WO2015196929A1 - 一种帧内深度图像块编码、解码的方法及装置 - Google Patents

一种帧内深度图像块编码、解码的方法及装置 Download PDF

Info

Publication number
WO2015196929A1
WO2015196929A1 PCT/CN2015/081078 CN2015081078W WO2015196929A1 WO 2015196929 A1 WO2015196929 A1 WO 2015196929A1 CN 2015081078 W CN2015081078 W CN 2015081078W WO 2015196929 A1 WO2015196929 A1 WO 2015196929A1
Authority
WO
WIPO (PCT)
Prior art keywords
block
depth image
value
image sub
mode
Prior art date
Application number
PCT/CN2015/081078
Other languages
English (en)
French (fr)
Inventor
陈旭
郑萧桢
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2015196929A1 publication Critical patent/WO2015196929A1/zh
Priority to US15/389,196 priority Critical patent/US10218985B2/en
Priority to US16/261,097 priority patent/US10951901B2/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134Methods 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
    • H04N19/157Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
    • H04N19/159Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134Methods 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
    • H04N19/146Data rate or code amount at the encoder output
    • H04N19/147Data rate or code amount at the encoder output according to rate distortion criteria
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/46Embedding additional information in the video signal during the compression process
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/593Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/597Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding specially adapted for multi-view video sequence encoding
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/90Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
    • H04N19/96Tree coding, e.g. quad-tree coding
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/172Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a picture, frame or field
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/182Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a pixel
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/184Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being bits, e.g. of the compressed video stream

Definitions

  • the present invention relates to the field of coding and decoding technologies, and in particular, to a method and an apparatus for encoding and decoding intra-frame depth image blocks.
  • the image encoding process is that the encoding end obtains the prediction information of the encoded image and the encoded image, and further obtains the residual of the encoded image.
  • the decoding end decodes the image before encoding according to the prediction information and the residual of the image.
  • DMM Depth modeling modes
  • DMM technology can be implemented in combination with RQT technology. According to the characteristics of the depth map, when the DMM is applied to the RQT process, the DMM can be implemented in four modes, and the four modes can be marked by numbers. In this way, when encoding and decoding a depth image block, it is necessary to detect the four modes, and the two modes are also required to be detected during decoding, resulting in high complexity of codec and low codec efficiency.
  • the embodiments of the present invention provide a method for encoding and decoding intra-frame depth image blocks, which can reduce the detection mode in the encoding and decoding process, thereby reducing the complexity of coding and decoding, and improving the coding.
  • the efficiency of decoding Embodiments of the present invention also provide corresponding devices.
  • a first aspect of the present invention provides a method for intra-frame depth block coding, including:
  • the depth image block is respectively performed using the DMM1 mode and the DMM4 mode in the DMM. Detecting, obtaining a rate distortion result of the depth image block in the DMM1 mode, and a rate distortion result of the depth image block in the DMM4 mode;
  • Determining a DMM mode in which the rate distortion result in the DMM1 and the DMM4 is minimized as a DMM mode used in encoding applying the used mode in the RQT or the SDC to performing the depth image block Encode and write the DMM mode used to the code stream.
  • the method when the DMM mode is used to write a code stream, the method further includes:
  • the applying the mode to apply to the depth image in the RQT or the SDC is encoded, including:
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the method before the obtaining the residual information of each pixel in the depth image sub-block, the method further includes:
  • the original pixel value of each pixel in the same depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • the adding by using the block prediction value of the same depth image sub-block and the block offset value, The value of the region constant of the same depth image sub-block, including:
  • the predicted value of the depth image sub-block with the block offset value of 0 is used as the depth image sub-block with the block offset value of 0
  • the value of the region constant is 0
  • a second aspect of the present invention provides a method for decoding an intra-depth image block, including:
  • Decoding the depth image block by applying the mode of the DMM used in encoding to a recursive quadtree RQT or a simplified depth map encoding SDC;
  • a decoded depth image block is obtained.
  • the method further includes: when acquiring, from a code stream, a mode of decoding a DMM to be decoded, where the DMM is used for encoding, the method further includes:
  • the applying the mode of the DMM used in encoding to the recursive quadtree RQT or simplifying the depth Decoding the depth image block in the picture coding SDC including:
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original image of the depth image block.
  • the difference between the prime value and the predicted value of the block is further mapped according to a preset mapping relationship, and the residual information of each pixel in the depth image sub-block is the pixel of each pixel in the depth image sub-block.
  • the method before the obtaining the residual information of each pixel in the depth image sub-block, the method further includes:
  • the original pixel value of each pixel in the same depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • the adding the block prediction value of the same depth image sub-block and the block offset value including:
  • the predicted value of the depth image sub-block with the block offset value of 0 is used as the depth image sub-block with the block offset value of 0
  • the value of the region constant is 0
  • a third aspect of the present invention provides an encoding apparatus, including:
  • An obtaining unit configured to acquire a depth image block to be encoded
  • a detecting unit configured to use the DMM1 in the DMM when encoding the depth image block acquired by the acquiring unit in the recursive quadtree RQT or the simplified depth map encoding SDC by applying the depth modeling model DMM
  • the mode and the DMM4 mode detect the depth image block, and obtain a rate distortion result of the depth image block in the DMM1 mode, and a rate distortion result of the depth image block in the DMM4 mode;
  • a coding unit configured to determine a DMM mode in which the rate distortion result of the DMM1 and the DMM4 detected by the detecting unit is minimized as a DMM mode used in encoding, and apply the used mode to the RQT Or encoding the depth image block in the SDC and writing the used DMM mode to the code stream.
  • the coding unit is further configured to write, to the code stream, a coded identifier that is used to indicate the RQT or the SDC, where the coded identifier is used to indicate that the decoding end is in accordance with the RQT indicated by the coded identifier or The SDC performs decoding.
  • the coding unit is specifically configured to obtain, according to the used DMM mode, a block prediction value, a block offset value, and a depth of an image sub-block in the depth image sub-block obtained by using the DMM mode. Residual information of each pixel;
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the coding unit is specifically configured to add a block prediction value of the same depth image sub-block and the block offset value to obtain an area constant value of the same depth image sub-block, and the same one
  • the original pixel value of each pixel in the depth image sub-block is different from the regional constant value, and the residual information of each pixel in the depth image sub-block is obtained.
  • the coding unit is specifically configured to: when a block offset value in at least one of the depth image sub-blocks is 0, use a predicted value of a depth image sub-block with a block offset value of 0 as the block offset The value of the region constant of the depth image sub-block with a value of 0.
  • a fourth aspect of the present invention provides a decoding apparatus, including:
  • An acquiring unit configured to obtain, from the code stream, a mode of decoding a DMM used by the depth image block to be decoded at the time of encoding
  • a decoding unit configured to apply, in a recursive quadtree RQT or a simplified depth map encoding SDC, the mode of the DMM used in encoding to decode the depth image block acquired by the acquiring unit;
  • an obtaining unit configured to obtain a depth image block decoded by the decoding unit.
  • the obtaining unit is further configured to acquire, from the code stream, an identifier used to indicate that the RQT or the SDC is used when encoding;
  • the decoding unit is specifically configured to use the RQT or the SDC according to the encoding. And identifying, by using the mode of the DMM used in encoding, the depth image block in the RQT or the SDC indicated by the identifier.
  • the decoding unit is specifically configured to obtain, according to the mode of the DMM used in the encoding, a block prediction value and a block offset value of the depth image sub-block obtained by dividing the mode of the DMM used in encoding. And residual information of each pixel in the depth image sub-block;
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the decoding unit is specifically configured to add a block prediction value of the same depth image sub-block and the block offset value to obtain a region constant value of the same depth image sub-block, and the same one
  • the original pixel value of each pixel in the depth image sub-block is different from the regional constant value, and the residual information of each pixel in the depth image sub-block is obtained.
  • the decoding unit is specifically configured to: when a block offset value in at least one of the depth image sub-blocks is 0, use a predicted value of a depth image sub-block with a block offset value of 0 as the block offset The value of the region constant of the depth image sub-block with a value of 0.
  • the embodiment of the present invention adopts acquiring a depth image block to be encoded; when the depth modeling model DMM is applied to recursive quadtree RQT or simplified depth map coding SDC, the depth image block is encoded, and the DMM is used respectively.
