WO2003003745A1 - Image encoder, image decoder, image encoding method, and image decoding method - Google Patents
Image encoder, image decoder, image encoding method, and image decoding method Download PDFInfo
- Publication number
- WO2003003745A1 WO2003003745A1 PCT/JP2002/006615 JP0206615W WO03003745A1 WO 2003003745 A1 WO2003003745 A1 WO 2003003745A1 JP 0206615 W JP0206615 W JP 0206615W WO 03003745 A1 WO03003745 A1 WO 03003745A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- partial image
- appearance frequency
- information
- decoding
- frequency distribution
- Prior art date
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
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/97—Matching pursuit coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/103—Selection of coding mode or of prediction mode
- H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods 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
- H04N19/136—Incoming video signal characteristics or properties
-
- 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
- H04N19/134—Methods 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
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/137—Motion inside a coding unit, e.g. average field, frame or block difference
-
- 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
- H04N19/134—Methods 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
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/137—Motion inside a coding unit, e.g. average field, frame or block difference
- H04N19/139—Analysis of motion vectors, e.g. their magnitude, direction, variance or reliability
-
- 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
- H04N19/134—Methods 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
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/14—Coding unit complexity, e.g. amount of activity or edge presence estimation
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—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
- H04N19/184—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 bits, e.g. of the compressed video stream
-
- 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
-
- 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
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
-
- 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
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/513—Processing of motion vectors
- H04N19/517—Processing of motion vectors by encoding
- H04N19/52—Processing of motion vectors by encoding by predictive encoding
-
- 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
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/56—Motion estimation with initialisation of the vector search, e.g. estimating a good candidate to initiate a search
-
- 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
-
- 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
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
-
- 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
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/91—Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
Definitions
- the present invention relates to an image encoding device, an image decoding device, an image encoding method, and an image decoding method.
- the present invention relates to an image encoding device, an image decoding device, an image encoding method, and an image decoding method capable of transmitting and storing an image with a small encoding amount.
- DCT discrete cosine transform
- Matching Pursuits is a method of expressing inter-frame prediction residual signals as a linear sum of over-completed basis sets.
- the unit of basis representation is not limited to blocks.
- the ability to express irregular signals in a compact manner with a small number of basis coefficients using a simple basis pattern has the characteristic that visual quality superior to DCT coding can be obtained at low rate coding.
- the prediction residual image signal f to be encoded is a pre-prepared over-completion consisting of n types of bases g k EG (1 ⁇ k ⁇ n). Using the basis set G, it is expressed as T f
- m is the total number of base search steps
- i is the base search step number
- the prediction residual image signal is a target of the base search.
- r. f.
- S i and g ki are s and g k of the arbitrary partial region s (partial region in the frame) above and the arbitrary basis g k included in the basis set G in the i-th step basis search. These are the subregions and bases obtained by selecting the combination that maximizes the inner product value.
- the information to be encoded is:
- An index indicating g ki (g k is held in common on the encoding side and the decoding side, and the basis can be specified by exchanging only the index information),
- a set of these parameters is collectively called an atom.
- the number of atoms to be coded is increased, that is, as the number m of base search total steps is increased, the coding amount is increased and the distortion is reduced.
- Matching Pursuits video coding a frame-based prediction residual signal is prepared for each frame. Specify a partial signal to be expressed as coded information. This partial signal may be located at any position in the frame. However, in a normal video signal, a signal portion having large motion, that is, a signal portion having a large amount of information may be considered as a portion having a large residual signal power.
- a general method is to detect the position of the maximum power in the prediction residual frame as the partial signal.
- the partial signal is represented on a basis, and it is necessary to encode the position information in the frame.
- a base that is often expressed as having the partial signal is selected from among the bases included in the base code book given in advance, and the index and the base coefficient (the inner product value of the partial signal and the base) are selected. Are transmitted and recorded as encoded information.
- SAC Syntax—based Arithmetic Coding mode
- H.263 Syntax—based Arithmetic Coding mode
- arithmetic coding is performed based on the table. Therefore, in cases where the occurrence probability of certain encoded data is affected by the occurrence of other encoded data, or when a video frame is encoded in units of areas such as fixed blocks, certain encoding of each block is performed. No consideration is given to cases where the data is affected by the occurrence of the same encoded data in the surrounding blocks. For this reason, it is not possible to make full use of the efficiency of arithmetic coding, especially for signals having temporal and spatial dependencies such as video.
- an object of the present invention is to improve the entropy coding efficiency of the atom parameter of Matching Pursuits coding by introducing arithmetic coding. Furthermore, in order to improve the efficiency of arithmetic coding, we focus on the interdependency of the atom parameters in Matching Pursuits coding, and define a context (context) model based on conditional probabilities. The purpose is to improve the efficiency of the event-to-peak coding of atom parameters by switching the appearance frequency table.
- an image encoding apparatus includes a position information acquisition unit that acquires position information indicating a position of a unit element of prediction residual data in a predetermined partial image region for each of the partial image regions.
- a context determining means for determining an appearance frequency distribution of the position information in the partial image area to be encoded according to the number of unit elements of the prediction residual data existing in the partial image area;
- Losless encoding means for arithmetically encoding the position information based on the determined appearance frequency distribution.
- the image decoding apparatus corresponding to the above-mentioned image decoding apparatus comprises: an irreversible decoding means for decoding position information indicating a position of a unit element of prediction residual data in a predetermined partial image area for each of the partial image areas; A context determining means for determining an appearance frequency distribution of the position information in the partial image area to be decoded according to the number of unit elements of the prediction residual data existing in the Arithmetic decoding means for performing arithmetic decoding of the position information based on the information.
- an image encoding method comprises: position information for acquiring position information indicating a position of a unit element of prediction residual data in a predetermined partial image region for each of the partial image regions An acquiring step; a context determining step of determining an appearance frequency distribution of the position information in the partial image area to be encoded according to the number of unit elements of the prediction residual data existing in the partial image area; A lossless encoding step of arithmetically encoding the position information based on the determined appearance frequency distribution.
- a corresponding image decoding method includes: an irreversible decoding step of decoding position information indicating a position of a unit element of prediction residual data in a predetermined partial image region for each of the partial image regions; A context determining step of determining the appearance frequency distribution of the position information in the partial image area to be decoded according to the number of unit elements of the prediction residual data existing in the And an arithmetic decoding step of performing arithmetic decoding of the position information based on the arithmetic operation.
- the distribution of position information values largely depends on the number of unit elements (atoms) of the prediction residual data existing in the partial image area. Therefore, the appearance frequency distribution of the position information in the partial image region to be encoded is determined according to the number of unit elements of the prediction residual data existing in the partial image region, and the appearance frequency determined in this manner is determined. By performing arithmetic coding of the position information based on the distribution, it is possible to improve the efficiency of the entity-to-peak coding of the position information.
