WO2003063503A1 - Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus - Google Patents
Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus Download PDFInfo
- Publication number
- WO2003063503A1 WO2003063503A1 PCT/JP2003/000664 JP0300664W WO03063503A1 WO 2003063503 A1 WO2003063503 A1 WO 2003063503A1 JP 0300664 W JP0300664 W JP 0300664W WO 03063503 A1 WO03063503 A1 WO 03063503A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- motion vector
- accuracy
- block
- decoding
- decoded
- 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
- H04N19/50—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
- H04N19/57—Motion estimation characterised by a search window with variable size or shape
-
- 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
-
- 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
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
- H04N19/43—Hardware specially adapted for motion estimation or 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
- H04N19/46—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
- 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
- H04N19/523—Motion estimation or motion compensation with sub-pixel accuracy
-
- 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
- H04N19/60—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
Definitions
- the present invention relates to a moving picture signal coding method, a decoding method, a coding apparatus and a decoding apparatus, and more particularly to a motion vector coding and decoding technique.
- FIG. 3 An outline of the encoding / decoding processing of a moving image will be described.
- moving picture coding is performed by the coding apparatus shown in FIG.
- one frame of the video to be encoded is composed of one luminance signal (Y signal: 61) and two color difference signals (Cr signal: 62, Cb signal: 63).
- the image size of the chrominance signal is half that of the luminance signal both vertically and horizontally.
- each frame of a moving image is divided into small blocks, and playback processing is performed in block units called macro blocks.
- the macro block includes one Y signal block 30 of 16 ⁇ 16 pixels, and a Cr signal block 31 and a Cb signal block 31 of 8 ⁇ 8 pixels which spatially match the Y signal block 30.
- the encoding of the image signal is processed on a macroblock basis. Coding is roughly classified into intra coding (intra mode) and predictive coding (inter mode).
- the input Mac mouth block image 201 to be encoded is divided into DCT encoding blocks, and the DCT converter 203 directly performs DCT 203, Converted to coefficients. Replacement) 5 ⁇
- Each transform coefficient is 204 by a quantizer and encoded by a multiplexer 206.
- the DCT coding block size is generally 8 ⁇ 8 pixels in the conventional coding method, but recently, DCT transform using a 4 ⁇ 4 pixel size has been studied in MPEG-4 Part 10 (Advanced Video Coding) and the like.
- each macro block is divided into 24 4 ⁇ 4 pixel blocks, and each DCT is converted by DCT converter 203 into 16 DCT coefficients.
- a part of the quantized DCT coefficient is decoded into an input macroblock by an inverse quantization unit 207 and an inverse DCT unit 208 of a local decoder 220 and synthesized into a frame memory 210.
- Local decoder 220 creates a decoded image similar to that of a decoding device described later.
- the image stored in the frame memory 210 is used for prediction in the time direction described later.
- This intra coding is placed in macroblocks (including the first coded frame) that do not have a part similar to the previous frame, or in parts where it is desired to eliminate the accumulation operation error associated with DCT.
- motion compensation In the case of inter (predictive coding) coding, motion compensation between the input macroblock image 201 and the decoded image (reference frame) of the coded frame stored in the frame memory 210 is performed by the motion compensator 211. Done. Motion compensation refers to searching a reference frame for a part similar to the contents of the target macroblock (generally, selecting a part with a small sum of absolute values or squared errors of the prediction error signal in the block). This is a temporal compression technique that encodes the amount of motion (motion vector).
- FIG. 4 shows the processing structure for motion compensation.
- FIG. 4 is a diagram showing a prediction block 75 and a motion vector 76 on a previous frame 73 (reference frame) for a luminance signal block 72 of a current frame 71 surrounded by a thick frame.
- the motion vector 76 indicates the movement from the block 74 (broken line) on the previous frame corresponding to the same position spatially to the thick frame block of the current frame to the predicted block 75 area on the previous frame (
- the motion vector length for the color difference signal is No. and no encoding).
- the detected motion vector is subjected to difference processing with the predicted motion vector obtained from the MV predictor 215 (predicted from the motion vector of the adjacent block), and then encoded by the multiplexer 206. Is done.
- the predicted macroblock image 213 extracted from the previous frame due to the motion compensation is subjected to a difference process between the input macroblock image 201 of the current frame and the differentiator 202 to generate a differential macroblock image. Is
- the differential macroblock image is divided into 24 4 ⁇ 4 pixel blocks, and each is converted into 16 DCT coefficients by the DCT converter 203.
- Each DCT coefficient is quantized by a quantizer 204 and encoded by a multiplexer 206.
- the quantized DCT coefficients are decoded into a difference macroblock image by the inverse quantizer 207 and the inverse DCT unit 208 of the local decoder 220, and added to the predicted macroblock image by the adder 209. After that, it is synthesized into the frame memory 210.
- the number of motion vectors to be encoded in one macroblock is one, but the macroblock is further divided into small blocks, and the motion vector is calculated and encoded for each small block.
- the method is common.
- Fig. 7 shows the macroblock division type (luminance component) when performing motion compensation.
- Type 51 to Type 54 is selected for each of the four 8 ⁇ 8 blocks 54-0 to 54-3, and is further selected from five types of types 54a, 54b, 54c, 54d and intra coding.
- the encoding accuracy of a motion vector is handled down to an integer pixel or less.
- 1/2 pixel precision and 1/4 pixel precision can be selected for each bit data.
- the method of calculating the predicted value with integer pixel accuracy or less is omitted.
- a 1/2 pixel accuracy a 2-tap linear filter is used, and for a 1/4 pixel accuracy, an 8-tap filter and a 2-tap filter are adaptively applied according to the pixel position.
- MPEG-4 Part 10 is studying to represent a motion vector with 1/4 pixel accuracy, and a predicted value is calculated using a 6-tap filter and a 2-tap filter.
- the prediction performance is improved by using a filter with a long tap number.
- P_Pi a frame constituted by forward predictive coding or Intora encoding C ture, bidirectional coded coding or intra-coding
- the frame composed of these is called a BP image.
- the INTRA / INTER decision unit 214 shown in Fig. 13 decides whether to perform each macroblock by predictive coding or intra coding, and multiplexes the decision result 218.
- Encoder 206 for encoding Decoding of a coded signal of a moving image is performed in the reverse procedure of coding. Generally, this is performed by the moving picture decoding apparatus shown in FIG.
- the decoding unit 501 analyzes the input coded data and quantizes the motion vector related information and macroblock type information to the MV predictor 508.
- the DCT coefficient information is distributed to the inverse quantizer 502. If the macroblock type is intra-coded, the decoded quantized DCT coefficient information is converted by the inverse quantizer 502 and inverse DCT unit 503 into inverse quantum Performs inverse DCT processing and reproduces macroblock images. If the prediction mode of the macroblock is predictive coding, the MV predictor 508 calculates the decoded motion vector from the decoded difference motion vector and the predicted motion vector, and the motion compensator 504 Is input to The motion compensator 504 extracts a predicted macroblock image from the frame memory 507 in which the decoded image of the previous frame is stored according to the amount of motion.
- the coded data relating to the prediction error signal is subjected to inverse quantization / inverse DCT processing for each 4 ⁇ 4 pixel block in an inverse quantizer 502 and an inverse DCT unit 503 to reproduce a differential macroblock image.
- the predicted macroblock image and the difference macroblock image are added by the adder 505 to reproduce the macroblock image.
- the reproduced macroblock image is combined with the decoded frame image by the combiner 506.
- the decoded frame image is stored in the frame memory 50 for prediction of the next frame.
- Conventionally known motion compensation coding of video signals includes multiple types of motion vectors included in a block, and the coding accuracy of the motion vector is as high as 1/8 pixel accuracy.
- a method is being considered.
- due to the relationship between the number of motion vector codes there is a tendency to select a coding method with a small number of motion vectors even when a block includes a plurality of motions.
- a predicted pixel with integer pixel precision or less is generated using a filter with a long tap number
- a memory access problem that the number of pixels in a reference image used to generate one predicted block increases. But appear. This problem is particularly problematic when performing motion compensation.
- an object of the present invention is to provide a moving image signal encoding method, a decoding method, and a method capable of reducing the amount of motion vector code and reducing motion prediction error information due to an increase in the number of blocks in an encoding block.
- An object of the present invention is to provide an encoding device and a decoding device.
- Another object of the present invention is to reduce the memory access range at the time of motion compensation in an encoding device and a decoding device. Disclosure of the invention
- the present invention divides an image into a plurality of coding blocks, predicts a motion vector in a coding block from a motion vector at an adjacent position, and calculates a motion vector to be coded.
- a moving image encoding method including a step of encoding a motion vector and a differential motion vector calculated from the predicted motion vector in block units.
- the motion vector is coded.
- the accuracy of the differential motion vector is selected for each coding block by preparing several types of differential motion vector accuracy, and information on the selected accuracy and the differential motion of the accuracy are selected.
- the determination of the accuracy value (hereinafter abbreviated as mv_shift value) associated with the encoding of the differential motion vector and the encoding process are performed for all macroblocks.
- the mv_shift value of a specific macroblock is used. This includes the case where encoding is omitted.
- an image of a coded block is reproduced by using a code obtained by the moving image coding according to the present invention, in a procedure reverse to the coding procedure.
