WO2009088038A1 - 画像符号化装置、画像復号装置、画像符号化方法、および画像復号方法 - Google Patents
画像符号化装置、画像復号装置、画像符号化方法、および画像復号方法 Download PDFInfo
- Publication number
- WO2009088038A1 WO2009088038A1 PCT/JP2009/050142 JP2009050142W WO2009088038A1 WO 2009088038 A1 WO2009088038 A1 WO 2009088038A1 JP 2009050142 W JP2009050142 W JP 2009050142W WO 2009088038 A1 WO2009088038 A1 WO 2009088038A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- motion vector
- prediction
- image
- unit
- encoding
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/103—Selection of coding mode or of prediction mode
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/117—Filters, e.g. for pre-processing or post-processing
-
- 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
- H04N19/46—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
- 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
- H04N19/60—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 an image encoding device, an image decoding device, an image encoding method, and an image decoding method used for image compression encoding technology, compressed image data transmission technology, and the like.
- a standardized input signal format called a 4: 2: 0 format has been used as a compression target signal.
- 4: 2: 0 means that RGB color video signals are converted into luminance components (Y) and two color difference components (Cb, Cr), and the number of samples of the color difference components in both the horizontal and vertical directions is halved. Reduced format. Since the color difference component is less visible than the luminance component, an international standard video encoding method (Non-Patent Document 1) such as MPEG-4 AVC / H.264 (hereinafter AVC) encodes in this way.
- AVC MPEG-4 AVC / H.264
- the color difference component is encoded with the same sample as the luminance component without down-sampling for the purpose of accurately reproducing the color expression at the time of content production at the time of screening 4: 4: 4.
- a direct encoding method in the format is recommended.
- FIG. 9 shows the difference between the 4: 2: 0 format and the 4: 4: 4 format.
- the 4: 2: 0 format is composed of luminance (Y) / color difference (Cb, Cr) signals, and one sample of the color difference signal is equivalent to 2 ⁇ 2 samples of the luminance signal
- 4: The 4: 4 format does not particularly limit the color space for expressing colors to Y, Cb, and Cr, and indicates that the sample ratio of signals of each color component is 1: 1.
- MPEG-4 AVC ISO / IEC 14496-10) / ITU-T H.264 standard MPEG-4 AVC (ISO / IEC 14496-10) / ITU-T H.264 Amendment2
- Non-Patent Document 2 For example, in the 4: 4: 4 format encoding according to Non-Patent Document 2, as shown in FIG. 10, first, the input video signal 1001 (4: 4: 4 format) to be encoded in advance is directly or appropriately selected. After conversion to a color space (Y, Cb, Cr, etc.), the block division unit 1002 divides the data into units of macroblocks (16 pixels ⁇ 16 lines rectangular blocks) and outputs the encoded video signal 1003 to the prediction unit 1004. input.
- a macroblock may be configured in units of three color components, or each color component may be regarded as an independent picture and configured as a single color component rectangular block, It is possible to select at the sequence level whether to use the structured macro block.
- the prediction unit 1004 predicts the image signal of each color component in the macroblock between frames and between frames to obtain a prediction error signal 1005.
- a motion vector is searched in units of sub-blocks obtained by further subdividing the macroblock itself or the macroblock, a motion compensated prediction image is generated based on the motion vector, and the encoded video
- a prediction error signal 1005 is obtained by taking a difference from the signal 1003.
- the compression unit 1006 performs conversion processing such as DCT (Discrete Cosine Transform) on the prediction error signal 1005 to remove signal correlation, and then quantizes it to obtain compressed data 1007.
- DCT Discrete Cosine Transform
- the compressed data 1007 is entropy-encoded by the variable-length encoding unit 1008 and output as a bit stream 1009 and sent to the local decoding unit 1010 to obtain a decoded prediction error signal 1011. This is added to the prediction signal 1012 used to generate the prediction error signal 1005 to obtain a decoded signal 1013.
- the decoded signal 1013 is stored in the memory 1014 for the purpose of generating a prediction signal 1012 for the subsequent encoded video signal 1003.
- the decoded signal is subjected to a deblocking filter to remove the block distortion.
- the prediction signal generation parameter 1015 determined by the prediction unit 1004 to obtain the prediction signal 1012 is sent to the variable length coding unit 1008 and output as a bit stream 1009.
- the prediction signal generation parameter 1015 includes, for example, an intra prediction mode indicating how to perform spatial prediction within a frame, a motion vector indicating a motion amount between frames, and the like.
- the prediction signal generation parameter 1015 is detected as a parameter that is commonly applied to the three color components, and each color component is regarded as an independent picture.
- the prediction signal generation parameter 1015 is detected as a parameter applied individually to each color component.
- the video signal of 4: 4: 4 format includes the same number of samples of each color component, and has strict color reproducibility compared with the conventional video signal of 4: 2: 0 format, but compression coding. In this sense, it contains redundant information.
- the redundancy included in the signal is further added to the conventional 4: 2: 0 format fixed color space definition (Y, Cb, Cr). It is necessary to reduce the degree.
- the encoded video signal 1003 is encoded regardless of the statistical and local properties of the signal, and each color component is encoded as a luminance signal, and the prediction unit In any of 1004, the compression unit 1006, and the variable length coding unit 1008, signal processing in consideration of the characteristics of the signal to be coded is not performed.
- the present invention provides a method for performing efficient information compression by better removing signal correlation according to the statistical and local properties of the 4: 4: 4 format video signal to be encoded,
- an image encoding device and an image decoding device that are highly optimized in encoding a moving image signal having no sample ratio distinction between color components as in the 4: 4: 4 format.
- An object is to provide an image encoding method and an image decoding method.
- An image encoding apparatus is an image encoding apparatus that divides each picture of a digital video signal into predetermined unit areas and performs compression encoding using motion compensated prediction in that unit.
- a prediction unit that performs a motion vector search based on virtual pixel accuracy designation information that designates an upper limit of the accuracy of the pixel position indicated by, and generates a motion compensated prediction image based on the searched motion vector, and the virtual pixel accuracy designation
- the information is multiplexed into the bitstream, and the motion vector data to be encoded is multiplexed into the bitstream based on the size of the searched motion vector and the size of the motion vector used for prediction of the searched motion vector.