  • a DMM mode in which DMM1 and the DMM4 have the lowest rate distortion result is determined as a DMM mode used in encoding, and the used mode is applied to encode the depth image block in the RQT or the SDC, and The used DMM mode is written to the code stream.
  • the coding method provided by the embodiment of the present invention only needs to detect two modes, thereby reducing the complexity of the coding and improving the coding. effectiveness.
  • FIG. 1 is a schematic diagram of an embodiment of an encoding method in an embodiment of the present invention.
  • FIG. 2 is a schematic diagram of a DMM 1 in an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of a DMM 4 in an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of an embodiment of a decoding method in an embodiment of the present invention.
  • FIG. 5 is a schematic diagram showing a process of codec in the embodiment of the present invention.
  • FIG. 6 is a schematic diagram showing another process of codec in the embodiment of the present invention.
  • FIG. 7 is a schematic diagram showing another process of codec in the embodiment of the present invention.
  • FIG. 8 is a schematic diagram of another process of decoding in an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of another process of decoding in an embodiment of the present invention.
  • FIG. 10 is a schematic diagram of an embodiment of an encoding apparatus according to an embodiment of the present invention.
  • FIG. 11 is a schematic diagram of an embodiment of a decoding apparatus according to an embodiment of the present invention.
  • FIG. 12 is a schematic diagram of an embodiment of an encoding apparatus according to an embodiment of the present invention.
  • FIG. 13 is a schematic diagram of an embodiment of a decoding apparatus according to an embodiment of the present invention.
  • the embodiments of the present invention provide a method for encoding and decoding intra-frame depth image blocks, which can reduce the detection mode in the encoding and decoding process, thereby reducing the complexity of encoding and decoding and improving the efficiency of encoding and decoding.
  • Embodiments of the present invention also provide corresponding devices. The details are described below separately.
  • an embodiment of a method for intra-frame depth image block coding includes:
  • DMM depth modeling model
  • RQT Recur Intra Coding Quart Tree
  • SDC Simplify Depth Coding
  • DMM Depth modeling modes
  • DMM Depth modeling modes
  • DMM1 (Depth modeling mode1) uses a wedge-shaped method to segment image blocks.
  • DMM4 (Depth modeling) Mode4) is based on the corresponding texture reference block for predictive segmentation.
  • the DMM1 can be understood with reference to FIG. 2, and the DMM4 can be understood with reference to FIG. 3.
  • the depth image block will have a rate distortion result in the DMM1 mode, and there will be a rate distortion result in the DMM4.
  • the calculation process of the rate distortion result is prior art, and will not be described in detail in this application.
  • the DMM mode with small selection rate distortion result is combined with the preset coding technique.
  • DMM1 is combined with RQT technology when encoding using RQT technology.
  • DMM1 When DMM1 is used, when RQT encoding is used, DMM1 is combined with RQT to encode the depth image block, and the DMM mode of DMM1 is written in the code stream.
  • DMM1 is represented by 0
  • DMM4 is represented by 1.
  • DMM1 is represented by 0
  • DMM4 is represented by 1.
  • the embodiment of the present invention adopts acquiring a depth image block to be encoded; and applying the depth modeling model DMM to the recursive quadtree RQT or the simplified depth map encoding SDC to edit the depth image block.
  • the depth image block is detected by using the DMM1 mode and the DMM4 mode in the DMM, respectively, to obtain a rate distortion result of the depth image block in the DMM1 mode, and the depth image in the DMM4 mode.
  • the coding method provided by the embodiment of the present invention only needs to detect two modes, thereby reducing the complexity of coding and improving the coding efficiency.
  • the used DMM mode is used in the foregoing embodiment of FIG.
  • the method further includes:
  • the DMM1 or DMM4 mode is marked by the default offset flag bit or 0.
  • the DMM is combined with the RQT according to whether the first offset flag is 0 or there is no first offset identifier.
  • the bit, and the second offset flag is 1 or there is no second offset flag to mark the DMM1 or DMM4 mode.
  • RQT or SDC can be used in the encoding and decoding process
  • the encoding technology is specifically used at the encoding end, and the decoding end needs to be notified, so that the decoding end decodes according to the encoding technology. Therefore, it is necessary to write RQT in the code stream. Or the identity of the SDC.
  • an identifier bit can be set. When the flag bit is 0, it indicates that RQT is used. When the flag bit is 1, it indicates that SDC is used.
  • the mode used to encode the depth image block in the RQT or the SDC may include:
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the depth image block is divided according to the DMM1 or divided according to the DMM4 to obtain a corresponding depth image sub-block, and then the depth image sub-block may be separately encoded.
  • the predefined prediction rules may be various, for example, selecting an average value of pixel values of each pixel point in an adjacent depth image block in the upper left corner of the depth image sub-block, or selecting an adjacent one of the left side of the depth image sub-block
  • the average of the pixel values of each pixel in the depth image block can also have a variety of prediction rules, which are not enumerated here.
  • the block raw pixel value of the depth image block is the average of the pixel values of each pixel point.
  • the preset mapping relationship Y X+1, where x is the difference between the block original pixel value and the block prediction value, and Y is the mapped block offset value.
  • the residual information of each pixel in the depth image sub-block is the difference between the original pixel value of each pixel in the depth image sub-block and the sum of the block prediction value and the block offset value.
  • Formula can be expressed as:
  • Z M - (A + Y) where Z is the residual information of the pixel, M is the original pixel value of the pixel, A is the predicted value of the block, and Y is the offset value of the block.
  • the obtaining the depth image sub-block is performed on the basis of the foregoing second optional embodiment.
  • the method may further include:
  • the original pixel value of each pixel in the same depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • the sum of the block prediction value and the block offset value is the area constant (DC) value.
  • the same depth image is used in the fourth optional embodiment.
  • the block prediction value of the sub-block and the block offset value are added to obtain a region constant value of the same depth image sub-block, which may include:
  • the predicted value of the depth image sub-block with the block offset value of 0 is used as the depth image sub-block with the block offset value of 0
  • the value of the region constant is 0
  • the predicted value of the depth image sub-block with the offset value of 0 is used as the depth image sub-block region DC value of the offset value of 0.
  • the DC value of the depth image sub-block area may be calculated as such, or the block offset value of all the depth image sub-blocks may be 0. This is the DC value of the depth image sub-block area.
  • an embodiment of a method for decoding an intra-depth image block includes:
  • the mode of the depth modeling model DMM is one of two modes, DMM1 and DMM4, corresponding to the encoding end.
  • a mode of decoding a DMM used to decode a depth image block to be decoded when encoding is obtained from a code stream; applying the mode of the DMM used in encoding to a recursive quadtree RQT or simplifying depth
  • the depth image block is decoded in the picture coding SDC; the decoded depth picture block is obtained.
  • the decoding method provided by the embodiment of the present invention does not need to detect the offset flag and decode the offset flag in the RQT to 0, thereby reducing the decoding process. Understand the complexity of the code and improve the efficiency of decoding.
  • the decoding is to be decoded from the code stream.
  • the method may further include:
  • the applying the mode of the DMM used in the encoding to the decoding of the depth image block in a recursive quadtree RQT or a simplified depth map encoding SDC may include:
  • the encoding end may perform encoding using RQT technology or SDC technology, and the decoding end needs to determine which encoding technology used by the encoding end to perform encoding, so as to determine a decoding technology corresponding to the encoding technology.
  • the coding technique used in determining the coding may be determined by the identification bit in the coding information. For example, when the identification bit is 0, the RQT coding technique is used when determining the coding, and when the identification bit is 1, the coding is indicated. SDC coding technology is used. In this way, when the flag is 0, the RQT technology can be used for decoding. When the flag is 1, the SDC technology can be used for decoding.
  • the second optional method for decoding the intra-depth image block in the embodiment of the present invention is optional.
  • the second optional method for decoding the intra-depth image block in the embodiment of the present invention is optional.
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the block prediction value is obtained in the same manner as the encoding end, and the decoding is performed. Based on the pixel values of the decoded adjacent depth image blocks, the predicted value of the current depth image may be obtained.
  • the block offset value and the residual information of each pixel in the depth image block are obtained from the code stream at the time of decoding.