- an image coding apparatus of the present invention classifies a prediction residual data distribution in a predetermined partial image region, and acquires class information for each of the partial image regions Prediction residual distribution classifying means, and context determining means for determining an appearance frequency distribution of basis coefficient information used for expressing a unit element of prediction residual data in the partial image area based on the class information of the partial image area And an irreversible encoding unit that performs arithmetic encoding of the base coefficient information based on the determined appearance frequency distribution.
- the corresponding image decoding device decodes the class information of the prediction residual data distribution in the predetermined partial image region and the base coefficient information used for expressing the unit element of the prediction residual data in the partial image region.
- Lossless decoding means, and context determining means for determining, based on the class information, an appearance frequency distribution of base coefficient information used to represent a unit element of prediction residual data in the partial image region;
- Arithmetic decoding means for performing arithmetic decoding of the base coefficient information based on the occurrence frequency distribution.
- an image encoding method of the present invention classifies a prediction residual data distribution in a predetermined partial image region, and obtains prediction information for acquiring class information for each of the partial image regions.
- the image decoding method corresponding thereto decodes the class information of the prediction residual data distribution in a predetermined partial image region and the base coefficient information used for expressing the unit element of the prediction residual data in the partial image region.
- the prediction residual data distribution is often divided into several classes (groups) using specific features.
- the appearance frequency distribution of the basis coefficient information is determined based on the class information, and the base coefficient information is arithmetically encoded based on the appearance frequency distribution determined in this manner, whereby the ent-peak coding of the transformed basis information is performed.
- the efficiency is improved.
- an image encoding device of the present invention acquires, for each of the partial image regions, number information indicating the number of unit elements of prediction residual data in a predetermined partial image region A number information obtaining means; and a context determination for determining an appearance frequency distribution of the number information of the partial image area to be coded according to the number information in the partial image area adjacent to the partial image area to be coded.
- Means, and lossless encoding means for performing arithmetic encoding of the number information based on the determined appearance frequency distribution.
- the image decoding apparatus corresponding to this is a reversible decoding unit that decodes, for each of the partial image regions, number information indicating the number of unit elements of the prediction residual data in a predetermined partial image region, and a decoding target.
- a context determining means for determining an appearance frequency distribution of the number information of the partial image area to be decoded in accordance with the number information in the partial image area adjacent to the partial image area;
- Arithmetic decoding means for performing arithmetic decoding of the number information based on the arithmetic information.
- an image encoding method comprises: a number information unit for obtaining, for each of the partial image regions, number information indicating the number of unit elements of prediction residual data in a predetermined partial image region.
- a corresponding image decoding method includes a reversible decoding step of decoding number information indicating the number of unit elements of prediction residual data in a predetermined partial image region for each of the partial image regions, and a decoding target.
- the number of unit elements (atoms) of prediction residual data in a predetermined partial image area largely depends on the number of unit elements (atoms) of prediction residual data in a partial image area adjacent to the partial image area. Therefore, the appearance frequency distribution of the number information of the predetermined partial image area to be coded is determined according to the number information of the unit elements (atoms) in the adjacent partial image areas. By performing arithmetic coding of the number information based on the appearance frequency distribution, it is possible to improve the efficiency of the entity-to-peak coding of the number information.
- an image encoding device of the present invention acquires, for each of the partial image regions, number information indicating the number of unit elements of prediction residual data in a predetermined partial image region
- a lossless encoding unit that arithmetically encodes the number information based on the determined appearance frequency distribution. Sign.
- the image decoding apparatus corresponding to this is a reversible decoding means for decoding the number information indicating the number of unit elements of the prediction residual data in the predetermined partial image area ⁇ for each of the partial image areas, and is a decoding target.
- Reversible decoding means for decoding motion information of an image in the partial image area for each of the partial image areas; Determining means for determining the appearance frequency distribution of the number information of the partial image area to be decoded according to the decoding information, and arithmetic decoding for arithmetically decoding the number information based on the determined appearance frequency distribution Means.
- an image encoding method comprises: An acquisition step; a motion detection step of detecting motion information of an image in a partial image area to be encoded; and a partial image to be encoded according to the motion information of an image in the partial image area to be encoded.
- a corresponding image decoding method includes: an irreversible decoding step of decoding the number information indicating the number of unit elements of the prediction residual data in a predetermined partial image region for each of the partial image regions; A lossless decoding step of decoding the motion information of the image in the partial image region for each of the partial image regions, and an appearance frequency distribution of the number information of the partial image region to be decoded according to the motion information of the image in the partial image region And an arithmetic decoding step of performing arithmetic decoding of the number information based on the determined appearance frequency distribution.
- the number of unit elements of the prediction residual data in the predetermined partial image area largely depends on the motion information of the image in the predetermined partial image area. Therefore, the appearance frequency distribution of the number information of the unit elements in the partial image region to be encoded is determined according to the motion information of the image in the partial image region to be encoded, and the appearance frequency thus determined is determined. By performing arithmetic coding of the number information based on the distribution, it is possible to improve the efficiency of the entity-to-peak coding of the number information.
- an image encoding device includes a predetermined partial image.
- Coefficient information obtaining means for obtaining base coefficient information for each unit element used for representing a unit element of the prediction residual data in the image area, and detecting the prediction residual data in the partial image area to be encoded
- a context determining means for determining an appearance frequency distribution of the base coefficient information of the partial image area to be encoded in accordance with the determined order; and an arithmetic code of the base coefficient information based on the determined appearance frequency distribution.
- Reversible encoding means for performing the conversion.
- the image decoding apparatus corresponding to the base coefficient information of the partial image area to be decoded according to the decoding order of the base coefficient information used to represent the unit element of the prediction residual data in the predetermined partial image area
- an arithmetic decoding means for arithmetically decoding the base coefficient information based on the determined occurrence frequency distribution.
- an image encoding method provides base coefficient information used for expressing a unit element of prediction residual data in a predetermined partial image area for each of the above unit elements.
- the appearance frequency distribution of the base coefficient information of the partial image area to be encoded is determined according to the coefficient information acquisition step to be acquired and the order in which the prediction residual data is detected in the partial image area to be encoded.
- the image decoding method corresponding to the above is based on the base coefficient information of the partial image area to be decoded according to the decoding order of the base coefficient information used for expressing the unit element of the prediction residual data in the predetermined partial image area. And an arithmetic decoding step for performing arithmetic decoding of the base coefficient information based on the determined occurrence frequency distribution.
- the basis coefficient information used for expressing the unit element of the prediction residual data in the predetermined partial image area largely depends on the order in which the prediction residual data is detected in the predetermined partial image area.