- an encoding unit to be converted is an image encoding apparatus that divides each picture of a digital video signal into predetermined unit areas and performs compression encoding using motion compensated prediction in that unit.
- encoding using various color spaces is not limited to a fixed color space such as Y, Cb, and Cr.
- various color spaces are used without being limited to fixed color spaces such as Y, Cb, and Cr.
- intra-prediction mode information and inter-prediction mode information used for each color component can be flexibly selected, and optimal encoding processing can be performed even when the color space definition is various. It can be carried out.
- FIG. 3 is an explanatory diagram illustrating a configuration of an image encoding device according to the first embodiment.
- 4 is a flowchart of adaptive motion vector search / encoding in the image encoding device of FIG. 3.
- Example 1 Explanatory drawing which shows the division
- FIG. 3 is an explanatory diagram illustrating a data array of a bit stream output from the image encoding device according to the first embodiment.
- FIG. 3 is an explanatory diagram illustrating a configuration of an image decoding device according to the first embodiment.
- Example 1 The flowchart of the adaptive motion vector decoding in the image decoding apparatus of FIG. (Example 1) Explanatory drawing of 4: 2: 0, 4: 4: 4 format. Explanatory drawing which shows the structure of the conventional image coding apparatus (nonpatent literature 2).
- Example 1 the motion vector detection accuracy when performing motion compensation prediction processing is dynamically determined for an image encoding device and an image decoding device that compress and expand a digital video signal input in 4: 4: 4 format. An image encoding device and an image decoding device to be switched will be described.
- a digital video signal is composed of discrete pixel information (hereinafter referred to as integer pixels) generated by sampling the original analog video signal, but is interpolated between adjacent integer pixels.
- a technique of creating a virtual sample (virtual pixel) by using the virtual pixel as a motion compensation prediction value is widely used. This technique has two effects, that is, prediction accuracy is improved by increasing the number of candidate points for prediction, and singular points of the predicted image are reduced by the smoothing filter effect accompanying the interpolation operation, thereby improving the prediction efficiency.
- the accuracy of the virtual pixel is improved, the dynamic range of the motion vector that expresses the amount of motion also increases, so that the code amount generally increases.
- the unit of the motion vector value may be an integer pixel unit.
- the unit of the value of the motion vector is a half-pixel unit. Double dynamic range is required.
- FIG. 1 shows how a virtual pixel with 1/2 pixel accuracy is generated.
- A, B, C, and D are integer pixels
- e, f, g, h, and i are virtual pixels with 1/2 pixel accuracy generated from A to D.
- the half pixel a is generated as follows using integer pixels in the vicinity of the surrounding eight.
- the following formula shows only the case of horizontal processing, and the relationship between the half pixel a created for 1/4 pixel generation and the X components X -4 to X 4 of the integer pixel of the following formula is The positional relationship is shown in FIG.
- the virtual pixels used in the motion compensation prediction process can use the accuracy of half pixels and 1/4 pixels.
- the image encoding device and the image decoding device according to the first embodiment set the upper limit of the accuracy of available virtual pixels for each color component of the 4: 4: 4 video signal according to the encoding / decoding situation. Configure to be flexible.
- the virtual pixel generated from it is used as the predicted value of the original signal. In some cases, the effectiveness is not sufficiently ensured. In this case, the balance between the prediction efficiency that can be earned by using the highly accurate virtual pixel and the code amount of the increasing motion vector is deteriorated. Therefore, it is convenient to design the image encoding device and the image decoding device so that the accuracy of the virtual pixel for which the motion vector can be specified can be flexibly changed according to the encoding situation.
- FIG. 3 shows the configuration of the image coding apparatus according to the first embodiment.
- the operations other than the prediction unit 4 and the variable length encoding unit 8 almost follow the encoding operation of Non-Patent Document 2 described in the related art.
- the prediction unit 4 receives the virtual pixel accuracy instruction information 16, and based on the information, determines the accuracy of virtual pixels used for motion vector detection between frames and performs processing.
- the virtual pixel accuracy instruction information 16 is defined as a value that defines the relationship between the magnitude of the motion vector and the virtual pixel accuracy. That is, the upper limit of the size of the motion vector using the virtual pixel up to 1/4 pixel accuracy and the upper limit of the size of the motion vector using the virtual pixel up to half pixel accuracy are specified.
- a motion vector having a size exceeding the upper limit of the size of a motion vector using virtual pixels up to half-pixel accuracy is configured to use only integer pixels. By adopting such a configuration, the following effects are obtained.
- the motion vector is an amount representing the degree of motion between adjacent frames for each block.
- the size is small, it indicates that the prediction target block does not move so much from the corresponding block on the reference image. That is, it may be considered that the block area is in a state close to rest.
- the magnitude of the motion vector is large, it indicates that the prediction target block has moved greatly from the corresponding block on the reference image. That is, the block area may be considered as an area with a large temporal movement between adjacent frames (such as a subject with intense movement).
- the resolution of an image is high in a still area, and the resolution tends to decrease in an area where movement is intense.
- the prediction accuracy is improved by generating virtual pixels with high accuracy and using them for prediction in an area where the size of the motion vector is small and close to static.
- the effect of reducing the upper limit of the accuracy of the virtual pixel and reducing the code amount of the motion vector can be expected.
- each color component is regarded as an independent picture, and a macro block is configured with a rectangular block of a single color component to make a color
- the motion vector adaptive encoding processing will be described in detail for each case where individual motion vectors are applied to each component.
- the block division unit 2 outputs a macroblock composed of the three color components, and performs encoding / decoding in a mode using the common motion vector for the three color components.
- the virtual pixel accuracy instruction information 16 uses virtual pixels up to 1/4 pixel accuracy when the motion vector mv common to the three color components is smaller than the value Lq.
- the virtual pixel is used up to half-pixel accuracy.
- the value is larger than the value Lh, only integer pixels are used for motion compensation prediction.
- the processing flow by the prediction unit 4 and the variable length coding unit 8 is shown in FIG.
- the prediction unit 4 first performs a motion vector search using only integer pixels (step S1), and determines which of the above equations (1a) to (3a) is met.
