WO2011052990A2 - Procédé et appareil de codage/décodage d'images sur la base d'une résolution adaptative - Google Patents

Procédé et appareil de codage/décodage d'images sur la base d'une résolution adaptative Download PDF

Info

Publication number
WO2011052990A2
WO2011052990A2 PCT/KR2010/007420 KR2010007420W WO2011052990A2 WO 2011052990 A2 WO2011052990 A2 WO 2011052990A2 KR 2010007420 W KR2010007420 W KR 2010007420W WO 2011052990 A2 WO2011052990 A2 WO 2011052990A2
Authority
WO
WIPO (PCT)
Prior art keywords
resolution
block
current block
image
reconstructed
Prior art date
Application number
PCT/KR2010/007420
Other languages
English (en)
Korean (ko)
Other versions
WO2011052990A3 (fr
Inventor
김수년
임정연
이규민
최재훈
김해광
전병우
문주희
이영렬
한종기
Original Assignee
에스케이텔레콤 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 에스케이텔레콤 주식회사 filed Critical 에스케이텔레콤 주식회사
Publication of WO2011052990A2 publication Critical patent/WO2011052990A2/fr
Publication of WO2011052990A3 publication Critical patent/WO2011052990A3/fr

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
    • H04N19/50—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
    • H04N19/53—Multi-resolution motion estimation; Hierarchical motion estimation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/462—Content or additional data management, e.g. creating a master electronic program guide from data received from the Internet and a Head-end, controlling the complexity of a video stream by scaling the resolution or bit-rate based on the client capabilities
    • H04N21/4621—Controlling the complexity of the content stream or additional data, e.g. lowering the resolution or bit-rate of the video stream for a mobile client with a small screen
    • 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 method and apparatus for adaptive resolution based image encoding / decoding. More specifically, the present invention relates to a method and apparatus for adaptively selecting a resolution for each predetermined coding unit and encoding the image at the selected resolution to improve compression efficiency.
  • Moving Picture Experts Group (MPEG) and Video Coding Experts Group (VCEG) have developed video compression techniques that are superior and superior to the existing MPEG-4 Part 2 and H.263 standards.
  • the new standard is called H.264 / AVC (Advanced video Coding) and co-published as MPEG-4 Part 10 AVC and ITU-T Recommendation H.264.
  • H.264 / AVC abbreviated as 'H.264'
  • intra prediction and / or inter prediction in units of macroblocks having various types of subblocks.
  • a residual signal is generated by performing a transform, and transform and quantization are performed on the generated residual signal and encoded.
  • Conventional video encoding techniques such as H.264 / AVC encode a video at a fixed resolution, and thus there is a problem in that the resolution of the video cannot be adaptively changed and encoded.
  • conventional video encoding techniques encode images in units of 16x16 pixel macroblocks, 8x8 pixel subblocks, or 4x4 pixel subblocks.
  • Such conventional image encoding techniques encode small images. Although it may be suitable, the efficiency of encoding a large sized image may not be good. This is because it is efficient to use a large macroblock to encode a large sized image, but the size of the macroblock is limited to 16 ⁇ 16 in a typical image encoding technique.
  • the present invention has a main object to improve compression efficiency by adaptively selecting a resolution for each predetermined coding unit and encoding the image at the selected resolution.
  • the present invention provides a method of encoding an image, the method comprising: converting a current block of an image into a current block having a plurality of resolutions; Selecting one of an original resolution and a plurality of resolutions; And generating a bitstream including prediction encoded image coded data, resolution data encoded with information on the selected resolution, and motion information data encoded with motion information.
  • a video encoding method is provided.
  • an apparatus for encoding an image comprising: a resolution converter for converting a current block of an image into a current block having a plurality of resolutions; A resolution selector for selecting one of an original resolution and a plurality of resolutions; A motion estimator for generating motion information by estimating the motion of the current block having the selected resolution in the reference picture converted to the selected resolution; A motion compensator for generating a prediction block having a selected resolution by compensating for a motion of a current block having a selected resolution using motion information in the reference picture converted to the selected resolution; A subtractor for generating a residual block having a selected resolution by subtracting a current block having a selected resolution and a prediction block having a selected resolution; a converter and a quantizer for converting and quantizing the residual block having the selected resolution; Encodes the residual and transformed quantized blocks to generate image coded data having a selected resolution, encodes information about the selected resolution to generate resolution data, encodes motion information having a selected resolution, and
  • a method of decoding an image comprising: restoring motion information, resolution information, and residual blocks by decoding a bitstream; Converting a reference picture indicated by the reconstructed motion information into a resolution identified by the reconstructed resolution information; Generating a reference block indicated by motion information reconstructed in the reference picture whose resolution is converted into a prediction block; Reconstructing the current block by adding the reconstructed residual block and the prediction block; And converting the resolution of the restored current block to the original resolution.
  • an apparatus for decoding an image comprising: a decoder for decoding a bitstream to reconstruct motion information, resolution information, and residual blocks that have been transformed and quantized; An inverse quantizer and an inverse converter that inverse quantizes and inverse transforms the quantized residual block to restore the residual block; A resolution adjuster for converting the reference picture indicated by the reconstructed motion information into a resolution identified by the reconstructed resolution information; A motion compensator for generating a reference block indicated by the motion information reconstructed in the reference picture whose resolution is converted into a prediction block; And an adder for reconstructing the current block by adding the reconstructed residual block and the prediction block, wherein the resolution adjuster converts the resolution of the reconstructed current block to the original resolution.
  • the compression efficiency can be improved by adaptively selecting a resolution for each predetermined coding unit and encoding the image at the selected resolution.
  • FIG. 1 is a block diagram schematically illustrating a video encoding apparatus according to an embodiment of the present invention
  • FIG. 2 is an exemplary diagram illustrating a process of converting a resolution of a block according to an embodiment of the present invention
  • FIG. 3 is a flowchart illustrating a video encoding method according to an embodiment of the present invention.
  • FIG. 4 is a flowchart illustrating an example of a method of selecting a resolution according to an embodiment of the present invention
  • FIG. 5 is a flowchart for explaining another example of a method for selecting a resolution according to an embodiment of the present invention.
  • FIG. 6 is a block diagram schematically illustrating an image decoding apparatus according to an embodiment of the present invention.