- Prediction residual data in the partial image area to be encoded The appearance frequency distribution of the base coefficient information of the partial image area to be encoded is determined according to the order in which the base coefficient information is detected, and the arithmetic code of the base coefficient information is determined based on the appearance frequency distribution determined in this manner. By performing the coding, it is possible to improve the efficiency of the event-to-peak coding of the base coefficient information.
- FIG. 1 is a configuration diagram of an image encoding device according to Embodiment 1.
- FIG. 2 is a configuration diagram of a decoding device according to Embodiment 1.
- 3A to 3G are diagrams showing macro blocks.
- FIGS. 4A and 4B are diagrams showing macroblocks and atoms.
- FIG. 5 is a diagram showing a configuration of the variable length coding unit.
- FIG. 6 is a diagram that defines a context for the encoding mode information of the macroblock C.
- FIGS. 7A and 7B are diagrams showing the distribution of position information values.
- 8A and 8B are diagrams showing the relationship between the atom g k massageand the residual signal f (n).
- FIG. 9 is a diagram showing a configuration of the variable length decoding unit.
- FIG. 10 is a diagram illustrating a state of a sequence (syntax) of encoded data in units of frames of a compressed stream.
- FIG. 11 is a diagram showing the relationship between the atom detection order ⁇ and the atom coefficient information.
- FIG. 12 is a diagram illustrating an example of the distribution of atoms.
- FIG. 13 is a diagram showing a binarization pattern of the encoding mode information.
- FIG. 14 shows an appearance frequency table for the context C M1.
- FIG. 15 shows an appearance frequency table for the context CM2.
- FIG. 16 is a diagram showing a binarization pattern of the atom position information.
- FIG. 17 shows an appearance frequency table for context CM3.
- FIG. 18 is a diagram showing a binarization pattern of the base index.
- FIG. 19 shows an appearance frequency table for context CM4.
- FIG. 20 is a diagram showing a binarization pattern of the number of atoms in a macroblock.
- FIG. 21 shows an appearance frequency table for context CM5.
- FIG. 22 shows an appearance frequency table for the context CM6.
- FIGS. 23A to 23C are diagrams showing the relationship between the atom detection order K and the atom coefficient information according to the value of ACT.
- FIG. 24 is a diagram showing a binarization pattern of the atom coefficient information.
- FIG. 25 shows an appearance frequency table for the context CM7 according to the value of ACT.
- An image coding apparatus and a decoding apparatus receive a video signal consisting of a sequence of image frames, and mainly include an inter-frame motion compensation prediction unit, a coding mode selection unit, and compression coding of a texture signal. And a decoding device that receives compressed video data (hereinafter, referred to as “bit stream”) generated by the coding device and reproduces a video signal.
- bit stream compressed video data
- the input video signal 101 is a time sequence of a frame image, and hereinafter, represents a signal of a frame image unit.
- a frame is divided into a square rectangular area (hereafter, referred to as a “macroblock”) with a fixed size of 16 pixels ⁇ 16 lines, and the following processing is performed on a unit basis. That is, the macroblock data of the input video signal 101 is first sent to the motion detection unit 102, and the motion vector 105 is detected in the motion detection unit 102.
- the motion vector 105 is obtained by referring to a predetermined search area of a past encoded frame 104 (hereinafter, referred to as a “locally decoded image”) stored in the frame memory 103. It finds a pattern similar to the macroblock to be coded and calculates the amount of spatial movement between that pattern and the current macroblock.
- the motion vector 105 is represented by a two-dimensional translation amount. This motion vector 1 0
- a method such as block matching is generally used.
- the unit to which the motion vector 105 is assigned can be defined as a unit in which the macroblock is equally divided into various rectangular areas, and indicates the type of block shape to be used.
- the identification information may be transmitted as coding mode information.
- MC Motion Compressive Mode
- MC mode 1 since the macroblock itself is used as a motion vector assignment unit, one motion vector is determined for each macroblock.
- MC mode 2 shown in Fig. 3B an area obtained by dividing the macro block into left and right halves is used as a motion vector assignment unit, and two motion vectors are determined for one macro block.
- MC mode 7 shown in Fig. 3G 16 motion vectors are determined per macroblock.
- the locally decoded image used for motion detection is not limited to a past frame, and a future frame can be coded first and stored in a frame memory for use. For example, B frame prediction in the MPEG series corresponds to this.
- the motion compensation unit 107 extracts the predicted image 106 from the local decoded image 104 in the frame memory using the motion vector 105.
- the motion detecting unit 102 and the motion compensating unit 107 perform processing for each macroblock, and a difference signal (prediction residual signal 108) from the input video signal 101 is obtained in units of frames. That is, the motion vector 105 of each macroblock is retained over the entire frame, and the predicted image 106 is It is configured as an image.
- a final parameter 112 is generated for the prediction residual signal 108 based on the above-mentioned antoregorism of Matching Pursuits.
- the basis set g k l 11 is stored in the basis codebook 110, from which a basis to be assigned to each partial signal is selected.
- Atom extraction is performed for the entire frame regardless of the macroblock structure. The operation is shown in Figs. 4A and 4B. Figures 4A and 4B show that the position of the atom does not depend on the structure of the macroblock.
- FIGS. 4A and 4B show a prediction residual frame, and a plurality of rectangular areas separated by dotted lines in the figure are macroblocks MB.
- the coding syntax of atoms uses rules that can be transmitted in macroblock units. Therefore, as can be seen from Equation (1), using the fact that the encoding order of the atoms does not affect the decoded image, the frames are sorted in order on the two-dimensional coordinates with the upper left corner of the frame as the origin. And the encoding order is configured so that atoms are counted in macroblock units. In this way, the configuration is such that the atom parameters 112 (basic index, position information, basis coefficients) are encoded for each macroblock by the number of atoms contained therein.
- the atom decoding unit 115 restores the local decoding residual signal 116 from the atom parameters 112 and adds the prediction image 106 to the local decoded image 117 to obtain a local decoded image 117.
- the locally decoded image 1 17 is stored in the frame memory 103 because it is used for motion compensation prediction of the next frame.
- the variable-length decoding unit 118 detects a synchronization word representing the start of each frame, and thereafter, the motion vector 105, The tom parameters 1 1 2 are restored at will.
- the motion vector 105 is sent to a motion compensation unit 107 to obtain a predicted image 106.
- the atom parameters 112 are decoded by the atom decoding unit 115.
- the basis is retrieved by giving the basis codebook 110 a basis index.
- the output 1 16 of the atom decoding unit 1 15 is added to the predicted image 106 to obtain a decoded image 1 17.
- the decoded image 1 17 is stored in the frame memory 103 because it is used for motion compensation prediction of the subsequent frames.
- the decoded image 1 17 is output to the display device at a predetermined display timing, and the video is reproduced.
- variable length coding unit 113 and the variable length decoding unit 118 serving as the points of the present invention will be described.
