned with a higher priority value.
  • the priority list Before the start of coding or decoding the area of the picture, the priority list can be initialized using a default priority list. The list is then updated every time after an intra block or unit inside the area is coded or decoded. For example, the list can be updated by adding to the list a new tuple (for an IPM not already included in the list), or the priority list can be updated by updating the priority value for an IPM already included in the priority list.
  • One example of updating the priority list is to first check if the priority list already contains a tuple with the same IPM as the one just coded or decoded. If the priority list already contains a tuple with the same IPM as the one just coded or decoded, then the priority value of that tuple is updated, otherwise a new tuple is added to the priority list.
  • the approach may use the priority list to generate the MPM set, selected modes set (SM set), and non- selected modes set (N-SM set).
  • SM set selected modes set
  • N-SM set non- selected modes set
  • the first step is similar to the current JEM method where the MPM set is derived from five neighboring intra modes plus the planar and DC modes as described earlier. If the MPM set is not fully populated, IPMs from the priority list that are not included in the MPM set are added into the MPM set in priority order until either the MPM set is full or until there are no more IPMs to be added from the priority list.
  • One example of using the priority list to generate the N-SM set is as follows: the priority list is scanned in priority order, and if an IPM from the list is not present in the MPM set or SM set, then the IPM is added to the N-SM set.
  • one way to generate the SM set is to include in the SM set the angular modes closest to angular modes in the MPM set.
  • An alternative is to not use the MPM, SM and N-SM sets, but rather order (at least a subset of) the elements in the priority list according to or based on their priority values such that IPMs with higher priority are represented with fewer bits, fewer binaries or lower entropy than modes with lower priority.
  • the encoding method can be seen as a way of mapping the IPMs to a fixed set of binary representations and the decoding method can be seen as a way of mapping a decoded value to an IPM.
  • the mapping is updated based (at least in part and for at least some of the modes) on the priority values.
  • Step 1 a priority list is initialized, where each tuple in the priority list consists of one value representing an IPM (i.e., an IPM index) and a priority value assigned to that IPM.
  • IPM an IPM index
  • Step 2 a coded intra mode value is decoded for a particular block or unit of a current picture. It is preferable that the decoding of intra mode values is not dependent on the priority list.
  • Step 3 An IPM is then is then derived from the decoded intra mode value by means of a mapping using the priority list. For example, if the decoded intra mode value is “7,” then the IPM identified by the IPM index stored in the 7 th entry of the priority list is the derived IPM.
  • Step 4 The priority list is updated by one or both of the following operations
  • the order of the paired elements in the priority list is modified and/orb) the priority value of the paired element in the priority list corresponding to the actual intra mode is modified, and optionally the priority values of other elements in the priority list are also updated.
  • the priority value assigned to an IPM in the priority list is a hit count (e.g., a number of occurrences) of the IPM.
  • the priority list may be referred to as a“count-mode buffer.”
  • An IPM with higher hit count means that is has higher priority and should be coded using fewer bits, fewer binaries or lower entropy.
  • the area of the picture is a CTU, and the priority list is then CTU based, which means that the priority list (count-mode buffer) is initialized at the start of encoding or compressing or decoding a CTU.
  • the priority list is updated every time when an IPM is added (encoded or decoded) for a CU.
  • the priority list is sorted according to the tuple element“count”. A large count value may be located at the bottom of priority list. If two or more modes have the same count, the last added mode may be located at the bottom of the priority list.
  • An example priority list is shown in FIG. 1. An alternative is to sort the priority list such that a large count value is located at the top of the priority list.
  • the first step for generating the MPM set is similar to the current JEM method: first the MPM set is populated by the IPMs left, above, planar, DC, below left, above right, and above left. If the MPM set is not complete, the priority list is scanned (e.g., from bottom to top when the priority list is sorted in reverse priority value order) and IPMs are inserted into the MPM set until it is completed. Only unique IPMs are inserted into the MPM set.
  • Table 1 below includes pseudo-code for populating the MPM set when the
  • the angular modes closest to angular modes in MPM set are added to the SM set.
  • offsets ⁇ 1, ⁇ 2, ⁇ 3 to the angular modes already in the MPM set are used to derive the 16 selected modes, only unique modes can be included into the SM set.
  • N-SM non-selected modes
  • the count-mode buffer is scanned from bottom to top (assuming the buffer is ordered in reverse priority order) and IPMs which do not exist in MPM set nor in the SM set are included to populate the N-SM set. Only unique modes are included into the N-SM set. The least frequently occurring IPMs will be put at the end of the N-SM set. Table 2 below includes pseudo-code to generate the SM set and N- SM set.
  • the count-mode buffer consists of a certain set of tuples (i.e., value pairs).