  • the method may further include:
  • the original pixel value of each pixel in the same depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • a fourth optional embodiment of the method for decoding a depth image block in the depth image sub-block, the predicted value of the pixel and the offset value are summed to obtain the DC value of the pixel of the pixel in the depth image sub-block, which may include:
  • the predicted value of each pixel in the one depth image sub-block is used as the regional DC value of each pixel .
  • a block prediction value of the image sub-block and the block offset value to obtain a region constant value of the same depth image sub-block, including:
  • the predicted value of the depth image sub-block with the block offset value of 0 is used as the depth image sub-block with the block offset value of 0
  • the value of the region constant is 0
  • FIG. 5 is a schematic diagram of a combined implementation of DMMs in RQT and SDC.
  • the depth image block starts the encoding process.
  • the SDC flag flag can be set to 1
  • RQT encoding if DMM1 is used for encoding, the flag of DMM can be set to 0. Whether it is SDC coding or RQT coding, when DMM1 is used for encoding, the flag of DMM can be set to 0. If DMM4 is used for encoding, the flag of DMM can be set to 1.
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the predefined prediction rules may be various, for example, selecting an average value of pixel values of each pixel point in an adjacent depth image block in the upper left corner of the depth image sub-block, or selecting an adjacent one of the left side of the depth image sub-block
  • the average of the pixel values of each pixel in the depth image block can also have a variety of prediction rules, which are not enumerated here.
  • the block raw pixel value of the depth image block is the average of the pixel values of each pixel point.
  • the preset mapping relationship Y X+1, where x is the difference between the block original pixel value and the block prediction value, and Y is the mapped block offset value.
  • the residual information of each pixel in the depth image sub-block is the difference between the original pixel value of each pixel in the depth image sub-block and the sum of the block prediction value and the block offset value.
  • Formula can be expressed as:
  • Z M - (A + Y) where Z is the residual information of the pixel, M is the original pixel value of the pixel, A is the predicted value of the block, and Y is the offset value of the block.
  • the sum of the block prediction value and the block offset value is the area constant (DC) value.
  • the residual information of each pixel in the depth image sub-block is further obtained according to the difference between the original pixel value of each pixel in the depth image block and the corresponding regional constant DC value.
  • the decoding end acquires the coding information to be decoded
  • the SDC Flag identifier bit and the identifier on the DMM Flag identifier bit are extracted, and the decoding technology and the DMM mode are determined according to the representation.
  • the specific process of implementing deep image block reconstruction can be:
  • the DMM mode uses the flag bit (isDimMode) to 1, indicating that the DMM mode is used, and the DMM mode flag (DimType) is 0, indicating that DMM1 is used. Further, DeltaDC0, DeltaDC1, and residual information are encoded. The decoding end decodes the DMM mode, and the identification bit (isDimMode) is 1. It can be known that the DMM mode is used in the encoding, and the DMM mode identification bit (DimType) is further decoded, and decoded to 0, indicating that DMM1 is used in encoding, and DeltaDC0 is used in encoding.
  • DeltaDC1 performs decoding to obtain the corresponding values 8 and 13, and finally decodes the residual information.
  • the depth image block is intra-coded
  • the 8 ⁇ 8-size depth image block block is intra-coded
  • the DMM1 mode in the RQT is selected for encoding, and the depth image block is performed according to the DMM1.
  • the residual information of each pixel in the depth image block shown in FIG. 7 can be obtained.
  • the DMM mode uses the flag bit (isDimMode) to 1, indicating that the DMM mode is used, and the DMM mode flag (DimType) is 0, indicating that DMM1 is used. Further, DeltaDC0, DeltaDC1 and residual information are encoded. The decoding end decodes the DMM mode, and the identification bit (isDimMode) is 1. It can be known that the DMM mode is used in the encoding, and the DMM mode identification bit (DimType) is further decoded, and decoded to 0, indicating that DMM1 is used in encoding, and DeltaDC0 is used in encoding. DeltaDC1 performs decoding to obtain corresponding values 3 and 0, and finally decodes the residual information to obtain residual information of the depth image block shown in FIG.
  • an embodiment of an encoding apparatus 30 includes:
  • An obtaining unit 301 configured to acquire a depth image block to be encoded
  • the detecting unit 302 is configured to use the DMM in the depth image block when the depth modeling model DMM is applied in the recursive quadtree RQT or the simplified depth map encoding SDC to encode the depth image block acquired by the acquiring unit 301 And detecting, by the DMM1 mode and the DMM4 mode, the depth image block, the rate distortion result of the depth image block in the DMM1 mode, and the rate distortion result of the depth image block in the DMM4 mode;
  • the encoding unit 303 is configured to determine a DMM mode in which the rate distortion result of the DMM1 and the DMM4 detected by the detecting unit 302 is minimized as a DMM mode used in encoding, and apply the used mode to the
  • the depth image block is encoded in the RQT or the SDC, and the used DMM mode is written into the code stream.
  • the acquiring unit 301 obtains a depth image block to be encoded, and the detecting unit 302 applies the depth modeling model DMM to the recursive quadtree RQT or the simplified depth map encoding SDC.
  • the depth image block acquired by the acquiring unit 301 is encoded, the depth image block is detected by using the DMM1 mode and the DMM4 mode in the DMM, respectively, to obtain the rate of the depth image block in the DMM1 mode.
  • the encoding unit 303 determines a DMM mode in which the DMM1 and the DMM4 have the lowest rate distortion result detected by the detecting unit 302 as encoding time
  • the DMM mode used, the used mode is applied to encode the depth image block in the RQT or the SDC, and the used DMM mode is written into the code stream.
  • the coding method provided by the embodiment of the present invention only needs to detect two modes, thereby reducing the complexity of coding and improving the coding efficiency.
  • the encoding unit 303 is further configured to: write, to the code stream, an encoding identifier that is used to indicate the RQT or the SDC, where the encoding identifier is used to indicate that the decoding end is in accordance with the RQT indicated by the encoding identifier. Or the SDC performs decoding.
  • the encoding unit 303 is specifically configured to obtain, according to the used DMM mode, a block prediction value, a block offset value, and the depth image sub-block of the depth image sub-block obtained by using the DMM mode. Residual information of each pixel in the middle;
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the encoding unit 303 is specifically configured to add the block prediction value of the same depth image sub-block and the block offset value to obtain a region constant value of the same depth image sub-block, and the same A raw pixel value of each pixel in a depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • the encoding unit 303 is specifically configured to: when a block offset value in at least one of the depth image sub-blocks is 0, use a predicted value of a depth image sub-block with a block offset value of 0 as the block bias The value of the region constant of the depth image sub-block with a value of 0.
  • an embodiment of a decoding apparatus 40 includes:
  • the obtaining unit 401 is configured to obtain, from the code stream, a mode for decoding a DMM used by the depth image block to be decoded when encoding;
  • a decoding unit 402 configured to apply, in a recursive quadtree RQT or a simplified depth map encoding SDC, the mode of the DMM used in encoding to decode the depth image block acquired by the acquiring unit 401;
  • the obtaining unit 403 is configured to obtain a depth image block decoded by the decoding unit 402.
  • the obtaining unit 401 acquires, from the code stream, a mode of decoding the DMM used when the depth image block to be decoded is encoded; the decoding unit 402 applies the mode of the DMM used in the encoding to the recursive four.
  • the decoding method provided by the embodiment of the present invention does not need to detect the offset flag and decode the offset flag in the RQT to 0, thereby reducing the decoding process. Understand the complexity of the code and improve the efficiency of decoding.
  • the obtaining unit 401 is further configured to: obtain, from the code stream, an identifier used to indicate that the RQT or the SDC is used when encoding;
  • the decoding unit 402 is specifically configured to apply, according to the identifier of the RQT or the SDC, the mode of the DMM used in encoding to the RQT or the location indicated by the identifier.
  • the depth image block is decoded in the SDC.
  • the decoding unit 402 is specifically configured to use, according to the mode of the DMM used in encoding, Obtaining a block prediction value of the depth image sub-block obtained by dividing the mode of the DMM used in encoding, a block offset value, and residual information of each pixel point in the depth image sub-block;
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block.
  • the difference between the block prediction values and the mapping is performed according to a preset mapping relationship.
  • the residual information of each pixel in the depth image sub-block is the original of each pixel in the depth image sub-block.