- the variable-length encoding unit 113 and the variable-length decoding unit 118 are used to determine the dependence of the atom parameters obtained in the image encoding device of FIG. By changing the appearance frequency table adaptively using the dependencies between the two, the most appropriate appearance frequency table is used according to the situation, and the coding efficiency of arithmetic coding is improved.
- FIG. 5 shows the configuration of the variable length coding unit 113.
- the input 1 19 also includes the coding mode information of the macroblock in addition to the above-described atom parameters 112.
- a context is defined for the following information in the input 1 19.
- the context determination unit 120 determines an appearance frequency tape lock 123 suitable for the context at that time according to the bit position 125 to be encoded given from the binarization processing unit 121, and performs arithmetic coding. To the chemical unit 1 24.
- the coding mode information is, for example, whether the macroblock is coded in any of the motion vector adding units shown in FIGS.
- Information such as whether it is inter-coded or intra-frame (intra) -coded is provided to the decoding device as macro-block coded data.
- the encoding mode options are determined, for example, as follows.
- Encoding mode 2 MC mode 1 (Fig. 3 A No atom)
- Encoding mode 3 MC mode 2 (Fig. 3B No atom)
- Encoding mode 4 MC mode 3 (Fig. 3 No atom
- Encoding mode 5 MC mode 4 (Fig. 3D No atom)
- Encoding mode 6 MC mode 5 (Fig. 3 E No atom)
- Encoding mode 7 MC mode 6 (Fig. 3 No F atom
- Encoding mode 8 MC mode 7 (Fig. 3G No atom)
- Coding mode 9 Intra coding, no atoms
- Encoding mode 10 MC mode 1 (Fig. 3A, with atoms
- Encoding mode 1 1 MC mode 2 (Fig. 3B, with atoms
- Encoding mode 12 MC mode 3 (Fig. 3C, with atoms
- Encoding mode 13 MC mode 4 (Fig. 3D, with atoms)
- Coding mode 14 MC mode 5 (Fig. 3E) with atoms
- Encoding mode 15 MC mode 6 (Fig. 3F) with atoms
- Encoding mode 16 MC mode 7 (Fig. 3G) with atoms
- Coding mode 17 Intra coding, with atoms
- “skip” in coding mode 1 is a case where the motion vector is zero and no atom is coded. This corresponds to a mode in which the image at the same spatial position of the reference image is copied as it is.
- coding modes 2 to 8 there are motion vectors corresponding to the individual MC modes in Figs. 3A to 3G, but if there are no atoms to be coded as residuals, coding modes 10 to 16 are If there are motion vectors corresponding to the individual MC modes in Figs.
- coding mode 9 is a case where intra coding is performed and the residual
- encoding mode 17 is a case where intra-coding is performed, and there is an atom to be encoded as a residual. Note that, here, the intra coding assumes that only the DC component is encoded, and that the AC component is encoded by an atom in the form of a residual.
- the context for the encoding mode information is defined as shown in Figure 6.
- a to C indicate adjacent macroblocks
- the context for the coding mode information of the macroblock C is defined as follows.
- CM 1 g (A) + 2 X g (B)
- CM2 h (A) + 2 X h (B)
- CM 2 sets the context as to whether A or B contains an atom to be encoded as a residual. Similarly to CM 1, the frequency of the presence or absence of an atom in macro block C changes according to the surrounding situation, so the appearance frequency table is switched according to the value of CM 2.
- the binarization processing unit 1221 encodes the encoding mode 1 to: L7 so that CM1 and CM2 correspond to the first bit and the second bit, respectively. Implement binarization. Thus, the appearance frequency table is switched according to the bit position. In addition, since the above context always requires information on the left and upper macroblocks, the current macroblock coding mode information is used to determine the context of subsequent macroblocks. Buffer to 120.
- Fig. 13 shows an example of the coding mode information binarization process.
- Fig. 14 shows the four types of appearance frequency table for the first bit.
- Fig. 15 shows the four types of appearance frequency for the second bit. 4 shows an example of a table.
- the value of CM 2 calculated by the above equation is 0. .
- the value of C M2 calculated by the above equation is 3.
- the position information is data expressing the position of an atom in the macroblock as a difference between them.
- the distribution of the position information value greatly depends on the number of atoms in the macroblock. For example, as shown in Fig. 7A, if there are few atoms in the macroblock, the distance between the atoms will be long, and as shown in Fig. 7B, the distance between the atoms will inevitably decrease if there are many atoms. In other words, the frequency distribution of the location information changes according to the value of the number of atoms in the macroblock. Therefore, the context is defined as follows based on this.
- CM 3 is a value for designating the appearance frequency table so as to best reflect the frequency distribution of the position information at that time when the number of atoms in the macroblock is within a predetermined range.
- the binarization processing section 121 binarizes the position information so that CM 3 can be used for switching the appearance frequency of the first bit.
- Fig. 16 shows an example of the binarization process of atom position information
- Fig. 17 shows an example of four types of appearance frequency tables for the first bit.
- the base index is an index number for specifying a base coefficient of each atom in the macroblock.
- a residual signal having a large power and containing a sharp waveform change such as a step edge is localized.
- the residual signal f (n + 1) obtained by subtracting the extracted atoms gkn from f (n) is divided into smaller waveforms as shown in Fig. 8B. The resulting atom tends to have a smaller base size.
- the base codebook is configured to assign smaller index numbers to bases with larger base sizes and higher index numbers to bases with smaller base sizes.
- the appearance frequency distribution of the atom's basis index tends to be biased toward larger values.
- CM4 is a value for designating the appearance frequency table so as to best reflect the frequency distribution of the base index at that time when the number of atoms in the macroblock is within a predetermined range.
- the binarizing processing section 121 binarizes the underlying index so that the CM 4 can be used to switch the appearance frequency of the first bit.
- FIG. 18 shows an example of the binarization process of the atom base index
- FIG. 19 shows an example of four types of appearance frequency tables for the first bit.
- the arithmetic coding of the base index of the atom according to the present invention will be described with reference to FIGS.
- the value of C M4 calculated by the above expression is 3.
- the number of atoms in the macroblock is the number Na of atoms in the macroblock.
- a residual signal having a large power and including a steep waveform change such as a step edge is generated. If it exists, there is a high probability that multiple atoms will be intensively extracted at the same location in subsequent atom extraction steps. Therefore, if the number of atoms in an adjacent macroblock is large, the possibility that the number of atoms in the macroblock is also increased is high.
- Figure 12 shows an example of the typical distribution of atoms in the screen.
- the white part is the atom AT.
- the macroblock adjacent to the macroblock C is A
- the macroblock adjacent to the left is B
- the numbers of atoms in the macroblocks of A and B are N a (A) and N a (B).
- the context for the number of atoms in a macroblock is defined as follows.