- multiple precisions of the motion vector are prepared, and different precisions are assigned to the horizontal and vertical components of the motion vector.
- Each component is corrected to the assigned precision, and a difference motion vector calculated from the corrected prediction motion vector and the motion vector with the assigned precision is encoded.
- the setting of the mv_shi ft value is performed in accordance with the process in which the code to be decoded is encoded. For example, in coding, the type of macroblock, or mv- in the surrounding situation. In the reproduction of a code whose shift value is not encoded, the decoding macroblock type or mv- in the surrounding situation. The shi ft value is determined and processed.
- FIG. 1 is a diagram showing an example of the data syntax of a forward prediction frame.
- FIG. 2 is a diagram illustrating an example of the data syntax of a bidirectional prediction frame.
- FIG. 3 is a diagram illustrating the macroblock division.
- FIG. 4 is a diagram illustrating the principle of motion compensation.
- FIG. 3 is a diagram showing a block configuration used for the following.
- FIG. 6 is a diagram showing a macro block configuration showing a block unit for performing DCT processing and encoding processing.
- FIG. 7 is a diagram showing a configuration of a luminance block indicating a block unit for performing motion compensation.
- FIG. 3 is a diagram illustrating generation of a predicted motion vector.
- FIG. 9 is a diagram showing an example of a code table of the motion vector accuracy.
- FIG. 10 is a diagram illustrating a method of predicting motion vector accuracy.
- FIG. 11 is a diagram showing a code table of a macroblock type and a code table of an 8x8 partition type in P-Picture according to the embodiment of the present invention.
- FIG. 12 is a diagram showing a code table of a macroblock type and a code table of an 8x8 partition type in B_Picture according to the embodiment of the present invention.
- FIG. 13 is a block diagram showing a general configuration of a video encoding device.
- FIG. 14 is a block diagram showing a general configuration of the main video decoding device.
- FIG. 15 is a diagram illustrating an example of a macroblock type code table in the conventional method.
- FIG. 16 is a diagram showing a configuration example of the universal coding table.
- FIG. 17 is a diagram illustrating a configuration example of a macroblock having a frame structure in a field signal.
- FIG. 18 is a flowchart of a process of switching the coding accuracy of the motion vector according to the block size.
- ⁇ 19 is a flowchart of a process of switching the encoding accuracy of a motion vector with a component of the motion vector according to the embodiment of the present invention.
- FIG. 20 is a flowchart of a process of switching the motion vector coding precision by the motion vector component and the macroblock configuration in the embodiment of the present invention.
- FIG. 21 shows the relationship between the block size and the motion vector in the embodiment of the present invention.
- 13 is a flowchart of a process of switching the encoding accuracy of a motion vector in a minute and a macro block configuration.
- FIG. 22 is another example of a flowchart of a process of switching the coding precision of a motion vector by a block size, a motion vector component, and a macroblock configuration in the embodiment of the present invention.
- FIG. 23 is an example of a flowchart of a process of switching the coding accuracy of a motion vector based on the components of the block size and the motion vector in the embodiment of the present invention.
- c Figure 2 5 is another example of a flow diagram of the processing for switching the coding accuracy of a motion base-vector in Ingredient block size and motion base-vector in the embodiment of the present invention, video according to the invention
- FIG. 4 is a configuration diagram of an embodiment of a motion vector encoding unit of the image encoding device.
- FIG. 26 is a configuration diagram of an embodiment of a decoding unit of a motion vector of the moving picture decoding apparatus according to the present invention.
- FIG. 27 is a configuration diagram of an embodiment of the motion vector encoding unit of the video encoding device according to the present invention.
- FIG. 28 is a configuration diagram of an embodiment of the motion vector decoding unit of the video decoding device according to the present invention.
- FIG. 29 is a configuration diagram of an embodiment of the motion vector coding unit of the video coding device according to the present invention.
- FIG. 30 is a configuration diagram of an embodiment of a motion vector decoding and decoding unit of the moving picture decoding apparatus according to the present invention.
- FIG. 31 is a configuration diagram of an embodiment of a motion vector coding unit of the video coding device according to the present invention.
- FIG. 32 is a configuration diagram of an embodiment of a motion vector decoding unit of the video decoding device according to the present invention.
- FIG. 33 is a block diagram of the motion vector coding unit of the video coding apparatus according to the present invention. It is a lineblock diagram of an example.
- FIG. 34 is a configuration diagram of an embodiment of the decoding unit of the motion vector of the moving picture decoding apparatus according to the present invention.
- FIG. 35 is a configuration diagram of an embodiment of a motion vector encoding unit of the video encoding device according to the present invention.
- FIG. 36 is a configuration diagram of an embodiment of a motion vector decoding unit of the video decoding device according to the present invention.
- FIG. 37 is a configuration diagram of an embodiment of the motion vector encoding unit of the video encoding device according to the present invention.
- FIG. 38 is a configuration diagram of an embodiment of the motion vector decoding unit of the video decoding device according to the present invention.
- FIG. 39 is a configuration diagram of an embodiment of a motion vector coding unit of the video coding device according to the present invention.
- FIG. 40 is a configuration diagram of an embodiment of a decoding unit of a motion vector of the moving picture decoding apparatus according to the present invention.
- FIG. 41 is a configuration diagram of an embodiment of the motion vector encoding unit of the video encoding device according to the present invention.
- FIG. 42 is a configuration diagram of an embodiment of the decoding unit of the motion vector of the moving picture decoding apparatus according to the present invention.
- FIG. 43 is a configuration diagram of an embodiment of a motion vector encoding unit of the video encoding device according to the present invention.
- FIG. 44 is a configuration diagram of an embodiment of a motion vector decoding unit of the video decoding device according to the present invention.
- FIG. 45 is a configuration diagram of an embodiment of a motion vector coding unit of the video coding device according to the present invention.
- FIG. 46 shows the motion vector decoding unit of the video decoding apparatus according to the present invention. It is a lineblock diagram of an example.
- FIG. 47 is a configuration diagram of an embodiment of the motion vector encoding unit of the video encoding device according to the present invention.
- FIG. 48 is a configuration diagram of an embodiment of the decoding unit of the motion vector of the moving picture decoding apparatus according to the present invention.
- Fig. 1 Data syntax of SP-Picture
- Fig. 2 Part of SB-Picture data syntax (CBP15 and later are omitted because they are the same as Fig. 1 after CBP15).
- the description will be made assuming that the data (motion vector accuracy information) 13 does not appear in the data syntax.
- the description of I-picture is omitted because it is not included in the features of the present invention.
- the picture header! 0 includes data related to the reproduction information of the frame such as the image size and the reproduction time.
- picture typell is frame type information, which is a type of I-picture, P-picture, B-picture, and is encoded here.
- Tcoeff chromal7 corresponds to the macroblock data, and the number of macroblocks in the frame is reduced.
- MB typel2 one mode is selected for each macroblock from the macroblock modes shown in Table 95 (P-picture) or Table 96 (B-Picture) in Fig. 15, and the c SKIP mode to be coded is This means that a macroblock image corresponding to the same spatial position is cut out and copied from the decoded image of the frame.
- M and N of IntraMXN in the table indicate the small block size when performing spatial prediction.
- MX N indicates a small block unit when performing motion compensation. Modes 1 to 4 are supported.
- Direct means direct prediction.
- Block 1 and Block 2 identify two small blocks in mode 2 or mode 3 in Fig. 7, and the prediction direction of each small block is Forward (forward prediction), Backward (backward prediction), Bi-directional ( Bidirectional prediction). If you select 8x8 (split) in MB typel2, 8x8 Partition 18 will occur. In other macroblock modes, 8x8 Partition 18 does not occur. In 8x8 Partition 18, one mode is selected for each 8x8 partition from the 8x8 partition modes shown in Table 92 (P_picture) in FIG. 11 or Table 94 (B-Picture) in FIG. 12 and encoded. Intra in the table indicates spatial prediction. MXN indicates a small block unit when performing motion compensation, and corresponds to partition 1 to partition 4 in Fig. 7, respectively. Direct means direct prediction.
- Prediction indicates whether the prediction direction of each small block is Forward (forward prediction), Backward (backward prediction), or Bi-directional (bidirectional prediction).
- MVD14 indicates the encoded data of the differential motion vector in P-Picture. Therefore, this data is not generated when MB typel2 is IntraMXN.
- MVD14 is repeatedly coded for the number of vectors indicated by MB type and 8X8 Partition. For example, in the case of 16 ⁇ 8, the number of differential motion vectors to be encoded is two, and each motion vector is composed of two components, horizontal and vertical, so that four MVDs are consecutive. Encoded.
- MVDFW23 and MVDBW24 indicate the coded data of the differential motion vector in B-Picture.
- MVDFW23 occurs when the macroblock contains a motion vector associated with forward prediction (including bidirectional prediction) indicated by MB type 12 and 8x8 Partition 18.
- MVDBW24 is backward prediction (including bidirectional prediction) indicated by MB type 12 and 8x8 Partition 18. Occurs when the macro block contains the motion vector associated with. Therefore, this data does not occur in the case of MB type12 S IntraMXN or Direct. If MB typel2 is MXN and there is a motion vector associated with forward prediction (including bi-directional prediction), MVDFW23 will perform forward prediction (including bi-directional prediction) indicated by MB type and 8X8 Partition.