  • the decoding unit 402 is specifically configured to add a block prediction value of the same depth image sub-block and the block offset value to obtain a region constant value of the same depth image sub-block, and the same A raw pixel value of each pixel in a depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • the decoding unit 402 is specifically configured to: when a block offset value in at least one of the depth image sub-blocks is 0, use a predicted value of a depth image sub-block with a block offset value of 0 as the block bias The value of the region constant of the depth image sub-block with a value of 0.
  • FIG. 12 is a schematic structural diagram of an encoding apparatus 30 according to an embodiment of the present invention.
  • Encoding device 30 may include input device 310, output device 320, processor 330, and memory 340.
  • Memory 340 can include read only memory and random access memory and provides instructions and data to processor 330. A portion of the memory 340 may also include non-volatile random access memory (NVRAM).
  • NVRAM non-volatile random access memory
  • the memory 340 stores the following elements, executable modules or data structures, or a subset thereof, or an extended set thereof:
  • Operation instructions include various operation instructions for implementing various operations.
  • Operating system Includes a variety of system programs for implementing various basic services and handling hardware-based tasks.
  • the processor 330 performs the following operations by calling an operation instruction stored in the memory 340, which can be stored in the operating system:
  • the depth image block is respectively performed using the DMM1 mode and the DMM4 mode in the DMM. Detecting, obtaining a rate distortion result of the depth image block in the DMM1 mode, and a rate distortion result of the depth image block in the DMM4 mode;
  • Determining a DMM mode in which the rate distortion result in the DMM1 and the DMM4 is minimized as a DMM mode used in encoding applying the used mode in the RQT or the SDC to performing the depth image block Encode and write the DMM mode used to the code stream.
  • the encoding device 30 only needs to detect two modes when encoding, thereby reducing the complexity of encoding and improving the efficiency of encoding.
  • the processor 330 controls the operation of the encoding device 30, which may also be referred to as a CPU (Central Processing Unit).
  • Memory 340 can include read only memory and random access memory and provides instructions and data to processor 330. A portion of the memory 340 may also include non-volatile random access memory (NVRAM).
  • NVRAM non-volatile random access memory
  • the various components of the encoding device 30 are coupled together by a bus system 350.
  • the bus system 350 may include a power bus, a control bus, a status signal bus, and the like in addition to the data bus. However, for clarity of description, various buses are labeled as bus system 350 in the figure.
  • Processor 330 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the foregoing method may be completed by an integrated logic circuit of hardware in the processor 330 or an instruction in a form of software.
  • the processor 330 described above may be a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or discrete hardware. Component.
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA off-the-shelf programmable gate array
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present invention may be implemented or carried out.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present invention may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in memory 340, and processor 330 reads the information in memory 340 and, in conjunction with its hardware, performs the steps of the above method.
  • the processor 330 may further: write, to the code stream, an encoding identifier used to indicate the RQT or the SDC, where the encoding identifier is used to indicate that the decoding end is instructed by the encoding identifier The RQT or the SDC is decoded.
  • the processor 330 may: obtain, according to the used DMM mode, a block prediction value, a block offset value, and the depth image sub-block of the depth image sub-block obtained by using the DMM mode. Residual information of each pixel in the block;
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the original pixel value of each pixel in the depth image sub-block. A difference from a sum of the block prediction value and the block offset value.
  • the processor 330 specifically:
  • the original pixel value of each pixel in the same depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • the processor 330 specifically:
  • the predicted value of the depth image sub-block with the block offset value of 0 is used as the depth image sub-block with the block offset value of 0
  • the value of the region constant is 0
  • FIG. 13 is a schematic structural diagram of a decoding apparatus 40 according to an embodiment of the present invention.
  • the decoding device 40 may include an input device 410, an output device 420, a processor 430, and a memory 440.
  • Memory 440 can include read only memory and random access memory and provides instructions and data to processor 430. A portion of memory 440 may also include non-volatile random access memory (NVRAM).
  • NVRAM non-volatile random access memory
  • Memory 440 stores the following elements, executable modules or data structures, or subsets thereof, or their extended sets:
  • Operation instructions include various operation instructions for implementing various operations.
  • Operating system includes various system programs for implementing various basic services and processing based on hard The task of the piece.
  • the processor 430 performs the following operations by calling an operation instruction stored in the memory 440 (the operation instruction can be stored in the operating system):
  • Decoding the depth image block by applying the mode of the DMM used in encoding to a recursive quadtree RQT or a simplified depth map encoding SDC;
  • a decoded depth image block is obtained.
  • the decoding device 40 does not need to detect the offset flag and decode the offset flag in the RQT, thereby reducing the complexity of the decoding and improving the decoding efficiency.
  • the processor 430 controls the operation of the decoding device 40, which may also be referred to as a CPU (Central Processing Unit).
  • Memory 440 can include read only memory and random access memory and provides instructions and data to processor 430. A portion of memory 440 may also include non-volatile random access memory (NVRAM).
  • NVRAM non-volatile random access memory
  • the components of the decoding device 40 are coupled together by a bus system 450.
  • the bus system 450 may include a power bus, a control bus, a status signal bus, and the like in addition to the data bus. However, for clarity of description, various buses are labeled as bus system 450 in the figure.
  • Processor 430 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the foregoing method may be completed by an integrated logic circuit of hardware in the processor 430 or an instruction in a form of software.
  • the processor 430 described above may be a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or discrete hardware. Component.
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA off-the-shelf programmable gate array
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present invention may be implemented or carried out.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present invention may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory 440 and the processor 430 reads The information in the memory 440 is taken together with the hardware to complete the steps of the above method.
  • the processor 430 may further obtain, from the code stream, an identifier used to indicate that the RQT or the SDC is used when encoding;
  • the processor may specifically: apply the mode of the DMM used in encoding to the RQT or the SDC indicated by the identifier according to the identifier of the RQT or the SDC when the encoding is used. Decoding the depth image block.
  • processor 430 specifically:
  • the block prediction value is obtained from a depth image block adjacent to the depth image sub-block according to a predefined prediction rule, where the block offset value is a block original pixel value of the depth image block. And the residual information of each pixel in the depth image sub-block is the pixel of each pixel in the depth image sub-block.
  • processor 430 specifically:
  • the original pixel value of each pixel in the same depth image sub-block is compared with the regional constant value to obtain residual information of each pixel in the depth image sub-block.
  • processor 430 specifically:
  • the predicted value of the depth image sub-block with the block offset value of 0 is used as the depth image sub-block with the block offset value of 0
  • the value of the region constant is 0
  • the storage medium may include: a ROM, a RAM, a magnetic disk, or an optical disk.