- CM5 is a value for designating an appearance frequency table so as to best reflect the frequency distribution of the number of atoms in the macroblock when the number of atoms in an adjacent macroblock is within a predetermined range. is there.
- the binarization processing unit 121 binarizes the number of atoms in the macroblock so that the CM5 can be used to switch the appearance frequency of the first bit.
- FIG. 20 shows an example of binarization processing of the number of atoms in a macroblock
- FIG. 21 shows an example of four types of appearance frequency tables for the first bit.
- CM 5 3
- CM 5 3
- a table with a low probability of becoming 0 that is, the number of atoms in the macroblock takes a value of 1 to 2).
- the context for the number of atoms in a macroblock may be defined as follows.
- MVD X and MVD y are differential motion vectors that are the difference between the horizontal and vertical components MVx and MVy of the motion vector of the macroblock and their predicted values PMVx and PMVy, and the predicted values PMVx and PMVy are For example, a median value of a motion vector value in a nearby macro block is used as used in an encoding method such as MPEG-14 or H.26L.
- Max (X) means that the processing is for the differential motion vector in which the absolute value of the differential motion vector of all the differential motion vectors X in the macroblock gives the maximum value.
- CM 6 a value for designating the appearance frequency table so as to reflect the frequency distribution of the number of atoms in the macroblock most effectively.
- the binarization processing unit 122 binarizes the number of atoms in the macroblock so that the CM6 can be used to switch the appearance frequency of the first bit.
- Fig. 22 shows an example of four types of appearance frequency tables for the first bit after binarization of the number of atoms in a macroblock. Another example of the arithmetic coding of the number of atoms in the macroblock according to the present invention will be described below with reference to FIGS.
- MV1 (0, 1)
- MV2 (2, 1 1)
- MV 3 (1 2, 1)
- CM6 calculated by applying the above equation to MV2 whose absolute value is the largest of MV2 and MV2 is 1.
- the motion in the macroblock to be encoded is considered to be complicated, and the number of atoms in the macroblock is likely to increase. Therefore, as shown in Fig. 22, the number of atoms in the macroblock is binarized.
- FIG. 9 shows the configuration of the variable-length decoding unit 118.
- the variable-length decoding unit 118 receives the compressed stream 114 as an input and, like the variable-length encoding unit 113, generates an appearance frequency table 1 in bit units according to the context determined by the context determining unit 120. Twenty three (The same table definition as the variable-length coding unit 113) is switched, and the arithmetic decoding unit 126 performs arithmetic decoding to binary data.
- the restored binarized data 122 is converted to final decoded data 119 by the inverse binarization processing unit 127 according to the same rules as those on the encoding side.
- the same distortion-free decoded value as the input 1 19 to the variable length coding unit 113 is obtained.
- data used for determining the context of the decoding process of the subsequent macroblock is sent to the context determination unit 120 and buffered.
- Fig. 10 shows the state of the encoded data arrangement (syntax) of the compressed stream 1 14 in frame units. Following the frame header that stores the time stamp of the frame, initial parameters, and activity information (details will be described later), for example, information on the movement of macroblocks and the like described in Embodiment 1 in fixed block units is collected. Multiplex.
- the counter N is the number of macroblocks in a frame, and is generally a constant uniquely determined for the image size. Next, a sync word is inserted.
- This synchronization word must be a unique code indicating the start of the atom parameter of the frame.
- the decoding device separates the motion information coded in macroblock units and the atom parameters coded in frame units in advance, and outputs the separated motion information to each decoding processing block. Since the processing can be performed in parallel, it is possible to speed up the decoding processing. Also, in the case of a line error or the like, if it is confirmed that an error has occurred after the synchronization word, it is possible to secure flexibility such as performing error concealment processing using only motion information.
- the atom parameters are multiplexed in the order in which they were detected in the frame.
- the atom parameter counter M generally varies from frame to frame and is not known.
- the value of the counter M is embedded in the synchronization word itself, a terminate code is provided in the position information which is the leading information of each atom parameter, and when the terminate code is detected, the atom parameter of the frame is decoded. There are methods such as terminating processing.
- Each atom parameter is encoded in the order in which it was detected in the frame.
- the basis coefficient for each detected atom is referred to as atom coefficient information or atom coefficient.
- Atoms are usually coded in order of importance in video information.
- the essential information in the video information is the degree of motion, and control is performed so that the motion is large and the inter-frame prediction residual power is large.
- the atom coefficient information has a large value at the beginning of detection, but is detected by gradually removing power by each atom. Each time, the value of coefficient information tends to decrease.
- the K-th atom coefficient information has a strong correlation with the K-1st atom coefficient information multiplexed immediately before. Therefore, the Kth atom coefficient is encoded as difference information from the K-1st atom coefficient.
- the distribution in FIG. 11 may deviate from the distribution in FIG.
- the activity information for classifying such a residual signal distribution is inserted into the frame header, and the frequency encoding table of the atom coefficient information is switched according to the activity information to perform the arithmetic coding. .
- the variance in the distribution of the power E of the inter-frame prediction residual signal is reduced, and the classification is performed by the following equation according to the value of the variance ⁇ .
- ACT is classified activity information
- TH1 and TH2 are preset thresholds.
- FIGS. 23A to 23C The distribution of typical atom coefficient information for the activity information ACT classified in this way is illustrated in FIGS. 23A to 23C.
- the larger the number of ACTs the larger the atom coefficient information changes from a large value to a small value as the atom detection order K increases. Conversely, as the value of ACT is smaller, the atom coefficient information does not change over a large range even if the atom detection order K changes.
- the context of the atom coefficient information is determined by the following equation.
- the ⁇ th atom coefficient information is encoded as difference information between the ⁇ -1th atom coefficient and the ⁇ th atom coefficient. As shown in FIG. 23, the difference information tends to decrease as the value of ⁇ increases.
- activity information ACT is calculated from the inter-frame prediction residual signal generated in the motion compensation unit 107.
- the variance ⁇ becomes small and the value of ACT becomes 0.
- the first bit after the binarization processing of the atom coefficient information becomes 0 as shown in Fig. 25 (that is, the atom (The coefficient information takes a value of 0 to 3.)
- the motion of the screen is complicated, and the power distribution of the inter-frame prediction residual signal is wide.
- the variance value ⁇ becomes a large value
- the value of ACT becomes 2, for example.
- the value of C M7 is calculated from the above equation according to each detection order ⁇ .
- the detection order K is 1 for the first detected atom
- the value of CM 7 is 0.
- the detection order K 1 0 for the 10th detected atom, so the value of CM 7 is 2.
- the appearance frequency table used for arithmetic coding of the atom coefficient information can be appropriately selected according to the motion information of the entire screen and the atom detection order, so that the coding efficiency can be improved. It becomes possible.