- step S2 determines which of the above equations (1a) to (3a) is met.
- step S3 determines which of the above equations (1a) to (3a) is met.
- step S6 When the expression (2a) is met (Yes in step S4), mv is output as a part of the prediction signal generation parameter 15.
- the equation (2a) is not matched, the equation (1a) is satisfied (No in step S4), and a motion vector search using 1/4 pixel is executed within the range of Lq (step S5).
- Mv is output as part of the prediction signal generation parameter 15.
- the variable length encoding unit 8 uses mv input as a part of the prediction signal generation parameter 15 and Lq and Lh ⁇ specified by the virtual pixel accuracy instruction information 16 to perform the above (1a) to (3a
- the motion vector is efficiently encoded based on the encoded representation of the motion vector according to the equation (step S6).
- the motion vector is generally encoded with the prediction difference using the motion vector of the neighboring block as the predicted value, but in this case, the neighboring block that is the predicted value is always the maximum. It is possible to store the values as virtual pixel accuracy values, and to convert the values according to the above formulas (1a) to (3a) and take the difference in the same way as mv only when taking the prediction difference.
- the virtual pixel accuracy instruction information 16 is multiplexed and output to the bit stream 9 because it is necessary to decode the motion vector by the method according to the above equations (1a) to (3a) on the image decoding apparatus side (step S6).
- the block division unit 2 outputs a macroblock composed of a single color component, and performs encoding / decoding in a mode using individual motion vectors for each of the three color components.
- the motion vector mv k ′ to be encoded can be encoded with the dynamic range adaptively reduced as follows. (In the following formula, when mv k > 0, the sign is inverted when mv k ⁇ 0)
- the processing flow by the prediction unit 4 and the variable length coding unit 8 is shown in FIG.
- the prediction unit 4 first performs a motion vector search using only integer pixels, and determines which of the above equations (1b) to (3b) is met. When the expression (3b) is satisfied, the prediction process is terminated without executing the subsequent motion vector search using the virtual pixel with the half pixel and the 1/4 pixel accuracy, and mv is used as a part of the prediction signal generation parameter 15 Output k . If the equation (3b) is not met, a motion vector search with half-pixel accuracy is further performed within the range of Lh k to determine whether the equation (2b) is met. If the equation (2b) is met, mv k is output as part of the prediction signal generation parameter 15.
- the expression (2b) does not match, the expression (1b) is satisfied, and a motion vector search using 1/4 pixel is further performed within the range of Lq k , and a part of the prediction signal generation parameter 15 is obtained.
- Outputs mv k as The variable length encoding unit 8 uses mv k input as a part of the prediction signal generation parameter 15 and Lq k and Lh k indicated by the virtual pixel accuracy indication information 16 to (1b) above.
- the encoded representation of the motion vector according to the equations (3b) is possible, and the motion vector is efficiently encoded.
- the motion vector is generally encoded with the prediction difference using the motion vector of the neighboring block as the predicted value.
- the value is stored as the maximum virtual pixel accuracy value and the difference is calculated by converting the value according to the above formulas (1b) to (3b) as in mv k only when the prediction difference is taken. Good. Since the virtual pixel accuracy instruction information 16 needs to decode the motion vector by the method according to the above equations (1b) to (3b) on the image decoding side, the values for the three color components are multiplexed and output to the bit stream 9. .
- the processing flow is equivalent to FIG. 4 when mv in FIG. 4 is read as mv k and Lq and Lh are read as Lq k and Lh k .
- Lq and Lh indicated by the virtual pixel accuracy indication information 16 are defined as parameters that adaptively change with respect to these factors in the sequence, or different values are multiplexed individually for each picture. It is desirable to be configured. For example, if the video has a lot of motion throughout the video, and the quantization step size is large, the quality of the reference image is not good at a low bit rate, and the rate of the coding amount of the motion vector increases.
- Lh may be set small or invalidated so that the virtual pixel can be used easily.
- the properties of the video and the bit rate (quantization step size) may be combined, or may be used individually as control factors for Lq and Lh.
- Non-Patent Documents 1 and 2 when a predicted image is selectively obtained from a plurality of reference images that are temporally different from each other, Lq and Lh are controlled according to the index of the reference image to be used. You may comprise as follows.
- the virtual pixel accuracy instruction information 16 may be configured to be linked with the size of the block of the motion vector search unit to be used.
- a block of a plurality of sizes can be used as a motion vector search unit block as shown in FIG. If the size of the block to which the motion vector is assigned is large, the image pattern can be captured efficiently even if the size of the motion vector itself is large. However, if the block is small, the effect of noise is greater than the image pattern. It becomes easy to receive. Therefore, when the block size of a block to which a motion vector is allocated is large, Lq and Lh are reduced or invalidated to increase the frequency of 1/4 pixel precision motion compensation prediction, and the block size is small. Lq and Lh may be made large or activated.
- the virtual pixel accuracy instruction information 16 may be configured to control Lq k and Lh k independently for each color component (k) when an individual motion vector is used for each color component. For example, when encoding is performed in a color space such as Y, Cb, and Cr, the characteristics of the signal are different for each color component, so the effects of Lq k and Lh k may be different for each color component.
- the virtual pixel accuracy instruction information 16 is targeted only for half pixels and 1/4 pixels in the above example, but in the case where finer virtual pixels such as 1/8 pixels and 1/16 pixels are used. Also, it can be easily expanded by providing a new upper limit setting similar to Lq and Lh.
- the input video signal 1 is encoded by the image encoding device in FIG. 3 based on the above processing, and output from the image encoding device as a bitstream 9 in units of slices in which a plurality of macroblocks are bundled. Is done.
- FIG. 6 shows a data array of the bit stream 9.
- the bit stream 9 is configured as a collection of encoded data for the number of macroblocks included in a picture, and the macroblocks are unitized into data units called slices.
- a picture level header that is referenced as a common parameter by macroblocks belonging to the same picture is prepared, and virtual pixel accuracy instruction information 16 is stored in the picture level header.
- the common / independent coding identification flag 17 multiplexed in the sequence level header indicates that a common motion vector is used for the three color components
- Lq and Lh are a set
- the common / independent coding identification flag 17 is In order to indicate that an individual motion vector is used for each color component, three sets of Lq k and Lh k (the number of color components) are multiplexed.