  • FIG. 7 is a flowchart illustrating an image decoding method according to an embodiment of the present invention.
  • a video encoding apparatus (Video Encoding Apparatus), a video decoding apparatus (Video Decoding Apparatus) to be described below is a personal computer (PC), notebook computer, personal digital assistant (PDA), portable multimedia player (PMP) : User terminal such as Portable Multimedia Player (PSP), PlayStation Portable (PSP: PlayStation Portable), Mobile Communication Terminal (Mobile Communication Terminal), or a server terminal such as an application server or a service server, and communicates with various devices or wired / wireless communication networks.
  • the present invention refers to various devices including a communication device such as a communication modem, a memory for storing various programs and data for encoding or decoding an image, a microprocessor for executing and operating a program, and the like.
  • the image encoded in the bitstream by the video encoding apparatus is real-time or non-real-time through the wired or wireless communication network, such as the Internet, local area wireless communication network, wireless LAN network, WiBro network, mobile communication network, or the like, or a cable, universal serial bus (USB: Universal) It may be transmitted to an image decoding apparatus through various communication interfaces such as a serial bus, and may be decoded by the image decoding apparatus to restore and reproduce the image.
  • wired or wireless communication network such as the Internet, local area wireless communication network, wireless LAN network, WiBro network, mobile communication network, or the like, or a cable, universal serial bus (USB: Universal) It may be transmitted to an image decoding apparatus through various communication interfaces such as a serial bus, and may be decoded by the image decoding apparatus to restore and reproduce the image.
  • USB universal serial bus
  • a moving picture is composed of a series of pictures, and each picture is divided into a predetermined area such as a block.
  • the divided blocks are largely classified into intra blocks and inter blocks according to encoding methods.
  • An intra block refers to a block that is encoded by using an intra prediction coding method.
  • An intra prediction coding is performed by using pixels of blocks that are previously encoded, decoded, and reconstructed in a current picture that performs current encoding.
  • a prediction block is generated by predicting pixels of a block, and a difference value with pixels of the current block is encoded.
  • An inter block refers to a block that is encoded using inter prediction coding.
  • Inter prediction coding generates a prediction block by predicting a current block within a current picture by referring to one or more past or future pictures, and then generates a current block. This is a method of encoding a difference value with.
  • a picture referred to for encoding or decoding the current picture is referred to as a reference picture.
  • An image encoding apparatus is a block, a slice, a picture when encoding an image including an image of a luminance component such as Y, Cr, and Cb and a color difference component.
  • an image having a selected resolution is encoded by adaptively selecting one of a plurality of resolutions according to various types of coding units such as a group of pictures (GOP).
  • GOP group of pictures
  • an arbitrary slice in an image of an image input to an image encoding apparatus is composed of 10 macroblocks, and each macroblock has six 64x64 pixel macroblocks and four 32x32 pixel macros.
  • the video encoding apparatus selects a horizontal resolution and a vertical resolution of 4: 1 for six 64x64 pixel size macroblocks and converts the encoded to 16x16 pixel sized macroblocks.
  • a horizontal resolution and a vertical resolution may be selected as 2: 1 and converted into a 16x16 pixel size macroblock, which is a block of the corresponding resolution, for encoding.
  • the macroblock of a large size is down-sampled to various resolutions, converted into a small macroblock, and then the macroblock of a small size
  • Predictive coding can be performed in units.
  • a 124x124 pixel sized macroblock, a 64x64 pixel sized macroblock, and a 32x32 pixel sized macroblock are down sampled at resolutions of 8: 1, 4: 1, and 2: 1, respectively, to 16x16 pixel sized blocks.
  • the prediction encoding may be performed in units of blocks of 16 ⁇ 16 pixels.
  • the resolution is 1/2 n (where n is equal to 1/16 resolution, 1/8 resolution, 1/4 resolution, 1/2 resolution, 1 resolution, 2 resolution, 4 resolution, 8 resolution, etc.).
  • Resolution represents a ratio of down-sampling or up-sampling based on an image of a predetermined unit, such as a block, slice, picture, or picture group, to be encoded. For example, assuming that the block to be encoded is a 16x16 pixel block, the original resolution of the block of 16x16 pixel block to be encoded is 1 resolution. A 16x16 pixel block is converted into a 4x4 pixel block. In this way, the resolution of 1/4 resolution can be achieved by down sampling the original block by 1/4 times.
  • the original resolution of the block of 16x16 pixel size which is the block to be encoded, becomes 1 resolution, and when the original resolution is converted to 4 resolution, The block is converted into a block of 64x64 pixel size. In this way, the resolution of 4 resolution can be achieved by up-sampling the original block four times.
  • Such resolution may include a horizontal resolution and a vertical resolution.
  • the horizontal resolution and the vertical resolution may not always be the same, and each may have independent resolution.
  • the horizontal resolution and the vertical resolution of the block may all be the same at 1/2 resolution, but the horizontal resolution and the vertical resolution may be different at 1/2 resolution and 1/4 resolution, respectively.
  • an image is encoded in a block unit and the resolution of the image is also selected in block units and predictively encoded according to the selected resolution.
  • the resolution must be encoded in block units or selected in block units.
  • the resolution may be coded in a slice unit, a picture unit, or a picture group unit, or a resolution may be selected.
  • FIG. 1 is a block diagram schematically illustrating a video encoding apparatus according to an embodiment of the present invention.
  • the image encoding apparatus 100 is an apparatus for encoding an image, and includes a resolution converter 110, a resolution selector 120, a motion estimator 130, and a motion.
  • the resolution converter 110 converts a current block of an image into a current block of a plurality of resolutions. That is, the resolution converter 110 generates a current block having a plurality of resolutions by converting the resolution of the current block, which is a block to be encoded in the input image.
  • the plurality of resolutions may be preset or adaptively determined according to characteristics of the input image such as the size of the current block. For example, two resolutions, such as 1/2 resolution and 1/4 resolution, may be set in advance in the resolution converter 110 as a plurality of resolutions, and the resolution converter 110 may be set to 1 / based on the size of the current block. Three resolutions, such as 2 resolution, 1/4 resolution, and 1/8 resolution, may be set as plural resolutions.
  • the plurality of resolutions may be set or determined to have different resolutions in the horizontal direction and in the vertical direction.
  • each resolution may have a horizontal 1/2 resolution and a vertical 1/2 resolution, a horizontal 1/2 resolution and a vertical 1/4 resolution, and a horizontal resolution. It may be in a quarter resolution in the direction and a half resolution in the vertical direction.