- the decoding side decodes the activity information included in the frame header of the compressed stream coded by the above-described coding method, and performs the same processing as the coding side in accordance with the decoded activity information.
- the correct decoding can be achieved by selecting the table set in a manner.
- the decoding can be correctly performed by selecting the appearance frequency table in the same manner as the encoding side according to the number of atoms in the macroblock decoded earlier.
- the present invention can be used as a moving image encoding device or a decoding device.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2003509782A JP3920849B2 (ja) | 2001-06-29 | 2002-06-28 | 画像符号化装置、画像復号装置、画像符号化方法、及び画像復号方法 |
EP02738884A EP1404133A4 (en) | 2001-06-29 | 2002-06-28 | IMAGE ENCODER, IMAGE DECODER, IMAGE ENCODING METHOD, AND IMAGE DECODING METHOD |
US10/362,364 US7292731B2 (en) | 2001-06-29 | 2002-06-28 | Image encoder, image decoder, image encoding method, and image decoding method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2001199008 | 2001-06-29 | ||
JP2001-199008 | 2001-06-29 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2003003745A1 true WO2003003745A1 (en) | 2003-01-09 |
Family
ID=19036358
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2002/006615 WO2003003745A1 (en) | 2001-06-29 | 2002-06-28 | Image encoder, image decoder, image encoding method, and image decoding method |
Country Status (5)
Country | Link |
---|---|
US (1) | US7292731B2 (ja) |
EP (1) | EP1404133A4 (ja) |
JP (1) | JP3920849B2 (ja) |
CN (1) | CN1305311C (ja) |
WO (1) | WO2003003745A1 (ja) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100409691C (zh) * | 2003-02-28 | 2008-08-06 | 株式会社Ntt都科摩 | 图像编码设备、方法和图像解码设备、方法 |
WO2013069255A1 (ja) * | 2011-11-07 | 2013-05-16 | パナソニック株式会社 | 画像符号化方法、画像復号方法、画像符号化装置、画像復号装置、および画像符号化復号装置 |
JP2016036176A (ja) * | 2009-05-29 | 2016-03-17 | 三菱電機株式会社 | 画像復号装置、画像符号化装置およびビットストリーム |
CN105551503A (zh) * | 2015-12-24 | 2016-05-04 | 武汉大学 | 基于原子预选择的音频匹配追踪方法与系统 |
US9628819B2 (en) | 2003-03-06 | 2017-04-18 | Thomson Licensing Dtv | Method for coding a video image taking into account the part relating to a component of a movement vector |
US9924199B2 (en) | 2005-08-26 | 2018-03-20 | Rgb Systems, Inc. | Method and apparatus for compressing image data using compression profiles |
Families Citing this family (42)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6965644B2 (en) * | 1992-02-19 | 2005-11-15 | 8×8, Inc. | Programmable architecture and methods for motion estimation |
US7336711B2 (en) * | 2001-11-16 | 2008-02-26 | Ntt Docomo, Inc. | Image encoding method, image decoding method, image encoder, image decode, program, computer data signal, and image transmission system |
JP4019790B2 (ja) * | 2002-05-23 | 2007-12-12 | 富士ゼロックス株式会社 | 画像処理装置および画像処理プログラム |
EP1453004A2 (en) * | 2003-02-28 | 2004-09-01 | NTT DoCoMo, Inc. | Image encoding apparatus and method |
US8705613B2 (en) * | 2003-06-26 | 2014-04-22 | Sony Corporation | Adaptive joint source channel coding |
US7746929B2 (en) * | 2003-08-05 | 2010-06-29 | Trident Microsystems (Far East) Ltd. | Video encoding and decoding methods and corresponding devices |
JP2007502561A (ja) | 2003-08-12 | 2007-02-08 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | ビデオエンコードおよびデコードの方法および対応する装置 |
US8582640B2 (en) * | 2003-12-16 | 2013-11-12 | Sony Corporation | Adaptive joint source channel coding |
US7689051B2 (en) | 2004-04-15 | 2010-03-30 | Microsoft Corporation | Predictive lossless coding of images and video |
US20060262860A1 (en) * | 2005-02-23 | 2006-11-23 | Chou Jim C | Macroblock adaptive frame/field coding architecture for scalable coding |
JP4444180B2 (ja) * | 2005-07-20 | 2010-03-31 | 株式会社東芝 | テクスチャ符号化装置、テクスチャ復号化装置、方法、およびプログラム |
JP4662171B2 (ja) * | 2005-10-20 | 2011-03-30 | ソニー株式会社 | 符号化装置および方法、復号化装置および方法、プログラム、並びに記録媒体 |
JP2007166039A (ja) * | 2005-12-09 | 2007-06-28 | Matsushita Electric Ind Co Ltd | 画像符号化装置 |
US8265392B2 (en) * | 2006-02-07 | 2012-09-11 | Qualcomm Incorporated | Inter-mode region-of-interest video object segmentation |
US8150155B2 (en) | 2006-02-07 | 2012-04-03 | Qualcomm Incorporated | Multi-mode region-of-interest video object segmentation |
US8265349B2 (en) * | 2006-02-07 | 2012-09-11 | Qualcomm Incorporated | Intra-mode region-of-interest video object segmentation |
JP4555257B2 (ja) * | 2006-06-06 | 2010-09-29 | パナソニック株式会社 | 画像符号化装置 |
JP4902854B2 (ja) * | 2006-09-12 | 2012-03-21 | パナソニック株式会社 | 動画像復号化装置、動画像復号化方法、動画像復号化プログラム、動画像符号化装置、動画像符号化方法、動画像符号化プログラム、及び動画像符号化復号化装置 |
US20080084924A1 (en) * | 2006-10-05 | 2008-04-10 | Donald Martin Monro | Matching pursuits basis selection design |
US7974488B2 (en) * | 2006-10-05 | 2011-07-05 | Intellectual Ventures Holding 35 Llc | Matching pursuits basis selection |
US7522076B1 (en) * | 2007-10-16 | 2009-04-21 | Mediatek Inc. | Parallel context adaptive binary arithmetic coding |
JPWO2010079736A1 (ja) * | 2009-01-09 | 2012-06-21 | パナソニック株式会社 | 通信端末および通信状態変更方法 |
KR20110090511A (ko) * | 2010-02-04 | 2011-08-10 | 삼성전자주식회사 | 통신 기기에서 3차원 영상을 위한 영상 처리 방법 및 장치 |
US9461872B2 (en) * | 2010-06-02 | 2016-10-04 | Hewlett Packard Enterprise Development Lp | Compressing data in a wireless network |