- MVDBW24 uses the backward prediction (bidirectional prediction) indicated by MB type and 8X8 Partition. (Including the number of differential motion vectors X2, horizontal and vertical components).
- CBP15 is coded data indicating whether 16 quantized DCT coefficients (significant coefficients) other than 0 are included in the 24 DCT blocks shown in FIG.
- coding the presence / absence information of the significant coefficient for each of the 24 DCT blocks increases the additional information.Therefore, here, as shown in Fig. 6, the 4 DCT blocks are significant in 8x8 block units. The presence or absence of the coefficient is coded.
- Tcoeff lumal6 indicates encoded data of the quantized DCT coefficient in the luminance block 40.
- 16 quantized coefficients are coded in ascending order of number.However, CBP shows that there are no significant coefficients. For blocks, the encoding process is omitted.
- Tcoeff chromal7 indicates encoded data of the quantized DCT coefficient in the luminance block 40.
- 16 quantization coefficients are coded in the order of 4X4 blocks 41-0, 41-2, 41-3, 41-4, 42-0, 42-1, 42-3, 42-4. It is. However, for signal components for which there is no significant coefficient in CBP15, In this case, the encoding process is omitted.
- the coding methods for macroblock-level coded data shown in Figs. 1 and 2 are as follows: Universal VLC (UVLC) using only one type of variable-length code table, fixed-length coding and variable-length coding (each coding An encoding method in which different code tables are prepared for the elements) and an arithmetic encoding (for example, see Non-Patent Document 1) can be considered.
- Table 83 in Figure 16 shows the structure of the UVLC, where the value of Xn is '0' or '1'.
- Table 84 shows an example of an actual variable-length code table.
- each code is replaced by binary data of several bits, and each bit is coded according to the occurrence probability model indicating the probability of occurrence of 0s and 1s. think of.
- This method is called CABAC (Context-based Adaptive Binary Arithmetic Coding) and is described in Witten et al., "Arithmetic Coding for Data Compression", Comm. Of the ACM, 30 (6), 1987, pp. .520-541).
- the present invention is particularly characterized by a motion vector encoder / decoder. That is, the processing of the MV Dprecision section 3 in the data MVD syntax shown in FIGS.
- the configuration and the signal processing method of the MV unit 508 and the motion compensation unit 504 have characteristics.
- the configuration of the other parts and the signal processing method are substantially the same as those of the conventional apparatus and method, and an embodiment of the characteristic portion of the present invention will be described below.
- the motion vector coding unit divides an image into a plurality of coding blocks, and converts an image signal of the coding block and a motion vector in the coding block from a motion vector at an adjacent position.
- the motion vector to be predicted and encoded and the differential motion vector calculated from the predicted motion vector In moving picture coding (hereinafter abbreviated as motion compensation coding of moving pictures), which is coded in units of ticks, several types of differential motion vector precision are prepared, and the differential motion vector is calculated for each decoding block.
- This method selects the accuracy and encodes the information of the selected accuracy and the difference motion vector of the information of the accuracy.
- the small block for which the motion vector is to be encoded is assumed to be 50.
- the motion vector of the three blocks located at the adjacent positions A, B, and C is used as an index to calculate the intermediate value, and the motion vector of the intermediate value is calculated. Is the predicted motion vector. If the block at position C is not coded or located outside the image due to the macroblock position, the motion of the block located at position D instead of position C The vector is used as one of the candidate motion vectors. If the blocks at positions A, B, C, and D do not have a motion vector, the prediction process is performed with the motion vector as the zero vector.
- the above-described prediction method is based on block 51-0 of mode 1 (51) and small blocks 54a-0, 54b-0, 54b-1, 54c-0, 54c-2 of mode 4 (54) in FIG. Pd 54d—0 to 54d-3 Pd Use the same prediction method.
- Figure 8 shows two small blocks (52-0, 52-1) of mode 2 (52) and two small blocks (53-0, 53-1) of mode 3 (53).
- the motion vector of the block located at the base of the arrow is used as the predicted value.
- the motion vector for the chrominance component is not coded, and the motion vector for the luminance component is divided by 2 and used. .
- MV 1/4 pixel accuracy table Assuming that the current motion vector component is quadrupled, and PMV is the quarter-pixel precision representation of the predicted motion vector component quadrupled, the MV value for each accuracy candidate in this representation Is restricted as follows.
- the effect of the motion vector code amount was added to the absolute value sum or square error sum of the prediction error signal in the motion vector estimation.
- Prepare an evaluation function At this time, the optimal evaluation value of each small block is calculated for each accuracy of (MV-PMV), and after the motion vector estimation processing of all small blocks in the macroblock is completed, the small block for each accuracy is calculated.
- the accuracy with which the sum of the evaluation values is small is determined as the accuracy of the difference motion vector.
- the differential motion vector MVD to be encoded is calculated by equation (1).
- MVD (MV-PMV) >> mv_shift (1)
- Mv-shi ft in equation (1) is the value shown in Table 81 in FIG. Equation (1) is obtained by subtracting the predicted motion vector component PMV from the motion vector component MV, and then arithmetically shifting the value to the right by the number indicated by mv—shift to obtain the differential motion vector component MVD. It is shown that.
- the value obtained by subtracting the predicted motion vector component PMV from the motion vector component MV is 1 times a multiple of mv—shift
- the differential motion vector component MVD processed by the arithmetic right bit shift is a multiple of 1 C Picture type power SP-Picture
- mv-shift power S Encoded data element shown in data MVD prec is ions in Fig.
- MVD shown in data MVD 14 Encoded as an encoding element.
- mv-shift is the encoded data element shown in data MVD precisionl3 in Fig. 2
- MVD is data MVDFW23 (when MVD is a forward motion differential motion vector) or data It is encoded as the encoded data element shown in MVDBW24 (when MVD is the differential motion vector for backward prediction).
- MVD precision coding is omitted for macroblocks in which the coded data (MVD14 or MVDFW23 and MVDBW24) of the differential motion vector are not generated.
- P-Picture MB type power S IntraMXN or MB type power S 8x8 (split) with 8x8 partition power of 4 / J blocks S all in intra
- UVLC or arithmetic coding is used as the coding method.
- Table 81 in FIG. 9 are examples of coding codes when UVLC is used.
- the generated value of the predicted motion vector of the coded MB (7C in Fig. 10) is the mv of the MB immediately above (7A in Fig. 10) and the MB directly above (7B in Fig. 10). Encode according to the occurrence probability table calculated from the shift value.
- Equation (2) shows that the value obtained by adding the predicted motion vector component PMV to the value obtained by arithmetically shifting the differential motion vector component MVD left by the number indicated by mv_shift to the decoded motion vector component MV is obtained. Is shown.
- the decoded differential motion vector component MVD is changed from a multiple of 1 to a multiple of l «mv_shift (corresponding to the MV-PMV on the encoding side). PMV By adding, the decoded motion vector component MV is restored to the original accuracy value.
- the method of motion estimation is the same as in the three cases of 1/4 pixel accuracy, 1/2 pixel accuracy, and 1 pixel accuracy, and a description thereof will be omitted.
- the differential motion vector MVD is calculated by equation (1).
- the value obtained by subtracting the predicted motion vector component PMV from the motion vector component MV is 1 times the multiple of mv_shift
- the difference motion vector component MVD obtained by the arithmetic right bit shift processing is a multiple of 1 And is calculated by the above equation (1).
- Mv-shift in equation (1) is the value shown in Table 82 in Fig. 9.
- the processing on the decoding side is also a 1/4 pixel precision candidate for the accuracy of the differential motion vector. This is the same as the three cases of 1/2 pixel accuracy and 1 pixel accuracy.
- the decoded MV value of each motion vector component is reproduced from the decoded mv_shift value, the decoded MVD and the PMV according to the above formula (2). Is done. Equation (2) indicates that the value obtained by adding the predicted motion vector component PMV to the value obtained by arithmetically shifting the differential motion vector component MVD to the left by the number indicated by mv_shift becomes the decoded motion vector component MV. Show.
- the decoded differential motion vector component MVD becomes a multiple of 1 to a multiple of 1 ⁇ mv_shift (corresponding to the MV-PMV on the encoding side).
- the decoded motion vector component MV is restored to the original accuracy value.
- MV value for all small blocks in the macroblock is a multiple of 4 for both horizontal and vertical components. MV.
- MV for which the MV value of all small blocks in a macroblock is a multiple of 2 for both horizontal and vertical components.
- MV 1/4 pixel accuracy: MV for which the MV value of all small blocks in the macroblock is a multiple of 1 for both horizontal and vertical components. Therefore, in order to reflect the effect of this method on the coding characteristics, the effect of the motion vector code amount is added to the sum of the absolute value or the square error of the prediction error signal in the motion vector estimation. Prepare an evaluation function. At this time, the optimal evaluation value of each small block is calculated for each accuracy of the MV, and after the motion vector estimation processing of all small blocks in the macroblock is completed, the evaluation value of the small block for each accuracy is calculated. By calculating the added value of, the accuracy of the difference motion vector can be determined with a small sum of the evaluation values.