  • FIG. 2 is an exemplary diagram illustrating a process of converting a resolution of a block according to an embodiment of the present invention.
  • FIG. 2 exemplarily illustrates a case in which a block is converted to four resolutions when the current block is a NxN pixel size block.
  • the current block can be converted to various resolutions regardless of the block of the luminance (Y) component and the block of the color difference (Cr, Cb) component.
  • NxN pixel size block is converted to a horizontal resolution of 1/4 and a vertical resolution of 1/4, it is downsampled four times in the horizontal direction and the vertical direction, respectively, so that a block of N / 4xN / 4 pixel size is obtained.
  • a block of N / 4xN / 4 pixel size is obtained.
  • the resolution selector 120 selects one of an original resolution and a plurality of resolutions. That is, the resolution selector 120 selects one resolution from the original resolution of the current block to be encoded in the input image and the plurality of resolutions of the current block of the plurality of resolutions converted by the resolution converter 110.
  • the resolution selector 120 may select one of an original resolution and a plurality of resolutions using, for example, a distortion cost. That is, the resolution selector 120 reconverts the current blocks of the plurality of resolutions to the original resolution to generate a plurality of reconstructed blocks, calculates the distortion cost between the current blocks of the original resolution and the plurality of reconstructed blocks, and distorts the resolution. If the cost is calculated and there is one or more distortion costs smaller than the threshold value among the distortion costs by resolution, the resolution having the smallest distortion cost among the distortion costs smaller than the threshold value is selected as one resolution, and the threshold value among the distortion costs by resolution is selected. If there is no distortion cost smaller than the value, the original resolution can be selected as one resolution.
  • a distortion cost for example, the resolution selector 120 reconverts the current blocks of the plurality of resolutions to the original resolution to generate a plurality of reconstructed blocks, calculates the distortion cost between the current blocks of the original resolution and the plurality of reconstructed blocks, and distorts the resolution.
  • sum of absolute difference (SAD) and sum of squared difference (SSD) may be used as the distortion cost.
  • the reconversion of the resolution of such a block may be performed by the resolution selector 120, but the resolution selector 120 may control the resolution converter 110 to convert and then receive the converted block.
  • the current block is a 64x64 pixel block (A block)
  • the original resolution of the current block is 1 resolution
  • the plurality of resolutions are 1/4 resolution and 1/8 resolution.
  • the 64x64 pixel A block is downsampled to 2x and 4x in both the horizontal and vertical directions, respectively, and converted into a 16x16 pixel block (A 1 / 4D ) and an 8x8 pixel block (A 1 / 8D ).
  • the downsampled blocks are again sampled up to 2x and 4x in both horizontal and vertical directions, respectively, and reconverted into two 64x64 pixel reconstruction blocks (A 1 / 8DU and A 1 / 8DU ).
  • the resolution selector 120 calculates the distortion cost for each of the A block, which is the current block of the original resolution, and the two reconstructed blocks A 1 / 8DU and A 1 / 8DU that have been reconverted , and is calculated as described above.
  • the distortion cost becomes the distortion cost for each resolution.
  • SAD may be used as the distortion cost. Therefore, the distortion cost for each resolution can be calculated as the distortion cost SAD (A, A 1 / 4D ) for 1/4 resolution and the distortion cost SAD (A, A 1 / 8D ) for 1/8 resolution.
  • the resolution selector 120 compares the distortion cost SAD (A, A 1 / 4D ) for 1/4 resolution and the distortion cost SAD (A, A 1 / 8D ) for 1/8 resolution by comparing the threshold with a preset threshold. It is determined whether there is a distortion cost smaller than the value.
  • the resolution selector 120 selects the original resolution of the current block as the resolution for the block or a predetermined coding unit if no distortion cost smaller than the threshold exists, and only one distortion cost smaller than the threshold exists.
  • the distortion cost smaller than the threshold value is selected as the resolution for the corresponding block or predetermined coding unit, and if both are smaller than the threshold value, SAD (A, A 1 / 4D ) and SAD (A, A 1 / 8D ), the resolution having the smallest value is selected as the resolution for the corresponding block or a predetermined coding unit.
  • the resolution selector 120 may select one of an original resolution and a plurality of resolutions using a rate-distortion cost. That is, the resolution selector 120 predictively encodes the current block of the original resolution and the current block of the plurality of resolutions to generate image encoded data for each resolution, calculates encoding cost for the image encoded data for each resolution, and The resolution having the smallest coding cost among the coding costs can be selected as one resolution. As described above, although the resolution selector 120 may perform the prediction encoding on the current block for each resolution, the motion selector 120 may perform the motion estimator 130, the motion compensator 140, the subtractor 150, the transformer, and the quantization.
  • encoder 170 After operation 160, encoder 170, inverse quantizer and inverse converter 180, adder 182, resolution adjuster 190, resolution selector 120 is used to calculate the rate-distortion cost.
  • the rate-distortion cost may be calculated by receiving the rate and distortion values from the motion compensator 140 and the encoder 170. If the resolution selector 120 selects the resolution by using the rate-distortion cost, since image coded data for each resolution has already been generated, the encoder 170 does not perform a separate predictive encoding process. Image encoded data may be included in the bitstream.
  • the motion estimator 130 generates motion information by estimating the motion of the current block of the selected resolution in the reference picture converted to the selected resolution. That is, the motion estimator 130 estimates the motion of the current block and generates motion information including a motion vector and a reference picture index. In this case, the motion estimator 130 estimates the motion of the current block by using the reference picture whose resolution of the reference picture stored in the memory 194 in the resolution adjuster 190 is converted to the resolution selected by the resolution selector 120.
  • the motion compensator 140 generates a predicted block of the selected resolution by compensating for the motion of the current block having the selected resolution using the motion information in the reference picture converted to the selected resolution. That is, the motion compensator 140 generates a prediction block by compensating for the motion of the current block by using the motion information of the current block estimated by the motion estimator 130.
  • the motion compensator 140 is indicated by the reference picture index of the motion information among the reference pictures whose resolution of the reference picture stored in the memory 194 in the resolution adjuster 190 is converted to the resolution selected by the resolution selector 120.
  • a block indicated by a motion vector of motion information in the reference picture is generated as a predictive block.