  • the first value of the value pair is a count
  • the second value of the value pair is an IPM index for identifying an IPM.
  • IPM index 43 has a has a count of 2
  • IPM index 17 has a count of 55.
  • an incomplete MPM set includes IPMs 0, 29, and 1
  • the complete MPM set that is generated using the count-mode buffer shown above includes the following six modes: 0, 29, 1, 30, 17, 28 (the modes 30, 17 and 28 are derived from count mode buffer).
  • the 16 modes in selected modes set are: 31 , 16, 18, 27, 32, 15, 19, 26, 33, 14, 20, 25, 34, 13, 21 , 24.
  • the 45 modes in non-selected modes set are: 7, 23, 37,
  • the priority value (at least partly) represents a size of the coded area on which the IPM is applied on.
  • the size of the coded area can be calculated as the sum of sizes of all the blocks or units that have the same IPM.
  • a decaying factor is applied to all the priority values in the list.
  • the default priority list is signaled in a parameter set, sequence parameter set, picture parameter set, or similar from an encoder to a decoder.
  • the default priority list could either be signaled explicitly or be signaled by an index to a pre defined priority lists set.
  • the default priority list could be derived by an encoder using offline statistics for deciding the priorities of the IPMs.
  • One alternative is to signal the default priority list on a picture or slice basis.
  • the default list can be different for an intra picture or an inter picture or can be dependent on the picture hierarchy level or the slice quantization parameter (QP). For these cases, multiple default lists may be signaled in e.g. a parameter set or slice header.
  • the default priority list is based on the priority lists from spatially neighboring areas inside a picture. For example, if the priority list is based on a CTU level, then the initialization of the priority list of the current CTU could be based on the priority list of a spatial neighboring CTU.
  • the neighboring CTU could be left CTU, above CTU, top left CTU and top right CTU.
  • the neighboring CTU which is not yet coded will not be used to initialize the current CTU.
  • the priority list of neighboring CTU can be set to the last updated priority list in that CTU.
  • the priority list to use could be constructed by dividing the priority values by a constant or dividing them with a value such that the sum of the priority values is equal to or smaller than a constant.
  • the default priority list could be generated using a combination of the above two described embodiments.
  • One example is to fetch N pair elements of the priority list from the neighboring CTU or CU and assign higher priority values to those N pair elements.
  • Other elements are fetched from the priority list from offline statistics, preferably assigned with lower priority values.
  • the number of considered neighboring blocks or units used to generate the MPM set is reduced compared to the prior art, for example 2 instead of 5. The idea is to make room for including more IPMs from the priority list. Another alternative is to derive all 6 MPM modes from the priority list, which does not need to check the IPMs of neighboring bocks.
  • the SM set is derived from the priority list.
  • One approach is to go through the priority list in priority order. If a mode is not used in MPM set, it is inserted into the SM set until the SM set is full.
  • IPMs there could be more than one priority list of IPMs to represent the intra prediction pattern.
  • NxN square CUs NxM rectangular CUs
  • MxN rectangular CUs where N>M
  • the approach is applied to luma component Y, chroma component Cb and Cr.
  • the priority list could be based on the CTU level.
  • the priority list could alternatively be based on the picture level or slice level.
  • FIG. 2 is a flow chart illustrating a process 200, according to some embodiments, for coding or decoding a video sequence.
  • Process 200 may begin with step s202 where, for each IPM included in a certain set of one or more IPMs, a priority value is assigned to the IPM.
  • step s204 an intra prediction characteristic based on the priority values assigned to the IPMs is derived.
  • step s206 the derived intra prediction characteristic is used in: i) a video sequence encoding process and/or ii) a video sequence decoding process.
  • the derived intra prediction characteristic is represented by an ordered sequence of the IPMs ordered in accordance with the assigned priority values (e.g., in some embodiments, deriving the intra prediction characteristic comprises creating the ordered sequence of IPMs).
  • the derived intra prediction characteristic is used to determine which IPMs should be used more often. That is, in one embodiment, the ordered sequence of IPMs is used to determine which IPMs should be used more often.
  • the priority values assigned to the IPMs in step s202 are derived from statistics regarding the IPMs that were used to encode blocks or units within an area of a picture of the video sequence. For example, as described above with respect to the count-mode buffer embodiment, the priority value is equal to, or based (e.g., a function of), the number of determined occurrences of the IPM in an area of a picture of the video sequence.
  • the video decoding process further includes selecting an
  • IPM to encode or decode a block or unit of an area of a picture of the video sequence, and, after selecting the IPM to encode or decode the block or unit: assigning a priority value to the selected IPM and adding the IPM to the ordered sequence of IPMs based on the priority value assigned to the selected IPM.
  • the ordered sequence of IPMs includes a particular IPM