KR20120014676A (ko) * | 2010-08-10 | 2012-02-20 | 에스케이 텔레콤주식회사 | 적응적 부호화/복호화 모드 인덱싱 기법을 이용한 영상 부호화/복호화 장치 및 방법 |
JP5485851B2 (ja) | 2010-09-30 | 2014-05-07 | 日本電信電話株式会社 | 映像符号化方法,映像復号方法,映像符号化装置,映像復号装置およびそれらのプログラム |
WO2012082029A1 (en) * | 2010-12-14 | 2012-06-21 | Telefonaktiebolaget L M Ericsson (Publ) | Method and device for storing a compression ratio indication in a pixel value buffer in tile encoding |
US8446301B2 (en) | 2011-04-15 | 2013-05-21 | Research In Motion Limited | Methods and devices for coding and decoding the position of the last significant coefficient |
CA2837532C (en) | 2011-06-23 | 2018-10-02 | Panasonic Corporation | Image decoding method, image coding method, image decoding apparatus, image coding apparatus, and image coding and decoding apparatus |
KR102062283B1 (ko) | 2011-06-24 | 2020-01-03 | 선 페이턴트 트러스트 | 화상 복호 방법, 화상 부호화 방법, 화상 복호 장치, 화상 부호화 장치 및 화상 부호화 복호 장치 |
CA2830046C (en) | 2011-06-24 | 2018-09-04 | Panasonic Corporation | Image decoding method, image coding method, image decoding apparatus, image coding apparatus, and image coding and decoding apparatus |
CN103563375B (zh) | 2011-06-27 | 2017-04-05 | 太阳专利托管公司 | 图像解码方法、图像编码方法、图像解码装置、图像编码装置及图像编码解码装置 |
CA2837535C (en) | 2011-06-28 | 2018-09-04 | Panasonic Corporation | Image decoding method, image coding method, image decoding apparatus, image coding apparatus, and image coding and decoding apparatus |
MX2013010892A (es) | 2011-06-29 | 2013-12-06 | Panasonic Corp | Metodo de decodificacion de imagenes, metodo de codificacion de imagenes, aparato de decodificacion de imagenes, aparato de codificacion de imagenes y aparato de codificacion y decodificacion de imagenes. |
CN103583048B (zh) | 2011-06-30 | 2017-05-17 | 太阳专利托管公司 | 图像解码方法、图像编码方法、图像解码装置、图像编码装置及图像编码解码装置 |
AU2012277219A1 (en) | 2011-06-30 | 2013-09-19 | Sun Patent Trust | Image decoding method, image encoding method, image decoding device, image encoding device, and image encoding/decoding device |
KR20140034209A (ko) | 2011-07-11 | 2014-03-19 | 파나소닉 주식회사 | 화상 복호 방법, 화상 부호화 방법, 화상 복호 장치, 화상 부호화 장치 및 화상 부호화 복호 장치 |
WO2013047805A1 (ja) * | 2011-09-29 | 2013-04-04 | シャープ株式会社 | 画像復号装置、画像復号方法および画像符号化装置 |
DK2763414T3 (da) | 2011-09-29 | 2020-11-30 | Sharp Kk | Billedafkodningsindretning og fremgangsmåde til billedafkodning til udførelse af konvertering af bi-forudsigelse til uni-forudsigelse |
US9154792B2 (en) * | 2011-11-08 | 2015-10-06 | Qualcomm Incorporated | Progressive coding of position of last significant coefficient |
BR122020010110B1 (pt) * | 2012-03-28 | 2021-08-17 | JVC Kenwood Corporation | Dispositivo de codificação de imagem |
CN103675444B (zh) * | 2012-08-30 | 2016-09-14 | 中国石油化工股份有限公司 | 一种高精度的时频分析方法 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09331535A (ja) * | 1995-11-02 | 1997-12-22 | Canon Inc | 情報処理装置及び方法 |
JPH11220728A (ja) * | 1998-02-02 | 1999-08-10 | Canon Inc | 画像圧縮時のコンテクスト生成方法及び画像圧縮方法、画像処理装置 |
JP2001103315A (ja) * | 1999-09-28 | 2001-04-13 | Matsushita Graphic Communication Systems Inc | 画像符号化方法および画像符号化装置 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5136371A (en) * | 1990-03-15 | 1992-08-04 | Thomson Consumer Electronics, Inc. | Digital image coding using random scanning |
US5150209A (en) * | 1990-05-11 | 1992-09-22 | Picturetel Corporation | Hierarchical entropy coded lattice threshold quantization encoding method and apparatus for image and video compression |
JPH06331535A (ja) * | 1993-05-25 | 1994-12-02 | Asahi Chem Ind Co Ltd | 流動状態における粉体付着力の測定方法 |
US5654702A (en) * | 1994-12-16 | 1997-08-05 | National Semiconductor Corp. | Syntax-based arithmetic coding for low bit rate videophone |
JP3115199B2 (ja) * | 1994-12-16 | 2000-12-04 | 松下電器産業株式会社 | 画像圧縮符号化装置 |
US6614847B1 (en) * | 1996-10-25 | 2003-09-02 | Texas Instruments Incorporated | Content-based video compression |
JP3132456B2 (ja) * | 1998-03-05 | 2001-02-05 | 日本電気株式会社 | 階層的画像符号化方式、及び階層的画像復号方式 |
US6563953B2 (en) * | 1998-11-30 | 2003-05-13 | Microsoft Corporation | Predictive image compression using a single variable length code for both the luminance and chrominance blocks for each macroblock |
-
2002
- 2002-06-28 WO PCT/JP2002/006615 patent/WO2003003745A1/ja active Application Filing
- 2002-06-28 EP EP02738884A patent/EP1404133A4/en not_active Withdrawn
- 2002-06-28 CN CNB028022653A patent/CN1305311C/zh not_active Expired - Fee Related
- 2002-06-28 JP JP2003509782A patent/JP3920849B2/ja not_active Expired - Fee Related
- 2002-06-28 US US10/362,364 patent/US7292731B2/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09331535A (ja) * | 1995-11-02 | 1997-12-22 | Canon Inc | 情報処理装置及び方法 |
JPH11220728A (ja) * | 1998-02-02 | 1999-08-10 | Canon Inc | 画像圧縮時のコンテクスト生成方法及び画像圧縮方法、画像処理装置 |
JP2001103315A (ja) * | 1999-09-28 | 2001-04-13 | Matsushita Graphic Communication Systems Inc | 画像符号化方法および画像符号化装置 |
Non-Patent Citations (3)
Title |
---|
GHARAVI-AIKHANSARI MOHAMMAD: "A model for entropy coding in matching pursuit", PROCEEDINGS. INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, vol. 1, 4 October 1998 (1998-10-04) - 7 October 1998 (1998-10-07), CHICAGO, IL, USA, pages 778 - 782, XP001044670 * |