- Each slice starts from a slice header, and then encoded data of each macroblock in the slice is arranged (in this example, the second slice includes M macroblocks).
- the common / independent encoding identification flag 17 indicates that an individual motion vector is used for each color component
- color component identification information 18 indicating which color component encoded data is included in the slice in the slice header is displayed.
- the virtual pixel accuracy instruction information 16 may be configured to multiplex Lq k and Lh k specified by the color component identification information 18 in the slice header.
- the coding mode, motion vector, quantization step size parameter, prediction error compressed data, and the like are arranged in the data of each macroblock.
- the motion vector is obtained as a difference between mv ′ determined by the above-described equations (1a) to (3a) (or (1b) to (3b)) and the predicted value pmv ′ converted in the same manner. Is encoded.
- the virtual pixel accuracy instruction information 16 is configured to be stored in a sequence level header that is assigned in units of a sequence in which a plurality of video frames are bundled, and is based on each encoded data such as a picture, a slice, and a macro block.
- Lq and Lh may be determined by adaptively changing the multiplexing information of the sequence level header. In this way, it is not necessary to encode and transmit the virtual pixel accuracy instruction information 16 with individual picture level headers, and the amount of header information can be reduced.
- FIG. 7 shows the configuration of the image decoding device according to the first embodiment.
- the variable length decoding unit 20 decodes the bit stream 9 shown in FIG. 6 and extracts / interprets the common / independent coding identification flag 17 so that the macroblock is composed of three color components or a single color component.
- the bit stream analysis of subsequent slices and macroblocks is advanced.
- the virtual pixel accuracy instruction information 16 is extracted from the bitstream 9.
- a slice header, prediction error compressed data 22 of each macroblock, encoding mode, prediction signal generation parameter 15 including a motion vector, quantization step size parameter 23, and the like are extracted according to a predetermined rule (syntax).
- the prediction error compression data 22 and the quantization step size parameter 23 are input to the prediction error decoding unit 24 and restored to the decoded prediction error signal 25.
- the prediction unit 21 generates a prediction image 26 from the prediction signal generation parameter 15 decoded by the variable length decoding unit 20 and the reference image in the memory 28 (the motion vector detection operation of the prediction unit 4 in the image encoding device is Not included).
- the decoded prediction error signal 25 and the predicted image 26 are added by an adder to obtain a decoded signal 27.
- the decoded signal 27 is stored in the memory 28 for use in subsequent motion compensation prediction of the macroblock.
- a process of removing a block distortion by applying a deblocking filter to the decoded signal may be performed before the data is written in the memory 28, a process of removing a block distortion by applying a deblocking filter to the decoded signal may be performed.
- the decoded signal 27 is restored as an image signal of either a macroblock including three color components or a macroblock including only a single
- the maximum accuracy of the virtual pixel indicated by the motion vector is 1/4 pixel
- the motion vector output from the variable length decoding unit 20 as part of the prediction signal generation parameter 15 is always It is assumed that the value is output to the prediction unit 21 as a value in which 1/4 pixel is 1. That is, the motion vector encoded by compressing the dynamic range by the above-described equations (1a) to (3a) (or (1b) to (3b)) in the image encoding device is a virtual pixel extracted from the bitstream.
- the accuracy instruction information 16 mvd extracted from the bitstream for each motion vector allocation block, and motion vector prediction value pmv '
- inverse transformation of the process at the time of encoding is performed, and the dynamic range is restored to predictor 21 To be output.
- Fig. 8 shows the processing flow of this inverse transformation.
- the variable length decoding unit 20 first extracts mvd that is motion vector encoded data from the bitstream (step S10). This corresponds to data encoded by compressing the dynamic range according to the above-described equations (1) to (3) at the time of encoding.
- pmv that is a predicted value of the motion vector is obtained, and this is encoded using the virtual pixel accuracy instruction information 16 according to the above-described equations (1a) to (3a) (or (1b) to (3b)).
- Conversion processing similar to that at the time of conversion is performed, and the dynamic range is compressed (step S11).
- This mv '' is output to the prediction unit 21 and held internally as a prediction value for subsequent motion vector decoding (step S13).
- the prediction unit 21 can always handle a motion vector in units of 1/4 pixels without being aware of the dynamic range of the encoded motion vector.
- Lq and Lh are used as values common to the three color components.
- (4) to (6) are used using Lq k and Lh k decoded for each color component (k). ) May be applied independently for each color component, or Lq k and Lh k may be set to the same value for all color components, and common Lq and Lh may be used.
- Lq and Lh are the frame resolution of the video to be decoded, the quantization step size parameter 23, the block size of the motion vector allocation unit (this is the coding It may be configured to change in conjunction with the encoding information included in the bit stream 9 such as the index of the reference image).
- the image encoding device and the image decoding device in Embodiment 1 described above in order to efficiently encode a 4: 4: 4 format color video signal, according to the nature of the signal of each color component, Since it is possible to dynamically switch the accuracy of virtual samples used for motion vector detection / prediction image generation, the amount of motion vector code can be effectively suppressed in low bit rate coding with a high compression rate. Thus, it is possible to provide an image encoding device and an image decoding device that perform encoding.
- the image encoding device and the image decoding device according to the first embodiment also have an effect of reducing the complexity of the image encoding process / decoding process for the following reason.
- the higher the video resolution and the greater the number of pixels in the screen the greater the number of moving pixels when the amount of movement of the subject is converted to the number of pixels compared to when the resolution is low. I need to take it.
- the number of evaluation points increases and the amount of evaluation calculation on the image coding apparatus side increases.
- the image coding apparatus of the first embodiment at the time of integer pixel search.
- the configuration is such that the search for virtual pixels is terminated, so that the amount of computation can be reduced.
- the reference image is a frame memory having a large data size, it is stored in a large-capacity memory such as an external DRAM (memory 14, memory 28, etc.).
- the reference image is stored on the external memory.
- a part of the reference image is mounted and configured to perform operations by fetching a part of the reference image into an internal cache.
- the virtual pixel generation process is performed only when the magnitude of the motion vector is smaller than a certain threshold value. Therefore, the memory band required for the interpolation filtering process It also has the effect of reducing the width and power consumption.