Landscapes

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

Abstract

本发明公开了一种帧内深度图像块编码的方法,包括:获取待编码的深度图像块,当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1和DMM4模式下所述深度图像块的率失真结果,将所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。本发明实施例提供的编码方法,可以减少编码过程中的检测模式,从而降低了编码的复杂度。

Description

一种帧内深度图像块编码、解码的方法及装置
本申请要求于2014年6月26日提交中国专利局、申请号为201410300985.X、发明名称为“一种帧内深度图像块编码、解码的方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明涉及编码解码技术领域,具体涉及一种帧内深度图像块编码、解码的方法及装置。
背景技术
目前图像编码的过程就是编码端获得编码图像和编码图像的预测信息,进而得到编码图像的残差,在解码端,解码端根据图像的预测信息和残差,解码出编码前的图像。
目前在3D编解码技术中有多种编解码技术,如:递归四叉树编码(Recur Intra Coding Quart Tree,RQT)技术等,3D编解码中为了能够反映出图像深度信息,需要对深度图进行编解码。因此较传统纹理图编解码增加了深度建模模型(Depth modeling modes,DMM)的图像块分割方法。DMM有两种模式,一种是DMM1模式,一种是DMM4模式。其中DMM1(Depth modeling mode1)是采用了楔形方式对图像块进行分割,DMM4(Depth modeling mode4)则是根据对应的纹理参考块进行预测分割。
DMM技术可以结合RQT技术进行实现。根据深度图特点,在DMM应用于RQT过程中,DMM可以通过四种模式进行实现,四种模式可以通过数字进行标记。这样,在对深度图像块进行编解码时,就需要对这四种模式进行检测,解码时也需要对两种模式进行检测,导致编解码的复杂度高,编解码效率低。
发明内容
本发明实施例提供本发明提出一种帧内深度图像块编码、解码的方法,可以减少编解码过程中的检测模式,从而降低了编解码的复杂度,提高了编 解码的效率。本发明实施例还提供了相应的装置。
本发明第一方面提供一种帧内深度图像块编码的方法,包括:
获取待编码的深度图像块;
当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;
将所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。
结合第一方面,在第一种可能的实现方式中,所述将所述所使用的DMM模式,写入码流时,所述方法还包括:
将用于指示所述RQT或所述SDC的编码标识写入所述码流,所述编码标识用于指示解码端按照所述编码标识所指示的所述RQT或所述SDC进行解码。
结合第一方面或第一方面第一种可能的实现方式,在第二种可能的实现方式中,所述将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,包括:
根据所述所使用的DMM模式,获得所述所使用的DMM模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
结合第一方面第二种可能的实现方式,在第三种可能的实现方式中,所述获得所述深度图像子块中各像素点的残差信息之前,所述方法还包括:
将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
结合第一方面第三种可能的实现方式,在第四种可能的实现方式中,所述将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,包括:
当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
本发明第二方面提供一种帧内深度图像块解码的方法,包括:
从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;
将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码;
获得解码后的深度图像块。
结合第二方面,在第一种可能的实现方式中,所述从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式时,所述方法还包括:
从所述码流中获取用于指示编码时使用所述RQT或所述SDC的标识;
所述将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码,包括:
根据所述编码时使用所述RQT或所述SDC的标识,将所述在编码时所使用的DMM的模式应用在所述标识所指示的所述RQT或所述SDC中对所述深度图像块进行解码。
结合第二方面或第二方面第一种可能的实现方式,在第二种可能的实现方式中,所述将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码,包括:
根据所述在编码时所使用的DMM的模式,获得所述在编码时所使用的DMM的模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像 素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
结合第二方面第二种可能的实现方式,在第三种可能的实现方式中,所述获得所述深度图像子块中各像素点的残差信息之前,所述方法还包括:
将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
结合第二方面第二种可能的实现方式,在第四种可能的实现方式中,所述将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,包括:
当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
本发明第三方面提供一种编码装置,包括:
获取单元,用于获取待编码的深度图像块;
检测单元,用于当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述获取单元获取的所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;
编码单元,用于将所述检测单元检测得到的所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。
结合第三方面,在第一种可能的实现方式中,
所述编码单元,还用于将用于指示所述RQT或所述SDC的编码标识写入所述码流,所述编码标识用于指示解码端按照所述编码标识所指示的所述RQT或所述SDC进行解码。
结合第三方面或第三方面第一种可能的实现方式,在第二种可能的实现方式中,
所述编码单元,具体用于根据所述所使用的DMM模式,获得所述所使用的DMM模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
结合第三方面第二种可能的实现方式,在第三种可能的实现方式中,
所述编码单元,具体用于将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
结合第三方面第三种可能的实现方式,在第四种可能的实现方式中,
所述编码单元,具体用于当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
本发明第四方面提供一种解码装置,包括:
获取单元,用于从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;
解码单元,用于将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述获取单元获取的所述深度图像块进行解码;
获得单元,用于获得所述解码单元解码后的深度图像块。
结合第四方面,在第一种可能的实现方式中,
所述获取单元,还用于从所述码流中获取用于指示编码时使用所述RQT或所述SDC的标识;
所述解码单元,具体用于根据所述编码时使用所述RQT或所述SDC的标 识,将所述在编码时所使用的DMM的模式应用在所述标识所指示的所述RQT或所述SDC中对所述深度图像块进行解码。
结合第四方面或第四方面第一种可能的实现方式,在第二种可能的实现方式中,
所述解码单元,具体用于根据所述在编码时所使用的DMM的模式,获得所述在编码时所使用的DMM的模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
结合第四方面第二种可能的实现方式,在第三种可能的实现方式中,
所述解码单元,具体用于将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
结合第四方面第三种可能的实现方式,在第四种可能的实现方式中,
所述解码单元,具体用于当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。本发明实施例采用获取待编码的深度图像块;当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;将所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。与现有技术中编解码时需要对四种模式进行检测相比,本发明实施例提供的编码方法,只需要检测两种模式,从而降低了编码的复杂度,提高了编码的 效率。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本发明实施例中的编码方法的一实施例示意图;
图2是本发明实施例中的DMM1的示意图;
图3是本发明实施例中的DMM4的示意图;
图4是本发明实施例中解码方法的一实施例示意图;
图5是本发明实施例中编解码的一过程描述示意图;
图6是本发明实施例中编解码的另一过程描述示意图;
图7是本发明实施例中编解码的另一过程描述示意图;
图8是本发明实施例中解码的另一过程描述示意图;
图9是本发明实施例中解码的另一过程描述示意图;
图10是本发明实施例中编码装置的一实施例示意图;
图11是本发明实施例中解码装置的一实施例示意图;
图12是本发明实施例中编码装置的一实施例示意图;
图13是本发明实施例中解码装置的一实施例示意图。
具体实施方式
本发明实施例提供本发明实施例提供一种帧内深度图像块编码、解码的方法,可以减少编解码过程中的检测模式,从而降低了编解码的复杂度,提高了编解码的效率。本发明实施例还提供了相应的装置。以下分别进行详细说明。
为了使本技术领域的人员更好地理解本发明方案,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分的实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获 得的所有其他实施例,都应当属于本发明保护的范围。
参阅图1,本发明实施例提供的一种帧内深度图像块编码的方法的一实施例包括:
101、获取待编码的深度图像块。
102、当将深度建模模型DMM应用在递归四叉树(Recur Intra Coding Quart Tree,RQT)或简化深度图编码(Simplify Depth Coding,SDC)中对所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果。
深度建模模型(Depth modeling modes,DMM)是一种图像的分割方法,DMM有DMM1和DMM4两种模式,其中DMM1(Depth modeling mode1)是采用了楔形方式对图像块进行分割,DMM4(Depth modeling mode4)则是根据对应的纹理参考块进行预测分割。关于DMM1可以参阅图2进行理解,DMM4可以参阅图3进行理解。
深度图像块在DMM1模式会有一个率失真结果,在DMM4会有一个率失真结果,率失真结果的计算过程为现有技术,本申请中不做过多赘述。