  • the subtractor 150 subtracts the current block of the selected resolution and the prediction block of the selected resolution to generate a residual block of the selected resolution. That is, the subtractor 150 subtracts the current block of the resolution selected by the resolution selector 120 and the prediction block generated by the motion compensator 140 to generate a residual block of the resolution selected by the resolution selector 120. do.
  • the residual block includes a residual signal, and the residual signal has a difference value between the pixel value of the pixel of the current block having the selected resolution and the pixel value of the pixel of the prediction block having the selected resolution.
  • Transducer and quantizer 160 transforms and quantizes the remaining blocks of the selected resolution. That is, the transformer and the quantizer 160 convert the residual signal of the residual block generated by the subtractor 150 into the frequency domain to generate a transformed residual block having a transform coefficient and convert the residual signal.
  • the transform coefficients of the residual block are quantized to produce transformed and quantized residual blocks.
  • a transform method used a method of transforming an image signal in a spatial domain into a frequency domain such as a Hadamard transform or a discrete cosine transform based integer transform may be used.
  • various quantization techniques such as Dead Zone Uniform Threshold Quantization (DZUTQ) or Quantization Weighted Matrix (DZUTQ) may be used as the quantization scheme.
  • DZUTQ Dead Zone Uniform Threshold Quantization
  • DZUTQ Quantization Weighted Matrix
  • the encoder 170 encodes the transformed and quantized residual blocks to generate image coded data having a selected resolution, encodes information about the selected resolution, generates resolution data, and encodes motion information having a selected resolution to encode motion information data. And generates a bitstream including video encoding data, resolution data, and motion information data. That is, the encoder 170 encodes the quantized transform coefficients of the residual block transformed and quantized by the transformer and the quantizer 160 to generate image coded data having a selected resolution, and the selection transmitted from the resolution selector 120. Encode resolution information, which is information about the resolution, to generate resolution data, encode motion information transmitted from the motion estimator 130 to generate motion information, and image coded data, resolution data, and motion information data of a selected resolution.
  • the encoder 170 may use an entropy encoding technique.
  • the present disclosure is not limited thereto and various other encoding techniques may be used. could be.
  • the motion estimator 130, the motion compensator 140, the subtractor 150, the converter and the quantizer 160, and the encoder 170 may have a resolution.
  • the motion estimator 130, the motion compensator 140, the subtractor 150, the converter and the quantizer 160, and the encoder 170 may have a resolution.
  • the resolution selector 120 selects the resolution using the rate-distortion cost
  • the motion estimator 130, the motion compensator 140, the subtractor 150, the converter and quantizer 160, and the encoder 170 For each of the original resolutions and the plurality of resolutions, i.e., the prediction block is generated by estimating and compensating the motion of the current block for each resolution, and the residual block is generated, and the residual block is transformed and quantized to be encoded. Create The image encoded data having the resolution selected by the resolution selector 120 among the image encoded data for each resolution generated as described above is included in the bitstream.
  • Inverse quantizer and inverse transformer 180 inverse quantizes and inverse transforms the transformed and quantized residual blocks to reconstruct the residual blocks of the selected resolution. That is, the inverse quantizer and inverse transformer 180 inverse quantizes the transformed and quantized residual blocks transmitted from the transformer and quantizer 160 to restore the residual block having the transform coefficients, and reconstructs the residual block having the transform coefficients. Inverse transform is used to recover the residual block with the residual signal. In this case, the inverse quantizer and the inverse transformer 180 may reconstruct the residual image by performing the transformed and quantized method in the inverse of the transformer and the quantizer 160.
  • the inverse quantizer and inverse transformer 180 does not inversely quantize and inversely transform all the transformed and quantized residual images transmitted from the transformer and quantizer 160, depending on the resolution selected by resolution selector 120. Only the transformed and quantized residual blocks may be inversely quantized and inversely transformed to restore only the residual image having the selected resolution.
  • the adder 182 adds the reconstructed residual block to the prediction block of the selected resolution to reconstruct the current block of the selected resolution.
  • the resolution adjuster 190 converts the resolution of the current block to be restored to the original resolution, stores the reference picture, and converts the reference picture to the selected resolution.
  • the resolution adjuster 190 may be configured to include an up sampler 192, a memory 194, and a down sampler 194 as shown.
  • the configuration of the resolution adjuster 190 is only an example when the resolution converter 110 downsamples the current block having the original resolution and converts the plurality of resolutions. Accordingly, when the resolution converter 110 upsamples the current block of the original resolution and converts the current block to a plurality of resolutions, the arrangement of the up sampler 192 and the down sampler 194 may be interchanged. If 110 may downsample or upsample the current block at its original resolution, upsampler 192 and downsampler 196 perform both upsampling and downsampling without performing only upsampling or downsampling. can do.
  • the up sampler 192 upsamples the current block reconstructed by the adder 182 to convert the reconstructed current block of the selected resolution into a reconstructed current block of the original resolution.
  • the up sampler 192 may determine whether to perform upsampling and an upsampling scheme using the resolution information transmitted from the resolution selector 120. For example, if the resolution selected by the resolution selector 120 is 1/2 resolution in both the horizontal and vertical directions, the resolution of the current block restored by the adder 182 is also 1/2 in both the horizontal and vertical directions. Therefore, the current block restored by the adder 182 may be upsampled twice in both the horizontal direction and the vertical direction to be converted into the restored current block of the original resolution.
  • the resolution selected by the resolution selector 120 is 1 resolution in both the horizontal and vertical directions, that is, the original resolution
  • the resolution of the current block restored by the adder 182 is the original resolution. 192 may bypass the current block restored by adder 182 to memory 194 without performing upsampling.
  • the memory 194 is a memory for storing a reference picture.
  • the memory 194 accumulates the current block transferred from the up sampler 192 in a picture unit and is stored as a reference picture. Blocks stored in memory 194 are always stored at the original resolution of the current block.