- adaptive encoding of a motion vector in the present invention is based on conventional luminance / Needless to say, the present invention can also be applied to the efficiency of motion vector coding in video coding for 4: 2: 0 and 4: 2: 2 formats in which color thinning is performed in the color difference component format. .
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
Description
本実施の形態では、4:4:4フォーマットで入力されるデジタル映像信号の圧縮伸張を行う画像符号化装置、画像復号装置につき、動き補償予測処理を行う際の動きベクトル検出精度を動的に切り替える画像符号化装置、画像復号装置について述べる。
f = (C+D)//2
g = (A+C)//2
h = (B+D)//2
i = (A+B+C+D)//4
(ただし、//は丸めつき除算を示す。)
以降、便宜上、1/2画素精度の仮想画素のことを単に「半画素」と記す。
(ただし、COEk: フィルタ係数(係数総和が256)。//は丸めつき除算を示す。)
AVC(ISO/IEC 14496-10)では、半画素を生成する際に[1,-5,20,20,-5,1]なる6タップのフィルタを採用しており、さらに1/4画素を上記MPEG-1、MPEG-2の半画素生成と同様の線形補間処理によって生成している。さらに、1/4画素と1/4画素の間に位置する1/8画素精度の仮想サンプルを同様の手順で求めて使用する事例もある。
本実施の形態1において、動き補償予測処理時に用いる仮想画素は、半画素、1/4画素の精度を使用可能とする。そのうえで、本実施の形態1の画像符号化装置、画像復号装置は、4:4:4映像信号の各色成分ごとに、利用可能な仮想画素の精度の上限を符号化・復号の状況に応じて柔軟に指定できるように構成する。
(i)仮想画素を用いた動き補償予測においては、画像符号化装置側・画像復号装置側で同一の参照画像を用いて仮想画素を生成しなければならない。一般に圧縮された映像信号では、圧縮率が高いほど、動き補償予測に用いる参照画像の品質は低下していく。精度の高い仮想画素を使用する効果は、参照画像が符号化前の原信号に近く、品質の高い信号である(すなわち圧縮率が低い、ないしは高ビットレート符号化)ほど顕著となるが、これは動きベクトルを符号化して伝送する情報量が増加しても、それを予測効率の改善分で補えるケースに相当する。しかし、圧縮率が高く(低ビットレート符号化時)、参照画像の品質が符号化前の原信号からかなり劣化している場合には、そこから生成した仮想画素は原信号の予測値としての有効性が十分に確保されないケースが発生し、その場合、精度の高い仮想画素を使用することによって稼ぐことができる予測効率と、増加する動きベクトルの符号量とのバランスが悪くなる。したがって、符号化の状況に応じて動きベクトルが指定できる仮想画素の精度を柔軟に変更できるように画像符号化装置・画像復号装置を設計しておくと都合が良い。
ブロック分割部2が3つの色成分からなるマクロブロックを出力し、3つの色成分に共通の動きベクトルを用いるモードで符号化・復号を行う場合は、仮想画素精度指示情報16は、3つの色成分に共通の動きベクトル mv に対し、値 Lq よりも小さい場合には1/4画素精度まで仮想画素を用いることとし、値 Lq 以上で値 Lh よりも小さい場合には半画素精度まで仮想画素を用いることとし、値 Lh よりも大きい場合は整数画素のみを動き補償予測に用いるものと規定する。このように定めることにより、符号化すべき動きベクトル mv' は、以下のようにダイナミックレンジを適応的に低減して符号化することが可能となる。(下式は mv > 0の場合、mv < 0の場合は符号反転)
→1/4画素精度を使用可
mv' = Lq + (mv Lq + 1)/2 (Lq <= mv < Lh) (2a)
→半画素精度を使用可
mv' = Lq + (Lh Lq + 1)/2 + (mv Lh + 2)/4 (Lh <= mv) (3a)
→整数画素精度のみ使用可
ブロック分割部2が単一色成分からなるマクロブロックを出力し、3つの色成分ごとに個別の動きベクトルを用いるモードで符号化・復号を行う場合は、仮想画素精度指示情報16は、3つの色成分の各動きベクトル mvk (k=0,1,2)に対し、値 Lqk よりも小さい場合には1/4画素精度まで仮想画素を用いることとし、値 Lqk 以上で値 Lhk よりも小さい場合には半画素精度まで仮想画素を用いることとし、値 Lhk よりも大きい場合は整数画素のみを動き補償予測に用いるものと規定する。このように定めることにより、符号化すべき動きベクトル mvk' は、以下のようにダイナミックレンジを適応的に低減して符号化することが可能となる。(下式は mvk > 0の場合、mvk < 0の場合は符号反転)
→1/4画素精度を使用可
mvk' = Lqk + (mvk Lqk + 1)/2 (Lqk <= mvk < Lhk) (2b)
→半画素精度を使用可
mvk' = Lqk + (Lhk Lqk + 1)/2 + (mvk Lhk + 2)/4 (Lhk <= mvk) (3b)
→整数画素精度のみ使用可
また、処理フローは、図4の mv を mvk 、Lq, Lh をLqk, Lhk と読み替えれば図4と等価である。
入力映像信号1は、上記処理に基づいて図3の画像符号化装置で符号化され、複数のマクロブロックを束ねたスライスの単位でビットストリーム9として画像符号化装置から出力される。図6に、ビットストリーム9のデータ配列を示す。ビットストリーム9は、ピクチャ中に含まれるマクロブロック数分の符号化データが集められたものとして構成され、マクロブロックは複数個集められたスライスというデータ単位にユニット化される。同一ピクチャに属するマクロブロックが共通パラメータとして参照するピクチャレベルヘッダが用意され、ピクチャレベルヘッダには、仮想画素精度指示情報16が格納される。シーケンスレベルヘッダに多重化される共通・独立符号化識別フラグ17が、3つの色成分に共通の動きベクトルを用いることを示す場合はLq、Lhが一組、共通・独立符号化識別フラグ17が、各色成分ごとに個別の動きベクトルを用いることを示す場合はLqk, Lhkが3組(色成分の数だけ)多重される。
図7に本実施の形態1における画像復号装置の構成を示す。可変長復号部20は、図6に示すビットストリーム9を復号し、共通・独立符号化識別フラグ17を抽出・解釈することにより、マクロブロックが3つの色成分から構成されるか、単一色成分で構成されるかを判断して、以降のスライス、マクロブロックのビットストリーム解析を進める。共通・独立符号化識別フラグ17の復号値に基づいて仮想画素精度指示情報16をビットストリーム9から抽出する。次いで、所定の規則(シンタックス)に従ってスライスヘッダ、各マクロブロックの予測誤差圧縮データ22、符号化モード、動きベクトルを含む予測信号生成用パラメータ15、量子化ステップサイズパラメータ23、などを抽出する。
mv'' = (mv' Lq) x 2 + Lq (Lq <= mv < (Lq + (Lh-Lq)/2)) (5)
mv'' = (mv' Lq (Lh-Lq)/2) x 4 + Lh ((Lq + (Lh-Lq)/2) <= mv) (6)
Claims (4)
- デジタル映像信号の各ピクチャを所定の単位領域に分割して、その単位で動き補償予測を用いて圧縮符号化を行う画像符号化装置であって、
動きベクトルが指し示す画素位置の精度の上限を指定する仮想画素精度指定情報に基づいて動きベクトル探索を行い、該探索された動きベクトルに基づいて動き補償予測画像を生成する予測部と、
前記仮想画素精度指定情報をビットストリームに多重化するとともに、前記探索された動きベクトルの大きさと、該探索された動きベクトルの予測に用いる動きベクトルの大きさとに基づいて、符号化する動きベクトルデータをビットストリームに多重化する符号化部とを備えることを特徴とする画像符号化装置。 - デジタル映像信号の各ピクチャを所定の単位領域に分割して、その単位で動き補償予測を用いて圧縮符号化された画像符号化ビットストリームを受信して映像信号を復元する画像復号装置であって、
動きベクトルが指し示す画素位置の精度の上限を指定する仮想画素精度指定情報をビットストリームから抽出するとともに、動きベクトルが割り当てられる領域の単位で動きベクトルの符号化データをビットストリームから抽出して動きベクトルを復元する復号部と、
該復号部によって復号される動きベクトルに基づいて動き補償予測画像を生成する予測部とを備え、