103、将所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。
在DMM1和DMM4的率失真结果都计算出来后,选择率失真结果小的DMM模式结合于预置的编码技术中。
当DMM1的率失真结果较小时,编码时使用RQT技术进行编码时,将DMM1结合于RQT技术。
当使用DMM1时,采用RQT编码时,将DMM1结合于RQT中对深度图像块进行编码,并在码流中写入DMM1这种DMM的模式。
对于在编码信息中DMM的模式的表示方式可以有多种,例如:用0表示DMM1,用1表示DMM4。当然,也可以有其他的表示方式。
本发明实施例采用获取待编码的深度图像块;当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行编 码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;将所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。与现有技术中编解码时需要对四种模式进行检测相比,本发明实施例提供的编码方法,只需要检测两种模式,从而降低了编码的复杂度,提高了编码的效率。
可选地,在上述图1对应的实施例的基础上,本发明实施例提供的帧内深度图像块编码的方法的第一个可选实施例中,所述将所述所使用的DMM模式,写入码流时,所述方法还包括:
将用于指示所述RQT或所述SDC的编码标识写入所述码流,所述编码标识用于指示解码端按照所述编码标识所指示的所述RQT或所述SDC进行解码。
本发明实施例中,当编码器中预置有RQT和SD编码技术时,DMM结合于SDC时,会有两种模式,两种模式分别对应DMM1和DMM4。而现有技术中针对SDC通过默认偏移标识位为0或者没有该偏移标识位来标记DMM1或DMM4模式,DMM结合于RQT时根据第一偏移标识位为0或者没有第一偏移标识位,以及第二偏移标识位为1或者没有第二偏移标识位来标记DMM1或DMM4模式,共有四种模式。而本发明实施例中只设置有DMM1和DMM4两种模式,无论在编解码过程中使用RQT技术还是使用SDC技术,在结合DMM时,都只结合DMM1和DMM4两种模式,进一步降低了编码的复杂度,提高了编码的效率。
虽然RQT或SDC都可以在编解码过程中使用,但在编码端具体使用了那种编码技术,需要通知解码端,以便解码端根据该编码技术进行解码,因此,需要在码流中写入RQT或者SDC的标识。
例如:可以设置一个标识位,当该标识位标识位为0时表示使用RQT,当该标识位为1时表示使用SDC。
可选地,在上述图1对应的实施例或第一个可选实施例的基础上,本发明实施例提供的帧内深度图像块编码的方法的第二个可选实施例中,所述将 所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,可以包括:
根据所述所使用的DMM模式,获得所述所使用的DMM模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
本发明实施例中,参阅图2和图3可知,深度图像块按照DMM1进行分割或者按照DMM4进行分割会得到相应的深度图像子块,然后可以分别对所述深度图像子块进行编码。
预先定义的预测规则可以有多种,例如:选择深度图像子块左上角的相邻的深度图像块的中各像素点的像素值的平均值,或者选择深度图像子块左方的相邻的深度图像块的中各像素点的像素值的平均值,还可以有很多种预测规则,本处不一一枚举。
深度图像块的块原始像素值为每个像素点的像素值的平均值。
预设的映射关系Y=X+1,其中,x为所述块原始像素值与所述块预测值的差值,Y为映射得到的块偏移值。
所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。用公式可以表示为:
Z=M-(A+Y)其中,Z为像素点的残差信息,M为像素点的原始像素值,A为块的预测值,Y为块的偏移值。
可选地,在上述第二个可选实施例的基础上,本发明实施例提供的帧内深度图像块编码的方法的第三个可选实施例中,所述获得所述深度图像子块中各像素点的残差信息之前,所述方法还可以包括:
将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
块预测值和块偏移值的和即为区域常数(DC)值。
可选地,在上述第三个可选实施例的基础上,本发明实施例提供的帧内深度图像块编码的方法的第四个可选实施例中,所述将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,可以包括:
当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
当一个深度图像子块的块偏移值为0时,将所述偏移值为0的深度图像子块的预测值作为所述偏移值为0的深度图像子块区域DC值。
可以是只有一个深度图像子块的块偏移值为0时,就可以这样就算深度图像子块区域DC值,也可以时必须所有深度图像子块的块偏移值都为0时,才可以这样就算深度图像子块区域DC值。
参阅图4,本发明实施例提供的帧内深度图像块解码的方法的一实施例包括:
201、从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式。
深度建模模型DMM的模式为与编码端对应的DMM1和DMM4两种模式中的一种。
202、将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码。
203、获得解码后的深度图像块。
本发明实施例中,从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码;获得解码后的深度图像块。与现有技术中解码时需要对两种模式进行检测相比,本发明实施例提供的解码方法不需要检测偏移标识位以及对RQT中偏移标识位为0的情况进行解码处理,从而降低了解码的复杂度,提高了解码的效率。
可选地,在上述图4对应的实施例的基础上,本发明实施例提供的帧内深度图像块解码的方法的第一个可选实施例中,所述从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式时,所述方法还可以包括:
从所述码流中获取用于指示编码时使用所述RQT或所述SDC的标识;
所述将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码,可以包括:
根据所述编码时使用所述RQT或所述SDC的标识,将所述在编码时所使用的DMM的模式应用在所述标识所指示的所述RQT或所述SDC中对所述深度图像块进行解码。
本发明实施例中,编码端可以使用RQT技术或者SDC技术进行编码,解码端需要确定编码端采用的哪种编码技术进行编码,以便确定与编码技术对应的解码技术。
确定编码时所使用的编码技术可以通过编码信息中的标识位来确定,例如:当该标识位标识位为0时确定编码时使用的是RQT编码技术,当该标识位为1时表示编码时使用的是SDC编码技术。这样,当标识位为0时就可以使用RQT技术进行解码,当标识位为1时就可以使用SDC技术进行解码。
可选地,在上述图4对应的实施例或以图4为基础的第一个可选实施例的基础上,本发明实施例提供的帧内深度图像块解码的方法的第二个可选实施例中,
所述将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码,包括:
根据所述在编码时所使用的DMM的模式,获得所述在编码时所使用的DMM的模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
本发明实施例中,在解码端,块预测值的获取方式与编码端相同,解码 端根据已解码的相邻深度图像块的像素值,可以得到当前所述深度图像的预测值。
块偏移值和深度图像块中各像素点的残差信息都是解码时从码流中获取到的。
可选地,在第二个可选实施例的基础上,本发明实施例提供的帧内深度图像块解码的方法的第三个可选实施例中,
所述获得所述深度图像子块中各像素点的残差信息之前,所述方法还可以包括:
将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
可选地,在解码端的第三个可选实施例的基础上,本发明实施例提供的深度图像块解码的方法的第四个可选实施例中,所述对所述深度图像子块中像素点的预测值和偏移值求和,得到所述深度图像子块中像素点的区域DC值,可以包括:
当一个深度图像子块中的每个像素点的偏移值都为0时,将所述一个深度图像子块中所述每个像素点的预测值作为所述每个像素点的区域DC值。
可选地,在解码端的第三个可选实施例的基础上,本发明实施例提供的帧内深度图像块解码的方法的第三个可选实施例中,所述将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,包括:
当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
为了便于理解,下面以几个应用场景为例,说明本发明实施例中编码和解码的过程:
参阅图5,图5为RQT和SDC中DMM的合并实现示意图。
将SDC编码过程中DMM实现的两种方式与RQT编码过程中DMM实现的四种方式合并为DMM1和DMM4两种模式,具体请参阅图5中S305中的过 程。
S300、深度图像块开始编码过程。
S305、无论是采用SDC编码还是采用RQT编码,根据深度图像块在DMM1和DMM4的率失真结果,确定在编码过程中采用DMM1还是采用DMM4。
S310、确定SDC Flag标识位和DMM Flag标识位的标识。
深度图像块在编码过程中,如果使用SDC编码,则在SDC flag标识位可以设置为1,如果使用RQT编码,则在SDC flag标识位可以设置为0。无论是SDC编码,还是RQT编码,当编码时所使用的是DMM1时,则在DMM的标识位可以设置为0,如果编码时所使用的是DMM4,则在DMM的标识位可以设置为1。
S315、计算深度图像块的预测值、偏移值和残差信息。
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
预先定义的预测规则可以有多种,例如:选择深度图像子块左上角的相邻的深度图像块的中各像素点的像素值的平均值,或者选择深度图像子块左方的相邻的深度图像块的中各像素点的像素值的平均值,还可以有很多种预测规则,本处不一一枚举。
深度图像块的块原始像素值为每个像素点的像素值的平均值。
预设的映射关系Y=X+1,其中,x为所述块原始像素值与所述块预测值的差值,Y为映射得到的块偏移值。
所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。用公式可以表示为:
Z=M-(A+Y)其中,Z为像素点的残差信息,M为像素点的原始像素值,A为块的预测值,Y为块的偏移值。
块预测值和块偏移值的和即为区域常数(DC)值。
深度图像子块中各像素点的的残差信息则进一步根据深度图像字块中每个像素点的原始像素值和对应的区域常数DC值的差值得到。
S316、在编码信息中设置S310中确定的SDC Flag标识位和DMM Flag标识位的标识。
S320、在解码过程中,解码端获取待解码的编码信息后,提取出SDC Flag标识位和DMM Flag标识位上的标识,根据表示确定采用的解码技术和DMM的模式。
S325、根据所述编码时所使用的DMM的模式和所述深度图像块编码时所RQT或者SDC,获取编码时所述深度图像子块中的块预测值和块偏移值,以及所述深度图像块中各像素点的残差信息。
S330、实现深度图像块的重建。
实现深度图像块重建的具体过程可以是:
根据所述编码时所使用的DMM的模式,确定所述深度图像块的深度图像子块块预测值;
从所述码流中获取所述深度图像子块的块偏移值,以及所述深度图像子块中各像素点的残差信息;
对所述深度图像子块的块预测值和块偏移值求和,得到所述深度图像子块的区域常数DC值;
根据所述深度图像子块的区域常数DC值和所述深度图像子块中各像素点的残差信息,得到所述深度图像子块中各像素点的重建值,最终得到深度图像块的重建图像。