  • the video encoding/decoding process further includes: selecting the particular IPM to encode or decode an intra block or unit within an area of a picture of the video sequence, and, after selecting the particular IPM to encode or decode the intra block or unit, modifying (e.g., increasing by a value of one) the priority value assigned to the particular IPM.
  • the ordered sequence of the IPMs is updated so that the IPMs maintain their priority value sequence. This feature is illustrated in FIG. 3. As shown in FIG.
  • a priority list 302 has been created, which in this case is a count-mode buffer and which includes an ordered sequence of IPMs (i.e., an ordered sequence of IPM index values). As further shown, the priority list 302 is updated, thereby creating a new priority list 304, so that the IPMs maintain their priority value sequence when the priority value of an IPM changes (in this example the priority value for IPM 63 has been increased from 1 to 2, which caused IPM 63 to move from the second position in the list to the fourth position in the list).
  • the set of IPMs includes a particular IPM
  • the process further comprises determining that the particular IPM has been applied on a particular intra block or unit, and the priority value assigned to the particular IPM is a function of the size of the particular intra block or unit.
  • the process also includes, a video encoder (e.g., apparatus
  • an encoder may send to the decoder a value of indicating the position of IPM 63 in the current priority list. For instance, assuming that priority list 304 is the current priority list, then to identify IPM 63 to the decoder the encoder may transmit a value of 4 because IPM 63 is in the fourth position of the list. Alternatively, the encoder may transmit a value of 3 since positions may be counted from 0 (or some other value to indicate the position of IPM 63 in priority list 304).
  • M may be incremented by 1 and then after step (5) steps 2-5 may be repeated so that the MPM set contains a total of T unique IPMs.
  • the encoder can transmit an MPM flag to indicate that the particular IPM is included in the MPM set and an MPM index value (e.g., a value between and including 1 to T or a value between and including 1 to T-l (assuming the MPM set contains a total of T IPMs)) to identify the particular IPM’s “position” within the MPM set.
  • an MPM index value e.g., a value between and including 1 to T or a value between and including 1 to T-l (assuming the MPM set contains a total of T IPMs)
  • selecting an IPM from the set of IPMs based at least in part on the position of the IPM within the ordered sequence of IPMs comprises: identifying an IPM having the highest priority based on the priority values assigned to the IPMs and, after identifying the IPM, selecting the identified IPM. In some embodiments, selecting an IPM from the set of IPMs based at least in part on the position of the IPM within the ordered sequence of IPMs further comprises determining that the identified IPM is not already included in the set of MPMs.
  • the process also includes, after adding the selected IPM to the set of MPMs, creating a selected set of IPMs based on (e.g. based at least in part on) the IPMs included in the set of MPMs (e.g., in the manner described above). In some embodiments, the process also includes, after creating the selected set of IPMs, creating a non-selected set of IPMs based on (e.g. based at least in part on) the IPMs included in the set of MPMs and the IPMs included in the selected set of IPMs (e.g., in the manner described above).
  • FIG. 4 is a block diagram of an apparatus 400 (e.g., an encoding apparatus or a decoding apparatus) according to some embodiments.
  • the apparatus may comprise: a data processing apparatus (DP A) 402, which may include one or more processors (P) 455 (e.g., a general purpose microprocessor and/or one or more other processors, such as an application specific integrated circuit (ASIC), field-programmable gate arrays (FPGAs), and the like); a network interface 448 comprising a transmitter (Tx) 445 and a receiver (Rx) 447 for enabling the apparatus to transmit data to and receive data from other nodes connected to a network 110 (e.g., an Internet Protocol (IP) network) to which network interface 448 is connected; and local storage unit (a.k.a.,“data storage system”) 408, which may include one or more non-volatile storage devices and/or one or more volatile storage devices (e.g., random access memory (RAM)).
  • DP A data processing apparatus
  • CPP 441 may be provided.
  • CPP 441 includes a computer readable medium (CRM) 442 storing a computer program (CP) 443 comprising computer readable instructions (CRI) 444.
  • CRM 442 may be a non-transitory computer readable medium, such as, but not limited, to magnetic media (e.g., a hard disk), optical media, memory devices (e.g., random access memory, flash memory), and the like.
  • the CRI 444 of computer program 443 is configured such that when executed by data processing apparatus 402, the CRI causes the apparatus to perform steps described herein (e.g., steps described herein with reference to the flow charts and/or message flow diagrams).
  • the apparatus may be configured to perform steps described herein without the need for code. That is, for example, DPA 402 may consist merely of one or more ASICs.
  • the features of the embodiments described herein may be implemented in hardware and/or software.
  • FIG. 5 is a diagram showing functional units of the apparatus 400 according to some embodiments.
  • the apparatus 400 may include: a value assigning unit 502 configured to assign a value to each IPM include in a certain set of one or more IPMs;
  • deriving unit 504 configured to derive an intra prediction characteristic based on the priority values assigned to the IPMs ; and an encoding or decoding unit 506 configured to use the derived intra prediction characteristic in a video sequence encoding process and/or a video sequence decoding process.