- the differential motion vector MVD to be encoded is given by the following equation:
- Equation (3) is a value shown in Table 81 of FIG. Equation (3) is obtained by subtracting the predicted motion vector component PMV from the motion vector component MV by the arithmetic bit shift to the right by the number indicated by mv_shift, and then immediately to the left by the arithmetic bit shift. On the other hand, it shows that the value obtained by performing the second arithmetic right bit shift by the number indicated by mv_shift is the differential motion vector component MVD.
- the continuous right and left bit shift processes for the predicted motion vector component serve to convert the value of the predicted motion vector component PMV into the same value as MV and a multiple of mv_shift. To fulfill. Also, the differential motion vector component MVD becomes a multiple of 1 by the second arithmetic right bit shift process.
- the forecast number converted to a multiple of 1 The value obtained by multiplying the turtle component by 1 mv_shift is equivalent to the “predicted motion vector component converted to a value of 1 ku mv-shift times” in the above equation, and the motion vector component MV is 1 mv_shift Therefore, no information is lost by using the following formula.
- Equation (5) Reproduce the decoded MV value of each motion vector component. Equation (5) is obtained by arithmetically shifting the differential motion vector component MVD to the left by the number indicated by mv-shift, and the predicted motion vector component PMV by the number indicated by mv_shift to the right by the arithmetic bit. Immediately after the shift, the value obtained by adding the arithmetic bit-shifted value to the left indicates that it is the motion vector component MV.
- the predicted vector component converted to a multiple of 1 is The value obtained by multiplying 1 by mv-1 shift is equivalent to the predicted motion vector component converted to a multiple of 1 by mv_shift in the above equation, and the differential motion vector component DMV is a multiple of 1, so No loss of information occurs by using equation (6).
- the candidates for the accuracy of the difference motion vector are 1/8 pixel accuracy, 1/4 pixel accuracy, 1/2 pixel accuracy, and 1 pixel accuracy.
- MV is a value obtained by multiplying the motion vector component of 1/8 pixel precision expression by 8 and PMV. Is a value obtained by multiplying the predicted motion vector component of 1/8 pixel precision expression by 8 times, The value of MV for accuracy detection is limited as follows.
- MV for which the MV value of all small blocks in the macroblock is a multiple of 8 for both horizontal and vertical components.
- the motion estimation method is the same as the three cases of 1/4 pixel accuracy, 1/2 pixel accuracy, and 1 pixel accuracy, and therefore, the encoding is omitted here.
- the differential motion vector MVD to be encoded is given by the following equation:
- MVD (MV-((PMV) mv-shift) ku mv-shif)> mv-shift (3)
- mv-shift is the value shown in Table 82 in Figure 9.
- the formula is the value obtained by subtracting the value of the motion vector component MV from the motion vector component MV by the arithmetic bit shift to the right by the number indicated by mv_shif1: and then immediately to the left by the arithmetic bit bit shift.
- the second calculation It shows that the value obtained by performing the right bit shift becomes the differential motion vector component MVD.
- the successive right and left bit shift processes for the predicted motion vector component convert the value of the predicted motion vector component PMV into the same multiple of l «mv_shift as the MV. Fulfill.
- the second motion right bit shift process results in a value that is a multiple of the differential motion vector component MVD force S1.
- the explanation of the accuracy of the difference motion vector is the same as that of the three cases of 1/4 pixel accuracy, 1/2 pixel accuracy, and 1 pixel accuracy. Is omitted, but the encoding code when UVLC is used follows Table 82 instead of Table 81. However, the number of operation steps can be reduced by simplifying the above equation as follows. This is effective when implementing circuit design. In this equation, the motion vector component MV. And the predicted motion vector component PMV are converted into multiples of 1 before performing the difference processing.
- MVD (MV >> mv_shif t)-(PMV >> mv_shif t) (4)
- the accuracy of the differential motion vector is the same as that of the three cases of 1/4 pixel accuracy, 1/2 pixel accuracy, and 1 pixel accuracy.
- MV (MVD x mv-shift) + ((PMV >> mv-shift) x mv_shift) (5)
- the decoded MV value of each motion vector component is reproduced. This equation is obtained by arithmetically shifting the differential motion vector component MVD to the left by the number indicated by mv_shift to the left, and the predicted motion vector component PMV by the number of mv-shifts to the right by the arithmetic bit shift. Immediately to the left plus the arithmetic bit-shifted value Is the decoded motion vector component MV.
- the continuous right and left bit shift processes for the predicted motion vector component and the process of arithmetically shifting the MVD to the left by the number indicated by the differential motion vector component mv_shi ft are as follows. It plays the role of converting each component into a multiple of mv_shi ft. Therefore, the decoded motion vector component MV is a multiple of the original 1 ⁇ mv ⁇ 1shift. Note that the number of operation steps can be reduced by simplifying this equation as follows. This is effective when implementing circuit design. In this equation, the left arithmetic bit shift is applied to the value obtained by adding the decoded differential motion vector DMV to the predicted motion vector component PMV converted to a multiple of 1 and the number indicated by mv—shi ft.
- the first embodiment adjusts the accuracy of the difference motion vector actually encoded. Therefore, the calculation is simpler than in the second embodiment.
- the accuracy of the motion vector to be encoded is adjusted, so that the search pixel assuming each accuracy does not depend on the prediction vector. Therefore, although when pursuing only coding performance is effective Example 1, the memory access reduction method described later, Example 2 c rather Example 3 is effective>
- the encoding method and the decoding method that are determined in advance according to the block mode without encoding the mv_shift value as much as possible are used.
- This is a method to prevent the code amount of the mv -shift value from increasing in the region where the selectivity of the clock is high.
- encoding and decoding of a differential motion vector are performed with a predefined default motion vector accuracy. By this processing, the encoding and decoding processing of the motion vector accuracy information is omitted.
- the encoding side and the decoding side are set to perform encoding and decoding processing with 1/4 pixel precision or 1/8 pixel precision.
- encoding / decoding of accuracy information is omitted.
- the macroblock of MB type 16x16 is set as the mode that omits the encoding / decoding processing of the accuracy information, but the target can be further extended.
- the macro block with MB type of 16x16 is set as the mode that skips the encoding / decoding processing of the accuracy information, and the number of motion vectors to be encoded is large.
- up to macro blocks with MB types of 16x16, 16x8 and 8x16 may be set as the mode in which encoding of precision information is omitted.
- the default value of the motion vector accuracy can be set in the range of 0-3 (0-2 for 1/4 MC type), like the mv-shift value. Switch by unit.
- the code of the default value shall be included in the header part at the beginning of the data in the case of the bit stream unit, and included in the picture header data in the case of the frame unit. In the case of bit stream unit, if it is determined in advance by the encoding side and the decoding side, it is not necessary to include it in the encoded data.
- the unit may be fixed for the entire frame or fixed for the entire sequence. Fixing the precision selection switching unit simplifies the encoding / decoding process. Therefore, in a simple encoding / decoding method that considers the implementation cost, it is effective to fix the accuracy switching unit in frame units or sequence units.
- the fourth embodiment is a method of uniquely determining a shift value from the surrounding situation. That is, by predicting the mv_shift value of the encoding target macroblock from the adjacent macroblock power, the code amount of the mv-shift value is reduced. For example, the mv-shift value of the current macroblock (macroblock 7C in FIG. 10) is predicted from the mv_shift value of the macroblock (7A) immediately above and the macroblock (7B) immediately above. As a prediction method, there is a method of predicting from the sum of the mv_shift values of 7A and 7B.
- the MB type combination there is a line control means to encode the mv_shift value. Note that this specific example is merely an example, and there are many combinations of the types of adjacent macroblocks used for prediction, the prediction method, the types of MB types for control, and the types of classification at the time of control.
- the amount of reduction in the code amount of the mv-shift value is large by extending the controlling MB type to two or less differential motion vectors, or by extending the control point of the added value to two or less. It becomes.
- the means for switching the coding accuracy of the motion vector in the above-described embodiment includes the following modifications.
- the prediction of the encoding accuracy of the motion vector is 1/8 pixel accuracy, 1/4 pixel accuracy, 1/2 pixel accuracy, and 1 pixel accuracy.
- the present invention is not limited to this. For example, when including up to 1/16 pixel accuracy, or when using two types of 1/2 pixel accuracy and 1 pixel accuracy.
- the encoding accuracy information of the motion vector is encoded in units of macroblocks.
- the encoding accuracy information of the motion vector is switched for each of a plurality of macroblock groups.
- the motion compensation block in the macroblock is divided into four types as shown in FIG. 7.
- the motion vector coding method in this specification uses the motion compensation block division mode. There are no restrictions on the configuration of the motion vector coding block, as it can be implemented even if the type and number of modes are changed.
- the encoding method of the encoded data is UVLC calculation.
- the motion vector coding method described in the present specification which describes the technique coding, can be implemented without being limited to the type of entropy coding.
- the configuration of the UVLC code table and the method of arithmetic coding / decoding are not limited to the tables in the embodiments.
- the default value of the mv_shift value is used for the 16x16, 16x8, and 8x16 modes, but this default value can be applied to a part of the 8x8 (split) Part mode.
- different values may be allocated to the power S for which the default value is set in units of bitstreams or frames, and to each macroblock.
- the default value is set to 1
- 16x8 and 8x16 modes the default value is set to 0.