参阅图6,假设当对深度图像块进行帧内编解码,当对4×4大小深度图像块像块进行帧内编码,假设选择了RQT中的DMM1模式进行编码实现,根据DMM1对深度图像块进行分割得到两个深度图像子块,计算得到第一深度图像子的块预测值PredDC0=5,块偏移值DeltaDC0=8,计算得到第二深度图像子块的块预测值PredDC1=6,块偏移值DeltaDC1=13,PredDC0=5与DeltaDC0=8的和即为的第一深度图像子的块的区域常数DC值,则可以计算出第一个深度图像子块的区域常数DC值为13,第二个深度图像子块的区域常数DC值为19,再用深度图像块中像素点的原始像素值与对应的区域常数DC值做差值,即可得等得到图6中所示的深度图像块中每个像素点的残差 信息。DMM模式使用标识位(isDimMode)为1,表示使用了DMM模式,且DMM模式标识位(DimType)为0,表示使用的是DMM1。进一步对DeltaDC0、DeltaDC1和残差信息进行编码。解码端解码DMM模式,通过标识位(isDimMode)为1,可知编码时使用了DMM模式,进一步对DMM模式标识位(DimType)解码,解码为0,表示编码时使用的是DMM1,后对DeltaDC0和DeltaDC1进行解码,得到对应值8和13,最后再对残差信息进行解码。解码端可以采用与编码端相同的策略,计算的到第一个深度图像子块的块预测值PredDC0=5,第二个深度图像子块的块预测值PredDC1=6,用PredDC0和DeltaDC0求和得到第一个深度图像子块的区域常数DC0=13,PredDC1和DeltaDC1求和得到第二个深度图像子块的区域常数区域DC1=19,再根据图6所示的残差信息,用区域常数DC0=13和区域常数DC1=19与对应像素点的像素值做求和运算,得到深度图像块中每个像素点的重建值,实现对该4×4大小深度图像块进行重构。
参阅图7,假设当对深度图像块帧内编解码,当对8×8大小深度图像块像块进行帧内编码,假设选择了RQT中的DMM1模式进行编码实现,根据DMM1对深度图像块进行分割得到两个深度图像子块,计算得第一个深度图像子块中每个像素点的块预测值PredDC0=13,块偏移值DeltaDC0=3,计算得到第二个深度图像子块中每个像素点的块预测值PredDC1=7,块偏移值DeltaDC1=0,PredDC0=13与DeltaDC0=3的和即为第一个深度图像子块的区域常数DC值,则可以计算出第一个深度图像子块的区域常数DC0=16,第二个深度图像子块的区域常数DC1=19,再用深度图像块中哥各像素点的原始像素值与对应的的区域常数DC值做差值,即可得等得到图7中所示的深度图像块中每个像素点的残差信息。DMM模式使用标识位(isDimMode)为1,表示使用了DMM模式,且DMM模式标识位(DimType)为0,表示使用的是DMM1。进一步对DeltaDC0,DeltaDC1和残差信息进行编码。解码端解码DMM模式,通过标识位(isDimMode)为1,可知编码时使用了DMM模式,进一步对DMM模式标识位(DimType)解码,解码为0,表示编码时使用的是DMM1,后对DeltaDC0和DeltaDC1进行解码,得到对应值3和0,最后再对残差信息进行解码,得到图7所示的深度图像块的残差信息。解码端可以采用与编码端相同的策略,计算的到第一个深度图像子块的块预测值PredDC0=13,第二个深度图像子块的块预测值PredDC1=7,用PredDC0和 DeltaDC0求和得到第一个深度图像子块的区域常数DC0=16,PredDC1和DeltaDC1求和得到第二个深度图像子块的区域常数DC1=7,因DeltaDC1=0,所以可以直接将PredDC1=7赋值给区域常数DC1=7。再根据图7所示的残差信息,用区域DC0=16和区域DC1=7与对应像素点的残差信息做求和运算,得到深度图像块中每个像素点的重建值,实现对该8×8大小深度图像块进行重构。
参阅图8,假设当对深度图像块帧内编解码,当对4×4大小深度图像块进行帧内编码,假设选择了RQT中的DMM1模式进行编码实现,计算得到PredDC0=5,PredDC1=6,DeltaDC0=0,DeltaDC1=13。由于DeltaDC0=0,则对于区域DC0则可以通过PredDC0直接赋值可得出为DC0=5,对于区域DC1则按照图6或图7所描述的步骤计算得出。
参阅图9,假设当对深度图像块帧内编解码,当对4×4大小深度图像块像块进行帧内编码,假设选择了RQT中的DMM1模式进行编码实现,计算得到PredDC0=5,PredDC1=6,DeltaDC0=0,DeltaDC1=0。由于DeltaDC0=DeltaDC1=0,则对于区域DC0和区域DC1均可以通过PredDC0和PredDC1直接赋值可得出分别为5和6。
参阅图10,本发明实施例提供的编码装置30的一实施例包括:
获取单元301,用于获取待编码的深度图像块;
检测单元302,用于当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述获取单元301获取的所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;
编码单元303,用于将所述检测单元302检测得到的所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。
本发明实施例中,获取单元301获取待编码的深度图像块,检测单元302当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对 所述获取单元301获取的所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;编码单元303将所述检测单元302检测得到的所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。与现有技术中编解码时需要对四种模式进行检测相比,本发明实施例提供的编码方法,只需要检测两种模式,从而降低了编码的复杂度,提高了编码的效率。
可选地,在上述图10对应的实施例的基础上,本发明实施例提供的编码装置的另一实施例中,
所述编码单元303,还用于将用于指示所述RQT或所述SDC的编码标识写入所述码流,所述编码标识用于指示解码端按照所述编码标识所指示的所述RQT或所述SDC进行解码。
可选地,在上述图10对应的实施例的基础上,本发明实施例提供的编码装置的另一实施例中,
所述编码单元303,具体用于根据所述所使用的DMM模式,获得所述所使用的DMM模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
可选地,在上述图10对应的实施例的基础上,本发明实施例提供的编码装置的另一实施例中,
所述编码单元303,具体用于将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
可选地,在上述图10对应的实施例的基础上,本发明实施例提供的编码装置的另一实施例中,
所述编码单元303,具体用于当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
参阅图11,本发明实施例提供的解码装置40的一实施例包括:
获取单元401,用于从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;
解码单元402,用于将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述获取单元401获取的所述深度图像块进行解码;
获得单元403,用于获得所述解码单元402解码后的深度图像块。
本发明实施例中,获取单元401从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;解码单元402将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述获取单元401获取的所述深度图像块进行解码;获得单元403,获得所述解码单元402解码后的深度图像块。与现有技术中解码时需要对两种模式进行检测相比,本发明实施例提供的解码方法不需要检测偏移标识位以及对RQT中偏移标识位为0的情况进行解码处理,从而降低了解码的复杂度,提高了解码的效率。
可选地,在上述图11对应的实施例的基础上,本发明实施例提供的解码装置的另一实施例中,
所述获取单元401,还用于从所述码流中获取用于指示编码时使用所述RQT或所述SDC的标识;
所述解码单元402,具体用于根据所述编码时使用所述RQT或所述SDC的标识,将所述在编码时所使用的DMM的模式应用在所述标识所指示的所述RQT或所述SDC中对所述深度图像块进行解码。
可选地,在上述图11对应的实施例的基础上,本发明实施例提供的解码装置的另一实施例中,
所述解码单元402,具体用于根据所述在编码时所使用的DMM的模式, 获得所述在编码时所使用的DMM的模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原
可选地,在上述图11对应的实施例的基础上,本发明实施例提供的解码装置的另一实施例中,
所述解码单元402,具体用于将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
可选地,在上述图11对应的实施例的基础上,本发明实施例提供的解码装置的另一实施例中,
所述解码单元402,具体用于当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
图12是本发明实施例编码装置30的结构示意图。编码装置30可包括输入设备310、输出设备320、处理器330和存储器340。
存储器340可以包括只读存储器和随机存取存储器,并向处理器330提供指令和数据。存储器340的一部分还可以包括非易失性随机存取存储器(NVRAM)。
存储器340存储了如下的元素,可执行模块或者数据结构,或者它们的子集,或者它们的扩展集:
操作指令:包括各种操作指令,用于实现各种操作。
操作系统:包括各种系统程序,用于实现各种基础业务以及处理基于硬件的任务。
在本发明实施例中,处理器330通过调用存储器340存储的操作指令(该操作指令可存储在操作系统中),执行如下操作:
获取待编码的深度图像块;
当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;
将所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。
本发明实施例中,编码装置30在编码时只需要检测两种模式,从而降低了编码的复杂度,提高了编码的效率。
处理器330控制编码装置30的操作,处理器330还可以称为CPU(Central Processing Unit,中央处理单元)。存储器340可以包括只读存储器和随机存取存储器,并向处理器330提供指令和数据。存储器340的一部分还可以包括非易失性随机存取存储器(NVRAM)。具体的应用中,编码装置30的各个组件通过总线系统350耦合在一起,其中总线系统350除包括数据总线之外,还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见,在图中将各种总线都标为总线系统350。
上述本发明实施例揭示的方法可以应用于处理器330中,或者由处理器330实现。处理器330可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器330中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器330可以是通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现成可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本发明实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本发明实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器340,处理器330读取存储器340中的信息,结合其硬件完成上述方法的步骤。
可选地,处理器330还可:将用于指示所述RQT或所述SDC的编码标识写入所述码流,所述编码标识用于指示解码端按照所述编码标识所指示的所述RQT或所述SDC进行解码。