Landscapes

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

Abstract

La présente invention concerne un procédé de codage et de décodage d'une séquence vidéo. Dans certains modes de réalisation, le procédé consiste, pour chaque mode de prédiction intra (IPM) présent dans un ensemble désigné d'IPM, à attribuer une valeur de priorité à l'IPM. Le procédé consiste également à dériver une caractéristique de prédiction intra sur la base des valeurs de priorité attribuées aux IPM, et à utiliser la caractéristique de prédiction intra dérivée dans un processus de codage de séquence vidéo et/ou dans un processus de décodage de séquence vidéo.
PCT/EP2018/084770 2018-01-12 2018-12-13 Codage en mode intra utilisant des statistiques d'utilisation de mode pour générer une liste mpm WO2019137732A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201862616626P 2018-01-12 2018-01-12
US62/616626 2018-01-12

Publications (1)

Publication Number Publication Date
WO2019137732A1 true WO2019137732A1 (fr) 2019-07-18

Family

ID=64899277

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2018/084770 WO2019137732A1 (fr) 2018-01-12 2018-12-13 Codage en mode intra utilisant des statistiques d'utilisation de mode pour générer une liste mpm

Country Status (1)

Country Link
WO (1) WO2019137732A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021032113A1 (fr) * 2019-08-19 2021-02-25 Beijing Bytedance Network Technology Co., Ltd. Mise à jour de mode de prédiction intra sur la base d'un compteur
WO2023051532A1 (fr) * 2021-09-29 2023-04-06 Beijing Bytedance Network Technology Co., Ltd. Procédé, dispositif et support de traitement vidéo

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20130079128A (ko) * 2011-12-26 2013-07-10 연세대학교 산학협력단 인트라 모드 부호화 방법 및 장치
US20140226912A1 (en) * 2011-09-28 2014-08-14 Electronics And Telecommunications Research Institute Method and apparatus for encoding/decoding intra prediction mode
WO2017086746A1 (fr) * 2015-11-19 2017-05-26 한국전자통신연구원 Procédé et appareil d'encodage/décodage d'un mode de prédiction intra-écran.

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140226912A1 (en) * 2011-09-28 2014-08-14 Electronics And Telecommunications Research Institute Method and apparatus for encoding/decoding intra prediction mode
KR20130079128A (ko) * 2011-12-26 2013-07-10 연세대학교 산학협력단 인트라 모드 부호화 방법 및 장치
WO2017086746A1 (fr) * 2015-11-19 2017-05-26 한국전자통신연구원 Procédé et appareil d'encodage/décodage d'un mode de prédiction intra-écran.