- the coding accuracy of the differential motion vector (or motion vector) to be applied is specified in units of frame or sequence. It may be encoded and transmitted (in the case of a sequence unit, it is not necessary to encode the accuracy information if the encoding precision in each mode is determined in advance on the encoding side and the decoding side). As described above, fixing the unit for switching the selection of the encoding accuracy simplifies the encoding / decoding process. The implementation cost is further reduced by increasing the mode that defines the default value. In an encoding / decoding system that requires a simpler configuration, it is effective to fix the unit of switching the encoding accuracy for more modes.
- the motion compensator 211 is provided to the motion compensator 211 in accordance with the combination of the MB type and the 8 ⁇ 8 Partion.
- mv Decision processing of 0N / 0FF encoding of shift value
- motion estimation processing considering mv_shift value
- the processing to be conveyed to the heavy pack 206 is added.
- a process of encoding the mv_shift value as MVD precision is added to the multiplexer 206.
- the decoder 501 determines the mv_shift value according to the decoded value of the MB type and 8x8 Partition by 0N / 0FF according to the decoding value of the MB type and 8x8 Partition, and decodes the MVD precision as mv_shift value ⁇ is provided with a processing unit for adding a process of transmitting the mv-shift value to the MV predictor 508. Also provided is a processing sound
- FIG. 25, FIG. 27, FIG. 29, FIG. 31, FIG. 33, FIG. 35, FIG. 37, FIG. 39 and FIG. 41 all show a moving picture coding apparatus according to the present invention.
- FIG. 2 is a block diagram showing a configuration of a main part (encoding unit of a motion vector) of the embodiment.
- Fig. 26, Fig. 28, Fig. 30, Fig. 32, Fig. 34, Fig. 36, Fig. 38, Fig. 40 and Fig. 42 are-Fig. 25, Fig. 27, Fig. 29 respectively , FIG. 31, FIG. 33, FIG. 35, FIG. 37, FIG. 39 and FIG. 41, a main part of an embodiment of the moving picture decoding apparatus according to the present invention corresponding to the coding apparatus shown in FIG.
- FIG. 4 is a block diagram illustrating a configuration.
- FIG. 4 is a block diagram showing a configuration of a main part (encoding section of a motion vector) of the embodiment of FIG.
- the apparatus of FIG. 2 5 and 2 6 are each comprised of circuits to execute the equation (1) and (2) of Example 1, shift (m V _Shift) determining the external input values (motion base-vector (Determined at the time of detection).
- the MV prediction unit 215 uses the memory 101 storing the motion vector MV of each block to create a predicted motion vector PMV.
- the subtractor 102 obtains the difference MV-PMV.
- the right shift processing unit 103 shifts the MV-PMV right.
- a memory 101 storing a motion vector MV, a left shift processing unit 113, and an adder 112 are provided in order to perform inverse processing on the code side. .
- the devices shown in FIGS. 27 and 28 are each configured by a circuit that executes the equations (1) and (2) of the first embodiment, and the shift (mv_shift) value is determined by an external input (when the motion vector is detected). Is determined and controlled by any one of the conditions such as the macro block type.
- Shift value selector 104 If you select a particular plot Kutaipu, m V - select shift value, the control circuit 1 0 5, the moiety force from Mv- shift values of al, corresponding to the block type mv — Process the right shift processing unit 103 with the shift value.
- the apparatus of Figure 2 9 and 3 0 is configured by a circuit for executing each type of each Example 1 (1) ⁇ Pi (2), the type of shift (m V _Shift) values determined macro block It is determined and controlled only by such conditions.
- the apparatus shown in FIGS. 31 and 32 is composed of a circuit for executing equations (3) and (5) of the second embodiment, respectively, and the shift (mv-shift) value is determined by an external input (motion vector detection). Is sometimes controlled).
- the precision correction processing unit 106 performs the processing of (PMV >> mv-shift) ⁇ mv_shift in Expression (3).
- the devices of FIGS. 33 and 34 are constituted by circuits for executing equations (3) and (5) of the second embodiment, respectively, and the determination of the shift (mv-shift) value is performed by an external input (motion vector detection). Is sometimes determined) and controlled by one of the conditions such as the type of macroblock.
- the devices of FIGS. 35 and 36 are each configured by a circuit that executes the equations (3) and (5) of the second embodiment, and the determination of the shift (mv_shift) value is controlled only by the macroblock type and the like. Is done.
- the devices shown in FIGS. 37 and 38 are constituted by circuits for executing equations (4) and (6) of the second embodiment, respectively, and the shift (mv-shift) value is determined by an external input (movement). (Determined at the time of vector detection).
- the devices shown in FIG. 39 and FIG. 4 ⁇ are each configured by a circuit that executes the equations (4) and (6) of the second embodiment, and the shift (mv—shi ft) value is determined by an external input (moving solid). (Determined at the time of detection of a torque) and macro block type.
- the apparatus of FIG. 41 and FIG. 42 each include a circuit for executing equations (4) and (6) of the second embodiment, and the determination of the shift (mv_shift) value is based on the conditions such as the type of the macro block. Determined and controlled by either.
- FIG. 18 is a processing flowchart of the encoding method according to the fifth embodiment.
- the fifth embodiment is an encoding / decoding method for switching the accuracy of an encoding motion vector in order to solve a problem that a memory access area is expanded by motion compensation using a filter having a long tap number.
- the predicted values generated using a long filter are only predicted pixels with pixel accuracy or less. Therefore, the memory access area can be reduced by limiting the motion vector coding accuracy to integer pixels. However, limiting the encoding accuracy of the motion vector to integer pixels reduces the prediction performance. Therefore, in the fifth embodiment, the pixel accuracy is limited in accordance with the degree to which the expansion of the memory access range affects device mounting or software mounting. In particular, the motion vector coding accuracy is limited only when the block size during motion compensation is small.
- the encoding apparatus of the present embodiment first determines the block size as shown in FIG. 18 (processing 801), and performs mv_shift Select a value.
- the 8x8 Partition 2 (8x4) t 8x8 Partition 3 (4x8) has 1/2 pixel precision (processing 8 1 1)
- the 8x8 Partition 3 (4x4) has integer pixel precision (processing 812).
- mv_shift is 1 for 8x4 and 4x8 pixel blocks
- mv—shift is 2 for 4x4 pixel blocks
- mv—shift is 0 for blocks of other pixel sizes.
- the above-mentioned precision switching is applied only to B-picture, using P-picture and B-picture, and in P-picture, motion compensation is performed at 1/4 pixel accuracy regardless of the block size. May be performed.
- the block size for performing motion compensation with integer pixel precision is only 4x4-but this may be extended to 8x4 and 4x8.
- mv-shift is set to 2 for blocks of 8x4, 4x8, and 4x4 pixel sizes
- mv-shift is set to 0 for blocks of other pixel sizes.
- the encoding unit and the decoding unit of the motion vector in the moving picture encoding apparatus and the decoding apparatus according to the present embodiment are configured in the same manner as in FIGS. 41 and 42, respectively.
- FIG. 19, FIG. 45 and FIG. 46 are a flowchart of an encoding method according to a sixth embodiment of the present invention, a main part configuration diagram of an encoding device, and a main configuration diagram of a decoding device, respectively.
- This embodiment also solves the problem that the memory access area is enlarged by motion compensation using a filter with a long tap number. Encode components with different encoding precisions to limit the memory access range.
- the shift selecting unit 124 of the encoding device determines whether or not the motion vector is a vertical component (process 801), and selects mv-1 shi ft.
- the type adaptation unit 124 drives the right shift processing unit 103 with the selected mv-shift to perform motion compensation.
- the vertical component of the motion vector is set to integer pixel precision (processing 812)
- the horizontal component is set to 1/4 pixel precision (processing 813)
- the vertical component is encoded.
- the m V _ S hift when / decrypt 2 the mv one shift at the time of encoding / decoding a horizontal component to zero.
- the method of the present embodiment for limiting the coding accuracy of the vertical component of the motion vector, which is greatly affected by the expansion of the memory access range, has a large effect.
- Image data is generally stored in memory in raster scan order from upper left to lower right. Therefore, when extending the access range from a certain pixel to an adjacent pixel, one pixel in the horizontal direction means one pixel, but one pixel in the vertical direction means one line. Therefore, the number of memory accesses in the vertical direction is larger than that in the horizontal direction.
- the horizontal memory Since the expansion of the access range occurs in word units or dword units, the effect is not so large.
- the method of limiting the coding accuracy of the vertical motion vector is effective from the viewpoint of reducing the memory access range.
- the processing as shown in FIG. 23 may be performed by combining the methods of the fifth and sixth embodiments. That is, the shift selection unit 124 calculates the vertical component of the motion vector only for blocks of 8x4, 4x8, and 4x4 pixel size. Limit to integer pixels. Furthermore, means for switching the vector components that limit the range according to the block size is also effective.
- the influence of the expansion of the memory access area has a larger effect as the block size is smaller, but this can also be applied to the aspect ratio of the block.
- the ratio of the vertical and horizontal sizes of the blocks is different, the influence of the expansion of the access range increases as the size of the component decreases.
- FIG. 24, FIG. 43 and FIG. 44 are a processing flow diagram of an encoding method according to a seventh embodiment of the present invention, a main configuration diagram of an encoding device, and a main configuration diagram of a decoding device, respectively.