  • the down sampler 196 downsamples the reference picture to convert the reference picture stored in the memory 194 to the resolution selected by the resolution selector 120. To this end, the down sampler 196 may determine whether to perform down sampling and the down sampling method using the resolution information transmitted from the resolution selector 120. For example, if the resolution selected by the resolution selector 120 is 1/2 resolution in both the horizontal and vertical directions, the reference picture of the original resolution stored in the memory 194 is downsampled to 1/2 resolution to be horizontal. The reference picture converted to 1/2 resolution in both the direction and the vertical direction is transmitted to the motion estimator 130 and the motion compensator 140.
  • the down sampler 196 is equal to the resolution of the reference picture stored in the memory 194.
  • the reference picture stored in the memory 194 is transferred to the motion estimator 130 and the motion compensator 140 without performing down sampling.
  • the image encoding apparatus 100 may further include an intra predictor for intra prediction, a deblocking filter, and the like.
  • the subtractor 150 may generate a residual block by subtracting the prediction block generated by the current block and the intra predictor
  • the transformer and quantizer 160 and the inverse quantizer and the inverse transformer 180 are residual blocks. Calculation may be further performed for transform and quantization for and inverse transform and inverse quantization for transform and quantized residual blocks.
  • the encoder 170 may generate image encoded data by encoding the transformed and quantized residual blocks, which are included in the bitstream.
  • the deblocking filter deblocks the input image to be reconstructed.
  • the deblocking filtering refers to an operation of reducing block distortion generated by encoding an image in block units, and applying a deblocking filter to a block boundary and a macroblock boundary, or applying a deblocking filter only to a macroblock boundary or a deblocking filter. You can optionally use one of the methods that does not use.
  • FIG. 3 is a flowchart illustrating an image encoding method according to an embodiment of the present invention.
  • the image encoding apparatus 100 converts a current block of an image into a current block of a plurality of resolutions (S310), and resolution of one of an original resolution and a plurality of resolutions. (S320), and generates a bitstream including image coded data in which the current block of the selected resolution is predictively encoded, resolution data encoded in information on the selected resolution, and motion information data encoded in motion information (S330). ).
  • the image encoding apparatus 100 may downsample the current block having the original resolution at a plurality of resolutions.
  • the resolution may include a horizontal resolution and a vertical resolution, but the horizontal resolution and the vertical resolution may have independent resolutions.
  • the image encoding apparatus 100 may select one resolution using the distortion cost. A method of selecting one resolution using the distortion cost will be described in detail with reference to FIG. 4 in a later process.
  • the image encoding apparatus 100 estimates the motion of a current block having a resolution selected from a reference picture whose resolution is converted to the selected resolution. Generates motion information and generates a prediction block of the selected resolution using the motion information, subtracts the current block of the selected resolution and the prediction block of the selected resolution, and generates a residual block of the selected resolution, and Transforms and quantizes the residual block, encodes the residual block to be transformed and quantized, generates image coded data having a selected resolution, generates resolution data by encoding information about the selected resolution, and encodes motion information to encode motion information.
  • the image encoding apparatus 100 reconstructs the residual block of the resolution selected by inverse quantization and inverse transformation of the transformed and quantized residual block, and adds the reconstructed residual block to the prediction block of the selected resolution to obtain a resolution of the selected resolution.
  • the current block may be restored and the reference picture of the original resolution may be stored by converting the resolution of the restored current block to the original resolution.
  • the image encoding apparatus 100 may select one resolution using the rate-distortion cost.
  • a method of selecting one resolution using the rate-distortion cost will be described in detail with reference to FIG. 5 in a later process.
  • the image encoding apparatus 100 may insert the resolution data into one or more of a block header, a slice header, and a picture header of the bitstream.
  • FIG. 4 is a flowchart illustrating an example of a method of selecting a resolution according to an embodiment of the present invention.
  • FIG. 4 illustrates a method of selecting a resolution using a distortion cost as an example of a method of selecting one of the original resolution and the plurality of resolutions in step S320 of FIG. 3.
  • the image encoding apparatus 100 generates a plurality of reconstructed blocks by reconverting the current blocks having the plurality of resolutions to the original resolutions (S410), and calculates a distortion cost between the current blocks having the original resolutions and the plurality of reconstructed blocks to resolve the resolution. If the distortion cost per resolution is determined (S420), it is determined whether one or more distortion costs smaller than the threshold value among the distortion costs by resolution (S430), and if there is one or more distortion costs smaller than the threshold value among the distortion costs by resolution The resolution having the smallest distortion cost among the distortion costs smaller than the threshold value is selected as one resolution (S440), and if there is no distortion cost smaller than the threshold value among the distortion costs by resolution, the original resolution may be selected as one resolution. There is (S450).
  • a function such as SAD, sum of squared difference (SSD), or the like may be used.
  • FIG. 5 is a flowchart illustrating another example of a method of selecting a resolution according to an embodiment of the present invention.
  • FIG. 5 illustrates a method of selecting a resolution using a rate-distortion cost as another example of a method of selecting one of the original resolution and the plurality of resolutions in step S320 of FIG. 3.
  • the image encoding apparatus 100 When the resolution is selected using the rate-distortion cost, the image encoding apparatus 100 generates image coded data for each resolution by predictively encoding the current block having the original resolution and the current block having the plurality of resolutions (S510). The encoding cost of the image encoded data for each resolution is calculated (S520), and a resolution having the smallest encoding cost among the encoding costs for each resolution may be selected as one resolution (S530).
  • the image encoding apparatus 100 since the image encoding apparatus 100 has all the image encoded data for each resolution in which the current block is predictively encoded for each resolution, when the resolution is selected, the image encoding apparatus 100 may include the image encoded data having the selected resolution in the bitstream and output the same.
  • FIG. 6 is a block diagram schematically illustrating an image decoding apparatus according to an embodiment of the present invention.
  • the image decoding apparatus 600 may include a decoder (Decoder, 610), an inverse quantizer and an inverse converter (620), a motion compensator (630), an adder (640), and a resolution adjuster (650). It can be configured to include.