前記復号部は、ビットストリームから抽出された動きベクトル符号化データと、該復号対象の動きベクトルの予測に用いる動きベクトルとから復元されるデータの大きさと、前記ビットストリームから抽出した仮想画素精度指定情報とに基づいて動きベクトルを復号することを特徴とする画像復号装置。 - デジタル映像信号の各ピクチャを所定の単位領域に分割して、その単位で動き補償予測を用いて圧縮符号化を行う画像符号化方法であって、
動きベクトルが指し示す画素位置の精度の上限を指定する仮想画素精度指定情報に基づいて動きベクトル探索を行い、該探索された動きベクトルに基づいて動き補償予測画像を生成する予測ステップと、
前記仮想画素精度指定情報をビットストリームに多重化するとともに、前記探索された動きベクトルの大きさと、該探索された動きベクトルの予測に用いる動きベクトルの大きさとに基づいて、符号化する動きベクトルデータをビットストリームに多重化する符号化ステップとを備えることを特徴とする画像符号化方法。 - デジタル映像信号の各ピクチャを所定の単位領域に分割して、その単位で動き補償予測を用いて圧縮符号化された画像符号化ビットストリームを受信して映像信号を復元する画像復号方法であって、
動きベクトルが指し示す画素位置の精度の上限を指定する仮想画素精度指定情報をビットストリームから抽出するとともに、動きベクトルが割り当てられる領域の単位で動きベクトルの符号化データをビットストリームから抽出して動きベクトルを復元する復号ステップと、
該復号される動きベクトルに基づいて動き補償予測画像を生成する予測ステップとを備え、
前記復号ステップは、ビットストリームから抽出された動きベクトル符号化データと、該復号対象の動きベクトルの予測に用いる動きベクトルとから復元されるデータの大きさと、前記ビットストリームから抽出した仮想画素精度指定情報とに基づいて動きベクトルを復号することを特徴とする画像復号方法。
Priority Applications (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP09701456A EP2234404A4 (en) | 2008-01-09 | 2009-01-08 | IMAGE ENCRYPTION DEVICE, IMAGE KEYING DEVICE, IMAGE ENCRYPTION METHOD AND IMAGE KEYING PROCESS |
| CN2009801020090A CN101911706A (zh) | 2008-01-09 | 2009-01-08 | 图像编码装置、图像解码装置、图像编码方法以及图像解码方法 |
| CA 2711742 CA2711742A1 (en) | 2008-01-09 | 2009-01-08 | Image encoding device, image decoding device, image encoding method, and image decoding method |
| US12/812,185 US20110032991A1 (en) | 2008-01-09 | 2009-01-08 | Image encoding device, image decoding device, image encoding method, and image decoding method |
| JP2009548946A JP5197630B2 (ja) | 2008-01-09 | 2009-01-08 | 画像符号化装置、画像復号装置、画像符号化方法、および画像復号方法 |
| BRPI0906824-4A BRPI0906824A2 (pt) | 2008-01-09 | 2009-01-08 | Dispositivo de codificação de imagem, dispositivo de decodificação de imagem, método de codificação de imagem e método de decodificação de imagem |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2008002223 | 2008-01-09 | ||
| JP2008-002223 | 2008-01-09 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2009088038A1 true WO2009088038A1 (ja) | 2009-07-16 |
Family
ID=40853150
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2009/050142 Ceased WO2009088038A1 (ja) | 2008-01-09 | 2009-01-08 | 画像符号化装置、画像復号装置、画像符号化方法、および画像復号方法 |
Country Status (9)
| Country | Link |
|---|---|
| US (1) | US20110032991A1 (ja) |
| EP (1) | EP2234404A4 (ja) |
| JP (1) | JP5197630B2 (ja) |
| KR (1) | KR20100099723A (ja) |
| CN (1) | CN101911706A (ja) |
| BR (1) | BRPI0906824A2 (ja) |
| CA (1) | CA2711742A1 (ja) |
| RU (1) | RU2010133237A (ja) |
| WO (1) | WO2009088038A1 (ja) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| JP2013520875A (ja) * | 2010-02-19 | 2013-06-06 | クゥアルコム・インコーポレイテッド | ビデオコーディングのための適応動き解像度 |
| JP2017508344A (ja) * | 2014-01-08 | 2017-03-23 | マイクロソフト テクノロジー ライセンシング,エルエルシー | 動きベクトル精度の選択 |
| US10327008B2 (en) | 2010-10-13 | 2019-06-18 | Qualcomm Incorporated | Adaptive motion vector resolution signaling for video coding |
| US10587891B2 (en) | 2014-01-08 | 2020-03-10 | Microsoft Technology Licensing, Llc | Representing motion vectors in an encoded bitstream |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8145002B2 (en) * | 2007-06-28 | 2012-03-27 | Mitsubishi Electric Corporation | Image encoding device and image encoding method |
| CN102742270B (zh) * | 2010-02-05 | 2016-02-03 | 瑞典爱立信有限公司 | 用于子像素插值的改进的方法和设备 |
| JP2012104945A (ja) * | 2010-11-08 | 2012-05-31 | Sony Corp | 画像処理装置と画像処理方法およびプログラム |
| PL2661892T3 (pl) | 2011-01-07 | 2022-08-16 | Nokia Technologies Oy | Przewidywanie ruchu w kodowaniu wideo |
| JP6056122B2 (ja) * | 2011-01-24 | 2017-01-11 | ソニー株式会社 | 画像符号化装置と画像復号装置およびその方法とプログラム |
| JP6008334B2 (ja) | 2011-08-25 | 2016-10-19 | サン パテント トラスト | 復号装置及び復号方法 |
| CN106791842A (zh) * | 2011-09-06 | 2017-05-31 | 英特尔公司 | 分析辅助编码 |
| MX2013008692A (es) | 2011-09-07 | 2013-08-21 | Panasonic Corp | Metodo de codificacion de imagenes, metodo de decodificacion de imagenes, aparato de codificacion de imagenes, aparato de decodificacion de imagenes y aparato de codificacion y decodificacion de imagenes. |