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
ALSHIN A ET AL: "Description of SDR, HDR and 360° video coding technology proposal by Samsung, Huawei, GoPro, and HiSilicon "" mobile application scenario", 10. JVET MEETING; 10-4-2018 - 20-4-2018; SAN DIEGO; (THE JOINT VIDEO EXPLORATION TEAM OF ISO/IEC JTC1/SC29/WG11 AND ITU-T SG.16 ); URL: HTTP://PHENIX.INT-EVRY.FR/JVET/,, no. JVET-J0024, 3 April 2018 (2018-04-03), XP030151189 *
JIANLE CHEN ET AL: "Further improvements to HMKTA-1.0", 52. VCEG MEETING; 19-6-2015 - 26-6-2015; WARSAW; (VIDEO CODING EXPERTS GROUP OF ITU-T SG.16),, no. VCEG-AZ07, 19 June 2015 (2015-06-19), XP030003885 *
NON-EE1: PRIORITY LIST BASED INTRA MODE CODING WITH 5 MPM: "Non-EE1: Priority List Based Intra Mode Coding with 5 MPM", 8. JVET MEETING; 18-10-2017 - 25-10-2017; MACAU; (THE JOINT VIDEO EXPLORATION TEAM OF ISO/IEC JTC1/SC29/WG11 AND ITU-T SG.16 ); URL: HTTP://PHENIX.INT-EVRY.FR/JVET/,, no. JVET-H0051-v2, 16 October 2017 (2017-10-16), XP030151042 *
SARWER M G ET AL: "CE3-related: Intra mode coding", no. JVET-K0243, 11 July 2018 (2018-07-11), XP030199431, Retrieved from the Internet <URL:http://phenix.int-evry.fr/jvet/doc_end_user/documents/11_Ljubljana/wg11/JVET-K0243-v2.zip JVET-K0243-v2.docx> [retrieved on 20180711] *
SEREGIN V ET AL: "Neighbor based intra most probable modes list derivation", 3. JVET MEETING; 26-5-2016 - 1-6-2016; GENEVA; (THE JOINT VIDEO EXPLORATION TEAM OF ISO/IEC JTC1/SC29/WG11 AND ITU-T SG.16 ); URL: HTTP://PHENIX.INT-EVRY.FR/JVET/,, no. JVET-C0055, 17 May 2016 (2016-05-17), XP030150153 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021032113A1 (fr) * 2019-08-19 2021-02-25 Beijing Bytedance Network Technology Co., Ltd. Mise à jour de mode de prédiction intra sur la base d'un compteur
US20220182667A1 (en) * 2019-08-19 2022-06-09 Beijing Bytedance Network Technology Co., Ltd. Updating for counter-based intra prediction mode
US11917196B2 (en) 2019-08-19 2024-02-27 Beijing Bytedance Network Technology Co., Ltd Initialization for counter-based intra prediction mode
WO2023051532A1 (fr) * 2021-09-29 2023-04-06 Beijing Bytedance Network Technology Co., Ltd. Procédé, dispositif et support de traitement vidéo

Similar Documents

Publication Publication Date Title
US11800117B2 (en) Enhanced intra prediction mode signaling
CN112438050B (zh) 使用二次变换的视频信号处理方法及装置
US11882284B2 (en) Determination of set of candidate transforms for video encoding
JP6598957B2 (ja) サンプル適応オフセットの処理を実行するための情報を符号化する符号化方法、および処理方法
JP7170669B2 (ja) 最後の有意係数フラグのコーディング
US20210099720A1 (en) Refined entropy coding for level maps
JP5180370B2 (ja) 並べ替えられた変換係数のマルチレベル表現
WO2015120818A1 (fr) Méthodes et dispositifs de codage et de décodage d&#39;image
US20220295052A1 (en) Video encoding method and apparatus, and video decoding method and apparatus
EP3732882A1 (fr) Conception de calcul de modèle contextuel efficace dans un codage de coefficient de transformée
AU2020360013A1 (en) Transform-based image coding method, and apparatus therefor
WO2019137732A1 (fr) Codage en mode intra utilisant des statistiques d&#39;utilisation de mode pour générer une liste mpm

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: 18826240

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18826240

Country of ref document: EP

Kind code of ref document: A1