  • the decoder 610 decodes the bitstream to restore motion information, resolution information, and residual blocks that have been transformed and quantized. That is, the decoder 610 extracts the motion information data, the resolution data, and the image encoded data from the bitstream, decodes the motion information data to restore the motion information, decodes the resolution data to restore the resolution information, and encodes the image. Decode the data to recover the transformed and quantized residual blocks. In this case, the decoder 610 may decode by using an encoding technique such as entropy encoding. The decoder 610 may reversely decode the encoder 170 described above with reference to FIG. 1.
  • Inverse quantizer and inverse transformer 620 inverse quantizes and inverse transforms the quantized residual block to restore the residual block. That is, the inverse quantizer and inverse transformer 620 inverse quantizes and inverse transforms the quantized transform coefficients of the transformed and quantized residual block delivered by the decoder 610 to restore the residual block having the residual signal. In this case, the inverse quantizer and the inverse transformer 620 may perform inverse quantization and inverse transformation by performing a process of converting and quantizing the converter and quantizer 160 described above with reference to FIG. 1.
  • the motion compensator 630 generates a reference block indicated by motion information reconstructed in the reference picture whose resolution is converted into a prediction block. That is, the motion compensator 630 is a reference block indicated by motion information reconstructed by the decoder 610 in a reference picture whose resolution is converted according to the resolution identified by the resolution information reconstructed by the decoder 610. Is generated as a prediction block.
  • the adder 640 reconstructs the current block by adding the residual block to be recovered and the prediction block. That is, the adder 640 reconstructs the current block by adding the residual block reconstructed by the inverse quantizer and the inverse transformer 620 and the predictive block generated by the motion compensator 630.
  • the resolution adjuster 650 converts the reference picture indicated by the reconstructed motion information to the resolution identified by the reconstructed resolution information, and converts the resolution of the current block to be reconstructed to the original resolution.
  • the resolution of the current block to be restored may be converted to the original resolution, accumulated in picture units, and output as a reconstructed image or stored as a reference picture.
  • the resolution adjuster 650 may include an up sampler 652, a memory 654, and a down sampler 656, as shown.
  • the configuration of the resolution adjuster 650 is only an example when the resolution information to be restored is a fractional resolution. Therefore, when the resolution information to be restored is an integer resolution, the arrangement of the up sampler 652 and the down sampler 656 may be interchanged, and if the resolution information to be restored may be a fractional resolution or an integer resolution, the up sampler ( 652 and the down sampler 656 may perform both upsampling and downsampling without performing only upsampling or downsampling.
  • the up sampler 652 upsamples the current block reconstructed by the adder 640 to convert the reconstructed current block of the resolution identified by the reconstructed resolution information into a reconstructed current block of the original resolution.
  • the up sampler 652 may determine whether to perform upsampling and an upsampling scheme according to the motion information transmitted from the decoder 610. For example, if the resolution identified by the reconstructed resolution information is 1/2 resolution in both the horizontal and vertical directions, the resolution of the current block reconstructed by the adder 640 is also 1/2 in both the horizontal and vertical directions.
  • the current block reconstructed by the adder 640 may be upsampled twice in both the horizontal direction and the vertical direction to be converted into the reconstructed current block of the original resolution. If the resolution identified by the resolution information to be restored is one resolution in both the horizontal direction and the vertical direction, that is, the original resolution, the resolution of the current block restored by the adder 640 is the original resolution. ) May bypass the current block restored by the adder 640 to the memory 654 without performing upsampling.
  • the memory 654 is a memory for storing a reference picture, and the current blocks transferred from the up sampler 652 are accumulated in units of pictures and stored as a reference picture. Blocks stored in memory 194 are always stored at the original resolution of the current block.
  • the down sampler 656 downsamples the reference picture to convert the reference picture stored in the memory 654 to the resolution identified by the reconstructed resolution information. To this end, the down sampler 656 may determine whether to perform down sampling and a down sampling method using the resolution information transmitted from the decoder 610. For example, if the resolution identified by the reconstructed resolution information is 1/2 resolution both in the horizontal direction and in the vertical direction, the horizontal direction is performed by down sampling the reference picture of the original resolution stored in the memory 654 to 1/2 resolution. The reference picture, which is converted to 1/2 resolution in both the and vertical directions, is transmitted to the motion compensator 630.
  • the down sampler 656 is down because it is the same as the resolution of the reference picture stored in the memory 654.
  • the reference picture stored in the memory 654 is transferred to the motion compensator 630 without performing sampling.
  • the image decoding apparatus 600 may include an intra predictor for intra prediction, a deblocking filter for deblocking filtering the reconstructed current block, and the like. It may further comprise.
  • FIG. 7 is a flowchart illustrating an image decoding method according to an embodiment of the present invention.
  • the image decoding apparatus 600 decodes a bitstream to restore motion information, resolution information, and residual blocks (S710), and the reference indicated by the restored motion information. Converts the picture to a resolution identified by the reconstructed resolution information (S720), generates a reference block indicated by motion information reconstructed from the reference picture whose resolution is converted into a prediction block (S730), and The prediction block is added to recover the current block (S740), and the resolution of the restored current block is converted to the original resolution (S750). The current block converted to the original resolution is accumulated in picture units and output as a reconstructed image.
  • the resolution may include a horizontal resolution and a vertical resolution, but the horizontal resolution and the vertical resolution may have independent resolutions.
  • the image decoding apparatus 600 may recover resolution information by extracting and decoding resolution data from at least one of a block header, a slice header, and a picture header of the bitstream.
  • the present invention is applied to an image processing or image compression field in which an image is encoded or decoded by changing the resolution of a video, and the image is adaptively selected for each predetermined coding unit and the image is encoded at the selected resolution. It is a very useful invention to generate an effect that can improve the compression efficiency.