- the shift value selecting section 124 of the encoding device (FIG. 43) and the decoding device (FIG. 44) has a block size of 8 ⁇ 4 (vertical X horizontal). ),
- FIG. 20, FIG. 47 and FIG. 48 are respectively a flowchart of an encoding method processing according to the eighth embodiment of the present invention, a main configuration diagram of an encoding device, and a main configuration diagram of a decoding device. It is.
- This embodiment is a method of limiting the vertical component of a motion vector to even pixel accuracy when an input image is an interlace signal and a macroblock has a frame structure in the seventh embodiment.
- FIG. 17 shows a macroblock when encoding / decoding an interlaced signal in a frame structure.
- the solid line in the figure indicates an odd field (top field), and the broken line indicates an even field (bottom field).
- the odd-valued motion vector is obtained by generating the predicted value on the odd field from the even field on the reference image and generating the predicted value on the even field from the odd field on the reference image.
- the display time is different between the odd field and the even field, so that the odd vertical component vector does not actually occur.
- the vertical component of the motion vector is limited to even pixel accuracy when the input image is an interlaced signal and the macroblock has a frame structure.
- mv-shift when encoding / decoding the vertical component is set to 3 (Fig. 20 Processing 803, 814), and mv_sh i when encoding / decoding the horizontal component is used.
- ft is set to 0 (FIG. 20, processing 813).
- FIG. 21 processing 801 only blocks of 8 ⁇ 4, 4 ⁇ 8 and 4 ⁇ 4 pixel sizes are used (FIG. 21 processing 801).
- the vertical component of the motion vector may be limited to integer pixels.
- an encoding method for interlaced signals there is an adaptive method in which a frame structure, a field structure, and a frame structure and a field structure are switched for each macroblock, and the encoding accuracy of a motion vector is determined according to the method. You may change the selection method.
- the encoding method of a frame structure may be the same regardless of signal scanning.
- the frame structure and the field structure Performs encoding / decoding of small-size blocks with the vertical motion vector encoding precision limited to integer pixel precision (Fig. 20 and Fig. 21 and ⁇ 22 processing 812).
- the coding accuracy of the vertical motion vector is limited to even pixel accuracy in the frame structure for small-sized blocks (see Fig. 20 and Fig. 20).
- encoding / decoding is performed only with integer pixel precision (the processing 812 in FIGS. 20 and 21).
- the processing 814 in Fig. 22 the coding accuracy of the vertical motion vector is limited to even pixel accuracy regardless of the block size (processing 814 in Fig. 22), and in the field structure, the size of the block is small.
- the encoding / decoding may be limited to the integer pixel precision (FIG. 22, process 801 and process 812).
- the encoding / decoding processing of the mv_shift value is not required. Therefore, the MVD prediction l 3 in FIGS. 1 and 2 and the mv_shi f1: in FIGS. 11 and 12 are not necessary. Also, in FIG. 13, there is no need to transmit the mv_shift value from the motion compensator 211 to the multiplexer 206. Further, in FIG. 14, there is no need to perform a process of transmitting the mv_shift value from the code decoder 501 to the MV predictor 508. The mv—shi ft value is determined by the MV predictor 508 according to the prediction mode.
- the memory access range reduction method according to the eighth embodiment uses a means for switching the accuracy between the coding motion vector and the prediction motion vector. Is limited to integer pixels. "
- Limiting the coding accuracy of the vertical component of the motion vector to integer pixels or even pixels '' means the accuracy of motion compensation, the type of motion vector coding accuracy to be prepared, and the type of motion vector.
- the block size for switching the encoding precision is not limited to the above embodiment.
- a method that limits the motion vector to 1/2 pixel precision instead of integer pixel precision is also proposed. It can be realized by the invention.
- the present invention is an apparatus and a system for transmitting, storing, and displaying moving images, and is used in an industrial field that requires a technology for encoding digital information with a smaller code amount of image information.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Description
Claims
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020047002817A KR100924850B1 (ko) | 2002-01-24 | 2003-01-24 | 동화상 신호의 부호화 방법 및 복호화 방법 |
US10/488,030 US8175159B2 (en) | 2002-01-24 | 2003-01-24 | Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus |
JP2003563227A JP4401783B2 (ja) | 2002-01-24 | 2003-01-24 | 動画像の符号化方法 |
EP03703048A EP1469682A4 (en) | 2002-01-24 | 2003-01-24 | SIGNAL CODING METHOD FOR MOVABLE IMAGES, DECODING METHOD, CODING DEVICE AND DECODING DEVICE |
US12/073,757 US8194746B2 (en) | 2002-01-24 | 2008-03-10 | Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus |
US12/073,756 US8233542B2 (en) | 2002-01-24 | 2008-03-10 | Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus |
Applications Claiming Priority (10)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2002-14974 | 2002-01-24 | ||
JP2002014974 | 2002-01-24 | ||
JP2002036130 | 2002-02-14 | ||
JP2002-36130 | 2002-02-14 | ||
JP2002093541 | 2002-03-29 | ||
JP2002-93541 | 2002-03-29 | ||
JP2002267555 | 2002-09-13 | ||
JP2002-267555 | 2002-09-13 | ||
JP2002330209 | 2002-11-14 | ||
JP2002-330209 | 2002-11-14 |
Related Child Applications (3)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10488030 A-371-Of-International | 2003-01-24 | ||
US12/073,757 Division US8194746B2 (en) | 2002-01-24 | 2008-03-10 | Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus |
US12/073,756 Division US8233542B2 (en) | 2002-01-24 | 2008-03-10 | Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2003063503A1 true WO2003063503A1 (en) | 2003-07-31 |
Family
ID=27617870
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2003/000664 WO2003063503A1 (en) | 2002-01-24 | 2003-01-24 | Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus |
Country Status (7)
Country | Link |
---|---|
US (2) | US8233542B2 (ja) |
EP (1) | EP1469682A4 (ja) |
JP (2) | JP4401783B2 (ja) |
KR (2) | KR100962759B1 (ja) |
CN (1) | CN1265649C (ja) |
HK (1) | HK1095461A1 (ja) |
WO (1) | WO2003063503A1 (ja) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004008772A1 (ja) * | 2002-07-15 | 2004-01-22 | Mitsubishi Denki Kabushiki Kaisha | 画像符号化装置、画像符号化方法、画像復号装置、画像復号方法、および通信装置 |
JP2007521696A (ja) * | 2003-10-09 | 2007-08-02 | トムソン ライセンシング | エラー・コンシールメントのための直接モード導出プロセス |
JP2007532036A (ja) * | 2003-09-07 | 2007-11-08 | マイクロソフト コーポレーション | インターレース・ビデオの進歩した双方向予測コーディング |
JP2008048200A (ja) * | 2006-08-17 | 2008-02-28 | Fujitsu Ltd | フレーム間予測処理装置、フレーム間予測方法、画像符号化装置及び画像復号装置 |
JP2008515298A (ja) * | 2004-09-29 | 2008-05-08 | ▲騰▼▲訊▼科技(深▲セン▼)有限公司 | 動き推定の実現方法 |