| PL2760205T3 (pl) | 2011-09-19 | 2021-05-17 | Sun Patent Trust | Sposób dekodowania obrazów, urządzenie do dekodowania obrazów |
| PL3742735T3 (pl) | 2011-10-19 | 2022-11-14 | Sun Patent Trust | Sposób kodowania obrazów, sposób dekodowania obrazów, urządzenie do kodowania obrazów, urządzenie do dekodowania obrazów, i urządzenie do kodowania i dekodowania obrazów |
| WO2017135151A1 (ja) | 2016-02-01 | 2017-08-10 | シャープ株式会社 | 予測画像生成装置、動画像復号装置、および動画像符号化装置 |
| KR102728610B1 (ko) | 2017-04-21 | 2024-11-08 | 제니맥스 미디어 인크. | 게임-생성된 모션 벡터들을 위한 시스템들 및 방법들 |
| WO2020084476A1 (en) | 2018-10-22 | 2020-04-30 | Beijing Bytedance Network Technology Co., Ltd. | Sub-block based prediction |
| CN111083489B (zh) | 2018-10-22 | 2024-05-14 | 北京字节跳动网络技术有限公司 | 多次迭代运动矢量细化 |
| CN111436229B (zh) | 2018-11-12 | 2024-06-28 | 北京字节跳动网络技术有限公司 | 帧间预测的带宽控制方法 |
| WO2020103872A1 (en) | 2018-11-20 | 2020-05-28 | Beijing Bytedance Network Technology Co., Ltd. | Prediction refinement for combined inter intra prediction mode |
| EP3861742A4 (en) | 2018-11-20 | 2022-04-13 | Beijing Bytedance Network Technology Co., Ltd. | DIFFERENCE CALCULATION BASED ON SPATIAL POSITION |
| CN111010581B (zh) * | 2018-12-07 | 2022-08-12 | 北京达佳互联信息技术有限公司 | 运动矢量信息的处理方法、装置、电子设备及存储介质 |
| CN113545065B (zh) | 2019-03-06 | 2023-12-12 | 北京字节跳动网络技术有限公司 | 转换后的单向预测候选的使用 |
| CN113875247A (zh) * | 2019-04-05 | 2021-12-31 | 巨人计划有限责任公司 | 高动态范围视频格式检测 |
| JP6835337B2 (ja) * | 2019-06-20 | 2021-02-24 | Kddi株式会社 | 画像復号装置、画像復号方法及びプログラム |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08251596A (ja) * | 1995-03-10 | 1996-09-27 | Toshiba Corp | 動画像符号化/復号化装置 |
| JP2004236049A (ja) * | 2003-01-31 | 2004-08-19 | Hitachi Ltd | 符号化方法及び復号化方法 |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5731840A (en) * | 1995-03-10 | 1998-03-24 | Kabushiki Kaisha Toshiba | Video coding/decoding apparatus which transmits different accuracy prediction levels |
| JP3856262B2 (ja) * | 1998-03-09 | 2006-12-13 | 日本ビクター株式会社 | 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 |
| GB2339101B (en) * | 1998-06-25 | 2002-09-18 | Sony Uk Ltd | Processing of compressed video signals using picture motion vectors |
| JP2003169338A (ja) * | 2001-09-18 | 2003-06-13 | Matsushita Electric Ind Co Ltd | 動きベクトル検出方法及び装置並びに方法プログラムを記録した媒体 |
| JP4724351B2 (ja) * | 2002-07-15 | 2011-07-13 | 三菱電機株式会社 | 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 |
| JP2004274212A (ja) * | 2003-03-06 | 2004-09-30 | Renesas Technology Corp | 画像符号化装置 |
| US7310451B2 (en) * | 2005-01-03 | 2007-12-18 | Intel Corporation | Sub-pixel image shifting in display device |
-
2009
- 2009-01-08 US US12/812,185 patent/US20110032991A1/en not_active Abandoned
- 2009-01-08 BR BRPI0906824-4A patent/BRPI0906824A2/pt not_active IP Right Cessation
- 2009-01-08 WO PCT/JP2009/050142 patent/WO2009088038A1/ja not_active Ceased
- 2009-01-08 KR KR20107015100A patent/KR20100099723A/ko not_active Ceased
- 2009-01-08 CA CA 2711742 patent/CA2711742A1/en not_active Abandoned
- 2009-01-08 JP JP2009548946A patent/JP5197630B2/ja not_active Expired - Fee Related
- 2009-01-08 CN CN2009801020090A patent/CN101911706A/zh active Pending
- 2009-01-08 RU RU2010133237/07A patent/RU2010133237A/ru not_active Application Discontinuation
- 2009-01-08 EP EP09701456A patent/EP2234404A4/en not_active Withdrawn
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08251596A (ja) * | 1995-03-10 | 1996-09-27 | Toshiba Corp | 動画像符号化/復号化装置 |
| JP2004236049A (ja) * | 2003-01-31 | 2004-08-19 | Hitachi Ltd | 符号化方法及び復号化方法 |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| KR101792041B1 (ko) * | 2009-10-14 | 2017-11-02 | 톰슨 라이센싱 | 움직임 정보의 적응적 코딩 방법 및 장치 |
| 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 |
| JP2013520875A (ja) * | 2010-02-19 | 2013-06-06 | クゥアルコム・インコーポレイテッド | ビデオコーディングのための適応動き解像度 |