Landscapes

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

Abstract

La présente invention porte sur un procédé et un appareil de codage/décodage d'images sur la base d'une résolution adaptive. La présente invention décrit un procédé de codage d'image caractérisé par les étapes consistant à : convertir un bloc présent d'une image en blocs présents ayant une pluralité de résolutions ; sélectionner l'une de la résolution initiale et des résolutions converties ; et générer un train binaire comprenant des données de codage d'image dans lesquelles le bloc présent ayant la résolution sélectionnée est prédit, des données de résolution dans lesquelles les informations pour la résolution sélectionnée sont codées, et des données d'informations de mouvement dans lesquelles des informations de mouvement sont codées. La présente invention permet d'améliorer le rendement de compression par sélection adaptative de la résolution de l'image dans chacune des unités de codage prédéterminées et de coder l'image ayant la résolution sélectionnée.
PCT/KR2010/007420 2009-10-28 2010-10-27 Procédé et appareil de codage/décodage d'images sur la base d'une résolution adaptative WO2011052990A2 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR1020090102664A KR101379190B1 (ko) 2009-10-28 2009-10-28 적응적인 해상도 기반의 영상 부호화/복호화 방법 및 장치
KR10-2009-0102664 2009-10-28

Publications (2)

Publication Number Publication Date
WO2011052990A2 true WO2011052990A2 (fr) 2011-05-05
WO2011052990A3 WO2011052990A3 (fr) 2011-10-13

Family

ID=43922823

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2010/007420 WO2011052990A2 (fr) 2009-10-28 2010-10-27 Procédé et appareil de codage/décodage d'images sur la base d'une résolution adaptative

Country Status (2)

Country Link
KR (1) KR101379190B1 (fr)
WO (1) WO2011052990A2 (fr)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110545436A (zh) * 2014-03-24 2019-12-06 索尼公司 图像编码装置及方法和图像解码装置及方法
CN110572654A (zh) * 2019-09-27 2019-12-13 腾讯科技（深圳）有限公司 视频编码、解码方法和装置、存储介质及电子装置
CN112911311A (zh) * 2021-01-28 2021-06-04 杭州朗和科技有限公司 编码方法、编码装置、存储介质及电子设备
CN112954360A (zh) * 2021-01-28 2021-06-11 杭州朗和科技有限公司 解码方法、解码装置、存储介质及电子设备
CN112997503A (zh) * 2019-01-02 2021-06-18 腾讯美国有限责任公司 用于帧间预测和显示的自适应图片分辨率重新缩放

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102085270B1 (ko) * 2013-08-12 2020-03-05 삼성전자 주식회사 가장 작은 왜곡 값을 갖는 해상도를 선택하는 이미지 처리 방법과 상기 방법을 수행하는 장치들
CN117981316A (zh) * 2021-09-23 2024-05-03 Lg电子株式会社 图像编码/解码方法、发送比特流的方法及存储比特流的记录介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20040056996A (ko) * 2002-12-24 2004-07-01 한국전자통신연구원 웨이브렛 변환 영역에서의 움직임 벡터 예측 방법 및 장치
KR20070039552A (ko) * 2004-06-23 2007-04-12 에이전시 포 사이언스, 테크놀로지 앤드 리서치 그리드 움직임 추정 및 보상을 갖는 스캘러블 비디오 코딩
JP2009206911A (ja) * 2008-02-28 2009-09-10 Mitsubishi Electric Corp 動画像変換装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20040056996A (ko) * 2002-12-24 2004-07-01 한국전자통신연구원 웨이브렛 변환 영역에서의 움직임 벡터 예측 방법 및 장치
KR20070039552A (ko) * 2004-06-23 2007-04-12 에이전시 포 사이언스, 테크놀로지 앤드 리서치 그리드 움직임 추정 및 보상을 갖는 스캘러블 비디오 코딩
JP2009206911A (ja) * 2008-02-28 2009-09-10 Mitsubishi Electric Corp 動画像変換装置