WO2012008162A1 (ja) * | 2010-07-15 | 2012-01-19 | パナソニック株式会社 | 画像復号化方法、画像符号化方法、画像復号化装置、画像符号化装置、プログラムおよび集積回路 |
JP2013509020A (ja) * | 2009-10-14 | 2013-03-07 | トムソン ライセンシング | 動き情報の適応型符号化方法及び装置 |
JP2013509063A (ja) * | 2009-10-20 | 2013-03-07 | トムソン ライセンシング | 画像シーケンスのブロックを符号化する方法およびこのブロックを再構成する方法 |
JP2013543713A (ja) * | 2010-10-13 | 2013-12-05 | クゥアルコム・インコーポレイテッド | ビデオ符号化のための適応動きベクトル解像度信号伝達 |
US8917768B2 (en) | 2003-07-18 | 2014-12-23 | Microsoft Corporation | Coding of motion vector information |
US9237355B2 (en) | 2010-02-19 | 2016-01-12 | Qualcomm Incorporated | Adaptive motion resolution for video coding |
WO2018110203A1 (ja) * | 2016-12-16 | 2018-06-21 | シャープ株式会社 | 動画像復号装置、および動画像符号化装置 |
JP2018533298A (ja) * | 2015-09-24 | 2018-11-08 | エルジー エレクトロニクス インコーポレイティド | 映像コーディングシステムにおけるamvrに基づく映像コーディング方法及び装置 |
JP2019169964A (ja) * | 2014-10-07 | 2019-10-03 | クゥアルコム・インコーポレイテッドQualcomm Incorporated | イントラbcとインターの統合 |
US10531113B2 (en) | 2014-10-31 | 2020-01-07 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
JP2020523954A (ja) * | 2017-06-12 | 2020-08-06 | ホアウェイ・テクノロジーズ・カンパニー・リミテッド | 動きベクトル(mv)精度の選択およびシグナリング |
Families Citing this family (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN100546385C (zh) * | 2002-01-24 | 2009-09-30 | 株式会社日立制作所 | 运动图像的编码方法、解码方法、编码装置及解码装置 |
CN1578477B (zh) * | 2003-07-16 | 2011-05-04 | 三星电子株式会社 | 用于色彩图像的视频编码/解码装置和方法 |
KR100694098B1 (ko) | 2005-04-04 | 2007-03-12 | 한국과학기술원 | 산술 복호 방법 및 그 장치 |
KR100827093B1 (ko) * | 2006-10-13 | 2008-05-02 | 삼성전자주식회사 | 영상 부호화 방법 및 장치 |
KR101365574B1 (ko) * | 2007-01-29 | 2014-02-20 | 삼성전자주식회사 | 영상 부호화 방법 및 장치, 복호화 방법 및 장치 |
US8942289B2 (en) * | 2007-02-21 | 2015-01-27 | Microsoft Corporation | Computational complexity and precision control in transform-based digital media codec |
KR100939917B1 (ko) * | 2008-03-07 | 2010-02-03 | 에스케이 텔레콤주식회사 | 움직임 예측을 통한 부호화 시스템 및 움직임 예측을 통한부호화 방법 |
WO2009115901A2 (en) * | 2008-03-19 | 2009-09-24 | Nokia Corporation | Combined motion vector and reference index prediction for video coding |
JP5422168B2 (ja) * | 2008-09-29 | 2014-02-19 | 株式会社日立製作所 | 動画像符号化方法および動画像復号化方法 |
US20110235931A1 (en) * | 2008-12-08 | 2011-09-29 | Tomoyuki Yamamoto | Image encoder and image decoder |
TW201041404A (en) * | 2009-03-06 | 2010-11-16 | Sony Corp | Image processing device and method |
WO2011021913A2 (ko) * | 2009-08-21 | 2011-02-24 | 에스케이텔레콤 주식회사 | 차분 움직임 벡터의 정밀도를 고려한 움직임 벡터 부호화/복호화 방법 및 장치, 및 그를 위한 영상처리 장치 및 방법 |
TW201515445A (zh) * | 2010-04-01 | 2015-04-16 | Sony Corp | 圖像處理方法 |
RU2549157C1 (ru) * | 2010-04-23 | 2015-04-20 | М энд К ХОЛДИНГС ИНК. | Устройство для кодирования движущегося изображения |
CN102378000B (zh) * | 2010-08-13 | 2013-07-17 | 炬力集成电路设计有限公司 | 一种视频解码装置和方法 |
US20120051431A1 (en) * | 2010-08-25 | 2012-03-01 | Qualcomm Incorporated | Motion direction based adaptive motion vector resolution signaling for video coding |
KR101479130B1 (ko) * | 2010-10-18 | 2015-01-07 | 에스케이 텔레콤주식회사 | 차분 움직임벡터 부호화/복호화 장치 및 방법, 및 그것을 이용한 영상 부호화/복호화 장치 및 방법 |
US9749657B2 (en) * | 2011-01-21 | 2017-08-29 | Sharp Kabushiki Kaisha | Buffer compression for motion vector competition |
JP2013172323A (ja) | 2012-02-21 | 2013-09-02 | Toshiba Corp | 動き検出装置、画像処理装置および画像処理システム |
GB2512829B (en) * | 2013-04-05 | 2015-05-27 | Canon Kk | Method and apparatus for encoding or decoding an image with inter layer motion information prediction according to motion information compression scheme |
US9774881B2 (en) * | 2014-01-08 | 2017-09-26 | Microsoft Technology Licensing, Llc | Representing motion vectors in an encoded bitstream |
US9749642B2 (en) | 2014-01-08 | 2017-08-29 | Microsoft Technology Licensing, Llc | Selection of motion vector precision |
US9942560B2 (en) | 2014-01-08 | 2018-04-10 | Microsoft Technology Licensing, Llc | Encoding screen capture data |
CN107809642B (zh) * | 2015-02-16 | 2020-06-16 | 华为技术有限公司 | 用于视频图像编码和解码的方法、编码设备和解码设备 |
CN109218733B (zh) * | 2017-06-30 | 2022-03-29 | 华为技术有限公司 | 一种确定预测运动矢量预测的方法以及相关设备 |
KR20210115052A (ko) | 2017-07-07 | 2021-09-24 | 삼성전자주식회사 | 적응적 움직임 벡터 해상도로 결정된 움직임 벡터의 부호화 장치 및 부호화 방법, 및 움직임 벡터의 복호화 장치 및 복호화 방법 |
EP3691274A4 (en) | 2017-09-28 | 2021-04-28 | Samsung Electronics Co., Ltd. | CODING METHOD AND DEVICE FOR IT AND DECODING METHOD AND DEVICE FOR IT |
GB2589221B (en) | 2018-06-19 | 2023-03-22 | Beijing Bytedance Network Tech Co Ltd | Mode dependent MVD precision set |
CN110891175B (zh) | 2018-09-08 | 2023-04-07 | 北京字节跳动网络技术有限公司 | 视频编解码和解码中的仿射模式 |
WO2020058886A1 (en) | 2018-09-19 | 2020-03-26 | Beijing Bytedance Network Technology Co., Ltd. | Fast algorithms for adaptive motion vector resolution in affine mode |
CN113498609B (zh) * | 2018-12-31 | 2023-06-20 | 北京达佳互联信息技术有限公司 | 用于视频编解码的依赖图片分辨率的配置 |
CN111416982B (zh) * | 2019-01-04 | 2024-06-07 | 北京三星通信技术研究有限公司 | 对运动矢量信息进行编/解码的方法及装置 |
CN113412623A (zh) | 2019-01-31 | 2021-09-17 | 北京字节跳动网络技术有限公司 | 记录仿射模式自适应运动矢量分辨率的上下文 |
CN113661709A (zh) * | 2019-03-27 | 2021-11-16 | 北京字节跳动网络技术有限公司 | 仿射高级运动矢量预测中的运动信息精度对齐 |
KR20220047729A (ko) * | 2019-06-20 | 2022-04-19 | 인터디지털 브이씨 홀딩스 프랑스 에스에이에스 | 비디오 인코딩 및 디코딩을 위한 움직임 벡터 처리 |
JP7323710B2 (ja) * | 2019-09-13 | 2023-08-08 | 北京字節跳動網絡技術有限公司 | 映像符号化における重み付けサンプルの双予測 |
CN111050182B (zh) * | 2019-12-27 | 2022-02-18 | 浙江大华技术股份有限公司 | 运动矢量的预测方法、视频编码方法及相关设备、装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09154140A (ja) * | 1995-11-29 | 1997-06-10 | Nec Corp | 動画像の動き補償回路 |
JPH1146364A (ja) * | 1997-07-28 | 1999-02-16 | Victor Co Of Japan Ltd | 動き補償符号化装置、復号化装置、符号化方法及び復号化方法 |
JPH1188888A (ja) * | 1997-06-25 | 1999-03-30 | Nippon Telegr & Teleph Corp <Ntt> | 動きベクトル予測符号化方法および動きベクトル復号方法、予測符号化装置および復号装置、並びに、動きベクトルの予測符号化プログラムおよび復号プログラムを記録した記録媒体 |
JPH11262018A (ja) * | 1998-03-09 | 1999-09-24 | Victor Co Of Japan Ltd | 動き補償符号化装置、動き補償符号化方法、及び動き補償符号記録媒体 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5428396A (en) * | 1991-08-03 | 1995-06-27 | Sony Corporation | Variable length coding/decoding method for motion vectors |
FR2725577B1 (fr) * | 1994-10-10 | 1996-11-29 | Thomson Consumer Electronics | Procede de codage ou de decodage de vecteurs mouvement et dispositif de codage ou de decodage mettant en oeuvre ledit procede |
US5731840A (en) | 1995-03-10 | 1998-03-24 | Kabushiki Kaisha Toshiba | Video coding/decoding apparatus which transmits different accuracy prediction levels |
JP4142180B2 (ja) * | 1998-10-29 | 2008-08-27 | 富士通株式会社 | 動きベクトル符号化装置および復号装置 |
US6738423B1 (en) | 2000-01-21 | 2004-05-18 | Nokia Mobile Phones Ltd. | Method for encoding and decoding video information, a motion compensated video encoder and a corresponding decoder |
KR100359115B1 (ko) | 2000-05-24 | 2002-11-04 | 삼성전자 주식회사 | 영상 코딩 방법 |
US6859494B2 (en) | 2001-07-27 | 2005-02-22 | General Instrument Corporation | Methods and apparatus for sub-pixel motion estimation |
US6950469B2 (en) | 2001-09-17 | 2005-09-27 | Nokia Corporation | Method for sub-pixel value interpolation |
-
2003
- 2003-01-24 CN CNB038009390A patent/CN1265649C/zh not_active Expired - Fee Related
- 2003-01-24 JP JP2003563227A patent/JP4401783B2/ja not_active Expired - Fee Related
- 2003-01-24 WO PCT/JP2003/000664 patent/WO2003063503A1/ja active Application Filing
- 2003-01-24 KR KR1020077023512A patent/KR100962759B1/ko not_active IP Right Cessation