NEFF RALPH ET AL.: "Very low bit-rate video coding based on matching pursuit", IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, vol. 7, no. 1, February 1997 (1997-02-01), pages 158 - 171, XP000678888 * |
See also references of EP1404133A4 * |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100409691C (zh) * | 2003-02-28 | 2008-08-06 | 株式会社Ntt都科摩 | 图像编码设备、方法和图像解码设备、方法 |
US9628819B2 (en) | 2003-03-06 | 2017-04-18 | Thomson Licensing Dtv | Method for coding a video image taking into account the part relating to a component of a movement vector |
US10244263B2 (en) | 2005-08-26 | 2019-03-26 | Rgb Systems, Inc. | Method and apparatus for packaging image data for transmission over a network |
US10051288B2 (en) | 2005-08-26 | 2018-08-14 | Rgb Systems, Inc. | Method and apparatus for compressing image data using a tree structure |
US9930364B2 (en) | 2005-08-26 | 2018-03-27 | Rgb Systems, Inc. | Method and apparatus for encoding image data using wavelet signatures |
US9924199B2 (en) | 2005-08-26 | 2018-03-20 | Rgb Systems, Inc. | Method and apparatus for compressing image data using compression profiles |
JP2016036176A (ja) * | 2009-05-29 | 2016-03-17 | 三菱電機株式会社 | 画像復号装置、画像符号化装置およびビットストリーム |
US8781001B2 (en) | 2011-11-07 | 2014-07-15 | Panasonic Corporation | Image coding method, image coding apparatus, image decoding method and image decoding apparatus |
RU2610249C2 (ru) * | 2011-11-07 | 2017-02-08 | ТАГИВАН II ЭлЭлСи | Способ кодирования изображения, устройство кодирования изображения, способ декодирования изображения и устройство декодирования изображения |
US9491465B2 (en) | 2011-11-07 | 2016-11-08 | Tagivan Ii Llc | Image coding method, image coding apparatus, image decoding method and image decoding apparatus |
US9712849B2 (en) | 2011-11-07 | 2017-07-18 | Tagivan Ii Llc | Image coding method, image coding apparatus, image decoding method and image decoding apparatus |
JP2017147742A (ja) * | 2011-11-07 | 2017-08-24 | タギバン ツー エルエルシー | 復号方法および復号装置 |
JP2014075807A (ja) * | 2011-11-07 | 2014-04-24 | Panasonic Corp | 復号方法および復号装置 |
JP5415656B2 (ja) * | 2011-11-07 | 2014-02-12 | パナソニック株式会社 | 符号化方法および符号化装置 |
WO2013069255A1 (ja) * | 2011-11-07 | 2013-05-16 | パナソニック株式会社 | 画像符号化方法、画像復号方法、画像符号化装置、画像復号装置、および画像符号化復号装置 |
US10334280B2 (en) | 2011-11-07 | 2019-06-25 | Tagivan Ii Llc | Image coding method, image coding apparatus, image decoding method and image decoding apparatus |
US10631014B2 (en) | 2011-11-07 | 2020-04-21 | Tagivan Ii Llc | Image coding method, image coding apparatus, image decoding method and image decoding apparatus |
CN105551503A (zh) * | 2015-12-24 | 2016-05-04 | 武汉大学 | 基于原子预选择的音频匹配追踪方法与系统 |
CN105551503B (zh) * | 2015-12-24 | 2019-03-01 | 武汉大学 | 基于原子预选择的音频匹配追踪方法与系统 |
Also Published As
Publication number | Publication date |
---|---|
US7292731B2 (en) | 2007-11-06 |
JP3920849B2 (ja) | 2007-05-30 |
US20040131268A1 (en) | 2004-07-08 |
CN1305311C (zh) | 2007-03-14 |
JPWO2003003745A1 (ja) | 2004-10-21 |
EP1404133A1 (en) | 2004-03-31 |
EP1404133A4 (en) | 2010-04-21 |
CN1465190A (zh) | 2003-12-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2003003745A1 (en) | Image encoder, image decoder, image encoding method, and image decoding method | |
JP3842803B2 (ja) | 信号符号化方法、信号復号方法、信号符号化装置、信号復号装置、信号符号化プログラム、及び信号復号プログラム | |
EP1379000B1 (en) | Signal encoding method and apparatus and decoding method and apparatus | |
JP5379020B2 (ja) | イントラ予測符号化、復号化方法及び装置 | |
CN101001374B (zh) | 为隔行扫描和逐行扫描视频编码和解码宏块和运动信息中的革新 | |
JP4993676B2 (ja) | 画像符号化装置及び画像符号化方法 | |
CN105556971A (zh) | 针对帧内块复制预测中的块翻动和跳跃模式的编码器侧判定 | |
CN102239693A (zh) | 运动图像解码方法和运动图像编码方法 | |
JPWO2006073116A1 (ja) | 映像符号化方法及び装置、映像復号方法及び装置、それらのプログラムおよびそれらプログラムを記録した記録媒体 | |
CN101005620B (zh) | 为隔行扫描和逐行扫描视频编码和解码宏块和运动信息中的革新 | |
JPWO2003043347A1 (ja) | 画像符号化方法、画像復号方法、画像符号化装置、画像復号装置、プログラム、コンピュータデータ信号、及び画像伝送システム | |
KR20100046202A (ko) | 이미지들을 처리하기 위한 방법 및 그 대응하는 전자 장치 | |
KR102267770B1 (ko) | 픽쳐들의 그룹에서 변경가능 엘리먼트들의 세트를 결정하기 위한 방법 및 디바이스 | |
CN101569201A (zh) | 基于帧内预测进行编码和解码的方法和装置 | |
CN101554058B (zh) | 基于帧内预测进行编码和解码的方法和装置 | |
CN100539670C (zh) | 已编码a/v序列的编辑 | |
CN106028029A (zh) | 低延时低复杂度的数字kvm切换器视频无损压缩方法及系统 | |
CN100551064C (zh) | 变长编码方法及装置 | |
JP3491001B1 (ja) | 信号符号化方法、信号復号方法、信号符号化装置、信号復号装置、信号符号化プログラム、及び、信号復号プログラム | |
KR20070090494A (ko) | 평균 움직임 벡터를 이용한 인터 프레임 에러 은닉 장치 및방법 | |
KR20040008360A (ko) | 개선된 움직임 벡터 부호화 및 복호화 방법과 그 장치 | |
CN102783036A (zh) | 使用辅码字指示符的编码和解码方法和设备 | |
Vidyavathi et al. | Motion-compensated video compression using variable length Huffman coding | |
JP2003209844A (ja) | 動画像符号化装置および動画像符号化方法 | |
JPH1051792A (ja) | 動画像符号化装置及び符号化方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CN JP US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR |
|
ENP | Entry into the national phase |
Ref country code: JP Ref document number: 2003 509782 Kind code of ref document: A Format of ref document f/p: F |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2003509782 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 028022653 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2002738884 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10362364 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 2002738884 Country of ref document: EP |