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110545436A (zh) * 2014-03-24 2019-12-06 索尼公司 图像编码装置及方法和图像解码装置及方法
CN112997503A (zh) * 2019-01-02 2021-06-18 腾讯美国有限责任公司 用于帧间预测和显示的自适应图片分辨率重新缩放
CN112997503B (zh) * 2019-01-02 2024-02-02 腾讯美国有限责任公司 对已编码视频序列进行解码的方法、装置及存储介质
CN110572654A (zh) * 2019-09-27 2019-12-13 腾讯科技（深圳）有限公司 视频编码、解码方法和装置、存储介质及电子装置
CN110572654B (zh) * 2019-09-27 2024-03-15 腾讯科技（深圳）有限公司 视频编码、解码方法和装置、存储介质及电子装置
CN112911311A (zh) * 2021-01-28 2021-06-04 杭州朗和科技有限公司 编码方法、编码装置、存储介质及电子设备
CN112954360A (zh) * 2021-01-28 2021-06-11 杭州朗和科技有限公司 解码方法、解码装置、存储介质及电子设备

Also Published As

Publication number Publication date
WO2011052990A3 (fr) 2011-10-13
KR20110045912A (ko) 2011-05-04
KR101379190B1 (ko) 2014-03-31

Similar Documents

Publication Publication Date Title
WO2013002549A2 (fr) Procédés et appareil de codage/décodage d'une image
WO2010074407A2 (fr) Procédé et appareil pour le codage/décodage d'une image en couleur
WO2013070006A1 (fr) Procédé et appareil de codage et de décodage vidéo faisant appel à un mode de saut
WO2010044569A2 (fr) Procédé et appareil permettant de générer une trame de référence et procédé et appareil permettant le coder/décoder une image au moyen de la trame de référence
WO2012077960A2 (fr) Procédé et dispositif de codage/décodage d'une image par inter-prédiction en utilisant un bloc aléatoire
WO2012033344A2 (fr) Procédé et appareil destinés à coder/décoder des images à l'aide de la sélection efficace d'un groupe de mode de prédiction intra
WO2012091461A2 (fr) Procédé et dispositif de codage / décodage d'une image en utilisant des vecteurs caractéristiques de blocs environnants
WO2011052990A2 (fr) Procédé et appareil de codage/décodage d'images sur la base d'une résolution adaptative
WO2013002550A2 (fr) Méthode et appareil de codage/décodage par décision de mode de l'unité de codage grande vitesse
WO2010047499A2 (fr) Appareil de codage/décodage vidéo, filtre de dégroupage et procédé de filtrage basé sur des directions de prédiction intra, et support d'enregistrement
WO2013062196A1 (fr) Appareil de décodage d'images
WO2013141671A1 (fr) Procédé et appareil de prédiction intra inter-couche
WO2012093854A2 (fr) Procédé et dispositif de codage et de décodage utilisant une prédiction interne parallèle par une unité de codage
WO2010047498A2 (fr) Procédé et dispositif de codage/décodage d'images utilisant des coefficients d'un filtre d'interpolation adaptatif
WO2012015275A2 (fr) Procédé et dispositif de codage/décodage d'image à l'aide d'une prédiction fragmentée de bloc
WO2011037337A2 (fr) Procédé et appareil de codage-décodage d'images tenant compte de composantes basse fréquence
WO2012077959A2 (fr) Procédé et dispositif pour coder/décoder une image dans un domaine spatial pour une composante de bruit
WO2012021040A2 (fr) Procédé et dispositif de codage/décodage d'image ayant un mode de filtrage amovible
KR20130095238A (ko) 고정밀 필터를 이용한 영상 부호화/복호화 방법 및 장치
WO2011111954A2 (fr) Procédé et appareil de codage/décodage de vecteur de mouvement, à l'aide de combinaison de résolutions de vecteur de mouvement, et procédé et appareil de codage/décodage d'images utilisant ceux-ci
WO2018074626A1 (fr) Procédé et appareil de codage vidéo utilisant un filtre d'interpolation adaptatif
WO2011068332A2 (fr) Procédé et appareil de prédiction spatiale, procédé et dispositif de codage d'image, et procédé et dispositif de décodage d'image utilisant lesdits procédé et appareil de prédiction
WO2012011683A2 (fr) Procédé et dispositif pour le codage en sous-bandes d'une unité de conversion de fréquence, et procédé et dispositif pour le codage/décodage d'images faisant intervenir ceux-ci
WO2011025301A2 (fr) Procédé et appareil permettant de coder/décoder des vecteurs de mouvement pour un codage vidéo, et procédé et appareil permettant de coder/décoder des images
WO2011053054A2 (fr) Procédé et appareil de codage/décodage de vecteur de mouvement au moyen d'une limitation de résolution de vecteur de mouvement, et procédé et appareil de codage/décodage d'image dans lesquels ils interviennent

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10827074

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS (EPO FORM 1205A DATED 02-08-2012)

122 Ep: pct application non-entry in european phase

Ref document number: 10827074

Country of ref document: EP

Kind code of ref document: A2