- 2003-01-24 EP EP03703048A patent/EP1469682A4/en not_active Withdrawn
- 2003-01-24 KR KR1020047002817A patent/KR100924850B1/ko not_active IP Right Cessation
-
2007
- 2007-03-07 HK HK07102503A patent/HK1095461A1/xx not_active IP Right Cessation
-
2008
- 2008-03-10 US US12/073,756 patent/US8233542B2/en not_active Expired - Fee Related
- 2008-03-10 US US12/073,757 patent/US8194746B2/en not_active Expired - Fee Related
-
2009
- 2009-04-28 JP JP2009108955A patent/JP5042267B2/ja not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09154140A (ja) * | 1995-11-29 | 1997-06-10 | Nec Corp | 動画像の動き補償回路 |
JPH1188888A (ja) * | 1997-06-25 | 1999-03-30 | Nippon Telegr & Teleph Corp <Ntt> | 動きベクトル予測符号化方法および動きベクトル復号方法、予測符号化装置および復号装置、並びに、動きベクトルの予測符号化プログラムおよび復号プログラムを記録した記録媒体 |
JPH1146364A (ja) * | 1997-07-28 | 1999-02-16 | Victor Co Of Japan Ltd | 動き補償符号化装置、復号化装置、符号化方法及び復号化方法 |
JPH11262018A (ja) * | 1998-03-09 | 1999-09-24 | Victor Co Of Japan Ltd | 動き補償符号化装置、動き補償符号化方法、及び動き補償符号記録媒体 |
Cited By (38)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8401079B2 (en) | 2002-07-15 | 2013-03-19 | Mitsubishi Denki Kabushiki Kaisha | Image coding apparatus, image coding method, image decoding apparatus, image decoding method and communication apparatus |
US8045616B2 (en) | 2002-07-15 | 2011-10-25 | Mitsubishi Denki Kabushiki Kaisha | Image coding device, image coding method, image decoding device, image decoding method and communication apparatus |
WO2004008772A1 (ja) * | 2002-07-15 | 2004-01-22 | Mitsubishi Denki Kabushiki Kaisha | 画像符号化装置、画像符号化方法、画像復号装置、画像復号方法、および通信装置 |
US8917768B2 (en) | 2003-07-18 | 2014-12-23 | Microsoft Corporation | Coding of motion vector information |
US9148668B2 (en) | 2003-07-18 | 2015-09-29 | Microsoft Technology Licensing, Llc | Coding of motion vector information |
JP2007532036A (ja) * | 2003-09-07 | 2007-11-08 | マイクロソフト コーポレーション | インターレース・ビデオの進歩した双方向予測コーディング |
JP2011139525A (ja) * | 2003-09-07 | 2011-07-14 | Microsoft Corp | インターレース・ビデオの進歩した双方向予測コーディング |
JP4763607B2 (ja) * | 2003-09-07 | 2011-08-31 | マイクロソフト コーポレーション | インターレース・ビデオの進歩した双方向予測コーディング |
JP2007521696A (ja) * | 2003-10-09 | 2007-08-02 | トムソン ライセンシング | エラー・コンシールメントのための直接モード導出プロセス |
US8804830B2 (en) | 2004-09-29 | 2014-08-12 | Tencent Technology (Shenzhen) Company Limited | Method for performing motion estimation |
JP2008515298A (ja) * | 2004-09-29 | 2008-05-08 | ▲騰▼▲訊▼科技(深▲セン▼)有限公司 | 動き推定の実現方法 |
JP4669517B2 (ja) * | 2004-09-29 | 2011-04-13 | ▲騰▼▲訊▼科技(深▲セン▼)有限公司 | 動き推定の実現方法 |
JP2008048200A (ja) * | 2006-08-17 | 2008-02-28 | Fujitsu Ltd | フレーム間予測処理装置、フレーム間予測方法、画像符号化装置及び画像復号装置 |
JP4712643B2 (ja) * | 2006-08-17 | 2011-06-29 | 富士通セミコンダクター株式会社 | フレーム間予測処理装置、フレーム間予測方法、画像符号化装置及び画像復号装置 |
JP2020188483A (ja) * | 2009-10-14 | 2020-11-19 | インターデジタル マディソン パテント ホールディングス | 動き情報の適応型符号化方法及び装置 |
JP2013509020A (ja) * | 2009-10-14 | 2013-03-07 | トムソン ライセンシング | 動き情報の適応型符号化方法及び装置 |
JP2016167837A (ja) * | 2009-10-14 | 2016-09-15 | トムソン ライセンシングThomson Licensing | 動き情報の適応型符号化方法及び装置 |
JP2018067949A (ja) * | 2009-10-14 | 2018-04-26 | トムソン ライセンシングThomson Licensing | 動き情報の適応型符号化方法及び装置 |
JP7179037B2 (ja) | 2009-10-14 | 2022-11-28 | インターデジタル マディソン パテント ホールディングス, エスアーエス | 動き情報の適応型符号化方法及び装置 |
JP2013509063A (ja) * | 2009-10-20 | 2013-03-07 | トムソン ライセンシング | 画像シーケンスのブロックを符号化する方法およびこのブロックを再構成する方法 |
US10142650B2 (en) | 2009-10-20 | 2018-11-27 | Interdigital Madison Patent Holdings | Motion vector prediction and refinement using candidate and correction motion vectors |
US9237355B2 (en) | 2010-02-19 | 2016-01-12 | Qualcomm Incorporated | Adaptive motion resolution for video coding |
WO2012008162A1 (ja) * | 2010-07-15 | 2012-01-19 | パナソニック株式会社 | 画像復号化方法、画像符号化方法、画像復号化装置、画像符号化装置、プログラムおよび集積回路 |
JP2013543713A (ja) * | 2010-10-13 | 2013-12-05 | クゥアルコム・インコーポレイテッド | ビデオ符号化のための適応動きベクトル解像度信号伝達 |
JP2019169964A (ja) * | 2014-10-07 | 2019-10-03 | クゥアルコム・インコーポレイテッドQualcomm Incorporated | イントラbcとインターの統合 |
US10531113B2 (en) | 2014-10-31 | 2020-01-07 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
US10848780B2 (en) | 2014-10-31 | 2020-11-24 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
US11483584B2 (en) | 2014-10-31 | 2022-10-25 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
US11818387B2 (en) | 2014-10-31 | 2023-11-14 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
US11818389B2 (en) | 2014-10-31 | 2023-11-14 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
US11818388B2 (en) | 2014-10-31 | 2023-11-14 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
US11831904B2 (en) | 2014-10-31 | 2023-11-28 | Samsung Electronics Co., Ltd. | Method and device for encoding/decoding motion vector |
JP2018533298A (ja) * | 2015-09-24 | 2018-11-08 | エルジー エレクトロニクス インコーポレイティド | 映像コーディングシステムにおけるamvrに基づく映像コーディング方法及び装置 |
WO2018110203A1 (ja) * | 2016-12-16 | 2018-06-21 | シャープ株式会社 | 動画像復号装置、および動画像符号化装置 |
JP2020523954A (ja) * | 2017-06-12 | 2020-08-06 | ホアウェイ・テクノロジーズ・カンパニー・リミテッド | 動きベクトル(mv)精度の選択およびシグナリング |
US11223847B2 (en) | 2017-06-12 | 2022-01-11 | Huawei Technologies Co., Ltd. | Selection and signaling of motion vector (MV) precisions |
US11272207B2 (en) | 2017-06-12 | 2022-03-08 | Futurewei Technologies, Inc. | Selection and signaling of motion vector (MV) precisions |
US11843800B2 (en) | 2017-06-12 | 2023-12-12 | Huawei Technologies Co., Ltd. | Selection and signaling of motion vector (MV) precisions |
Also Published As
Publication number | Publication date |
---|---|
JP2009171608A (ja) | 2009-07-30 |
US8233542B2 (en) | 2012-07-31 |
KR20040075317A (ko) | 2004-08-27 |
US20080310511A1 (en) | 2008-12-18 |
CN1265649C (zh) | 2006-07-19 |
KR20070108571A (ko) | 2007-11-12 |
KR100924850B1 (ko) | 2009-11-02 |
US20080165856A1 (en) | 2008-07-10 |
CN1550110A (zh) | 2004-11-24 |
EP1469682A1 (en) | 2004-10-20 |
HK1095461A1 (en) | 2007-05-04 |
JPWO2003063503A1 (ja) | 2005-05-26 |
EP1469682A4 (en) | 2010-01-27 |
US8194746B2 (en) | 2012-06-05 |
JP5042267B2 (ja) | 2012-10-03 |
KR100962759B1 (ko) | 2010-06-09 |
JP4401783B2 (ja) | 2010-01-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4401783B2 (ja) | 動画像の符号化方法 | |
JP6513847B2 (ja) | 動画像符号化方法 | |
JP4724351B2 (ja) | 画像符号化装置、画像符号化方法、画像復号装置、画像復号方法、および通信装置 | |
WO2005081541A1 (ja) | 画像情報符号化装置および画像情報符号化方法 | |
JP4355319B2 (ja) | 動画像復号化方法 | |
US8175159B2 (en) | Moving picture signal coding method, decoding method, coding apparatus, and decoding apparatus | |
JP4895995B2 (ja) | 動画像復号化方法 | |
CN100493200C (zh) | 运动图像的编码方法、解码方法、编码装置及解码装置 | |
JP2004048522A (ja) | 動画像の符号化方法及び復号化方法 | |
JP4363484B2 (ja) | 動画像復号化方法 | |
JP2008118701A (ja) | 動画像復号化方法 | |
JP5017087B2 (ja) | 動画像復号化方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CN JP KR US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PT SE SI SK TR |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
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: 2003563227 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2003703048 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020047002817 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 20038009390 Country of ref document: CN Ref document number: 10488030 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 2003703048 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020077023512 Country of ref document: KR |