| JP2015156653A (ja) * | 2010-02-19 | 2015-08-27 | クゥアルコム・インコーポレイテッドQualcomm Incorporated | ビデオコーディングのための適応動き解像度 |
| US9237355B2 (en) | 2010-02-19 | 2016-01-12 | Qualcomm Incorporated | Adaptive motion resolution for video coding |
| US10327008B2 (en) | 2010-10-13 | 2019-06-18 | Qualcomm Incorporated | Adaptive motion vector resolution signaling for video coding |
| JP2017508344A (ja) * | 2014-01-08 | 2017-03-23 | マイクロソフト テクノロジー ライセンシング,エルエルシー | 動きベクトル精度の選択 |
| US10313680B2 (en) | 2014-01-08 | 2019-06-04 | Microsoft Technology Licensing, Llc | Selection of motion vector precision |
| US10587891B2 (en) | 2014-01-08 | 2020-03-10 | Microsoft Technology Licensing, Llc | Representing motion vectors in an encoded bitstream |
Also Published As
| Publication number | Publication date |
|---|---|
| EP2234404A1 (en) | 2010-09-29 |
| EP2234404A4 (en) | 2011-06-22 |
| JP5197630B2 (ja) | 2013-05-15 |
| US20110032991A1 (en) | 2011-02-10 |
| RU2010133237A (ru) | 2012-02-20 |
| KR20100099723A (ko) | 2010-09-13 |
| CN101911706A (zh) | 2010-12-08 |
| JPWO2009088038A1 (ja) | 2011-05-26 |
| BRPI0906824A2 (pt) | 2015-07-14 |
| CA2711742A1 (en) | 2009-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5197630B2 (ja) | 画像符号化装置、画像復号装置、画像符号化方法、および画像復号方法 | |
| KR100966427B1 (ko) | 화상 복호 장치 및 화상 복호 방법 | |
| CN114830651B (zh) | 帧内预测方法、编码器、解码器以及计算机存储介质 | |
| KR102024516B1 (ko) | 화상 부호화 장치, 화상 복호 장치, 화상 부호화 방법, 화상 복호 방법 및 기억 매체 | |
| TWI411310B (zh) | Image processing apparatus and method | |
| WO2012172668A1 (ja) | 動画像符号化方法及び装置並びに動画復号化方法及び装置 | |
| JPWO2011039931A1 (ja) | 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 | |
| KR20150034699A (ko) | 인트라 모드를 이용한 쿼터 픽셀 해상도를 갖는 영상 보간 방법 및 장치 | |
| JPWO2013001720A1 (ja) | 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 | |
| KR20170114598A (ko) | 적응적 색상 순서에 따른 색상 성분 간 예측을 이용한 동영상 부호화 및 복호화 방법 및 장치 | |
| US20070133689A1 (en) | Low-cost motion estimation apparatus and method thereof | |
| KR20210023911A (ko) | 인트라 예측 방법 및 장치 | |
| KR20110087871A (ko) | 인트라 모드를 이용한 쿼터 픽셀 해상도를 갖는 영상 보간 방법 및 장치 | |
| KR102111437B1 (ko) | 인트라 모드를 이용한 쿼터 픽셀 해상도를 갖는 영상 보간 방법 및 장치 | |
| HK1148414A (en) | Image encoding device, image decoding device, image encoding method, and image decoding method | |
| JP4153774B2 (ja) | 動画像符号化方法とその復号化方法、およびそれらの装置 | |
| WO2022146215A1 (en) | Temporal filter | |
| JP2012209873A (ja) | 動画像符号化装置及び動画像符号化方法 | |
| HK40040676B (en) | Video decoding method, video encoding method, and related device | |
| HK40040676A (en) | Video decoding method, video encoding method, and related device | |
| KR20100041441A (ko) | 비디오 부호화/복호화 방법 및 장치 | |
| KR20190084929A (ko) | 인트라 모드를 이용한 쿼터 픽셀 해상도를 갖는 영상 보간 방법 및 장치 | |
| HK1260985A1 (en) | Image encoding device, image decoding device, and methods thereof | |
| HK1260985B (en) | Image encoding device, image decoding device, and methods thereof | |
| KR20190004247A (ko) | 인트라 모드를 이용한 쿼터 픽셀 해상도를 갖는 영상 보간 방법 및 장치 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| WWE | Wipo information: entry into national phase |
Ref document number: 200980102009.0 Country of ref document: CN |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09701456 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 2009548946 Country of ref document: JP Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 4082/CHENP/2010 Country of ref document: IN |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2009701456 Country of ref document: EP |
|
| ENP | Entry into the national phase |
Ref document number: 20107015100 Country of ref document: KR Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2711742 Country of ref document: CA |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2010133237 Country of ref document: RU |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 12812185 Country of ref document: US |
|
| ENP | Entry into the national phase |
Ref document number: PI0906824 Country of ref document: BR Kind code of ref document: A2 Effective date: 20100709 |