可选地,处理器330具体可:根据所述所使用的DMM模式,获得所述所使用的DMM模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
可选地,处理器330具体可:
将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
可选地,处理器330具体可:
当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
图13是本发明实施例解码装置40的结构示意图。解码装置40可包括输入设备410、输出设备420、处理器430和存储器440。
存储器440可以包括只读存储器和随机存取存储器,并向处理器430提供指令和数据。存储器440的一部分还可以包括非易失性随机存取存储器(NVRAM)。
存储器440存储了如下的元素,可执行模块或者数据结构,或者它们的子集,或者它们的扩展集:
操作指令:包括各种操作指令,用于实现各种操作。
操作系统:包括各种系统程序,用于实现各种基础业务以及处理基于硬 件的任务。
在本发明实施例中,处理器430通过调用存储器440存储的操作指令(该操作指令可存储在操作系统中),执行如下操作:
从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;
将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码;
获得解码后的深度图像块。
本发明实施例中,解码装置40不需要检测偏移标识位以及对RQT中偏移标识位为0的情况进行解码处理,从而降低了解码的复杂度,提高了解码的效率
处理器430控制解码装置40的操作,处理器430还可以称为CPU(Central Processing Unit,中央处理单元)。存储器440可以包括只读存储器和随机存取存储器,并向处理器430提供指令和数据。存储器440的一部分还可以包括非易失性随机存取存储器(NVRAM)。具体的应用中,解码装置40的各个组件通过总线系统450耦合在一起,其中总线系统450除包括数据总线之外,还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见,在图中将各种总线都标为总线系统450。
上述本发明实施例揭示的方法可以应用于处理器430中,或者由处理器430实现。处理器430可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器430中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器430可以是通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现成可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本发明实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本发明实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器440,处理器430读 取存储器440中的信息,结合其硬件完成上述方法的步骤。
可选地,所述处理器430还可从所述码流中获取用于指示编码时使用所述RQT或所述SDC的标识;
所述处理器具体可:根据所述编码时使用所述RQT或所述SDC的标识,将所述在编码时所使用的DMM的模式应用在所述标识所指示的所述RQT或所述SDC中对所述深度图像块进行解码。
可选地,所述处理器430具体可:
根据所述在编码时所使用的DMM的模式,获得所述在编码时所使用的DMM的模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的。
可选地,所述处理器430具体可:
将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
可选地,所述处理器430具体可:
当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件(例如处理器)来完成,该程序可以存储于一计算机可读存储介质中,存储介质可以包括:ROM、RAM、磁盘或光盘等。
以上对本发明实施例所提供的确定非连续性接收周期的方法、用户设备以及系统进行了详细介绍,本文中应用了具体个例对本发明的原理及实施方 式进行了阐述,以上实施例的说明只是用于帮助理解本发明的方法及其核心思想;同时,对于本领域的一般技术人员,依据本发明的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本发明的限制。

Claims (20)

  1. 一种帧内深度图像块编码的方法,其特征在于,包括:
    获取待编码的深度图像块;
    当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;
    将所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。
  2. 根据权利要求1所述的方法,其特征在于,所述将所述所使用的DMM模式,写入码流时,所述方法还包括:
    将用于指示所述RQT或所述SDC的编码标识写入所述码流,所述编码标识用于指示解码端按照所述编码标识所指示的所述RQT或所述SDC进行解码。
  3. 根据权利要求1或2所述的方法,其特征在于,所述将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,包括:
    根据所述所使用的DMM模式,获得所述所使用的DMM模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
    其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
  4. 根据权利要求3所述的方法,其特征在于,所述获得所述深度图像子块中各像素点的残差信息之前,所述方法还包括:
    将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
    将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数 值做差,得到所述深度图像子块中各像素点的残差信息。
  5. 根据权利要求4所述的方法,其特征在于,所述将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,包括:
    当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
  6. 一种帧内深度图像块解码的方法,其特征在于,包括:
    从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;
    将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码;
    获得解码后的深度图像块。
  7. 根据权利要求6所述的方法,其特征在于,所述从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式时,所述方法还包括:
    从所述码流中获取用于指示编码时使用所述RQT或所述SDC的标识;
    所述将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码,包括:
    根据所述编码时使用所述RQT或所述SDC的标识,将所述在编码时所使用的DMM的模式应用在所述标识所指示的所述RQT或所述SDC中对所述深度图像块进行解码。
  8. 根据权利要求6或7所述的方法,其特征在于,所述将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述深度图像块进行解码,包括:
    根据所述在编码时所使用的DMM的模式,获得所述在编码时所使用的DMM的模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
    其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述 深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
  9. 根据权利要求8所述的方法,其特征在于,所述获得所述深度图像子块中各像素点的残差信息之前,所述方法还包括:
    将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值;
    将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
  10. 根据权利要求9所述的方法,其特征在于,所述将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,包括:
    当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
  11. 一种编码装置,其特征在于,包括:
    获取单元,用于获取待编码的深度图像块;
    检测单元,用于当将深度建模模型DMM应用在递归四叉树RQT或简化深度图编码SDC中对所述获取单元获取的所述深度图像块进行编码时,分别使用所述DMM中的DMM1模式和DMM4模式对所述深度图像块进行检测,得到所述DMM1模式下所述深度图像块的率失真结果,以及所述DMM4模式下所述深度图像块的率失真结果;
    编码单元,用于将所述检测单元检测得到的所述DMM1和所述DMM4中率失真结果最小的DMM模式确定为编码时所使用的DMM模式,将所述所使用的模式应用在所述RQT或所述SDC中对所述深度图像块进行编码,并将所述所使用的DMM模式,写入码流。
  12. 根据权利要求11所述的编码装置,其特征在于,
    所述编码单元,还用于将用于指示所述RQT或所述SDC的编码标识写入所述码流,所述编码标识用于指示解码端按照所述编码标识所指示的所述RQT或所述SDC进行解码。
  13. 根据权利要求11或12所述的编码装置,其特征在于,
    所述编码单元,具体用于根据所述所使用的DMM模式,获得所述所使用的DMM模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
    其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
  14. 根据权利要求13所述的编码装置,其特征在于,
    所述编码单元,具体用于将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
  15. 根据权利要求14所述的编码装置,其特征在于,
    所述编码单元,具体用于当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
  16. 一种解码装置,其特征在于,包括:
    获取单元,用于从码流中获取解码待解码的深度图像块在编码时所使用的DMM的模式;
    解码单元,用于将所述在编码时所使用的DMM的模式应用在递归四叉树RQT或简化深度图编码SDC中对所述获取单元获取的所述深度图像块进行解码;
    获得单元,用于获得所述解码单元解码后的深度图像块。
  17. 根据权利要求16所述的解码装置,其特征在于,
    所述获取单元,还用于从所述码流中获取用于指示编码时使用所述RQT或所述SDC的标识;
    所述解码单元,具体用于根据所述编码时使用所述RQT或所述SDC的标识,将所述在编码时所使用的DMM的模式应用在所述标识所指示的所述RQT或所述SDC中对所述深度图像块进行解码。
  18. 根据权利要求16或17所述的解码装置,其特征在于,
    所述解码单元,具体用于根据所述在编码时所使用的DMM的模式,获得所述在编码时所使用的DMM的模式下分割得到的深度图像子块的块预测值、块偏移值和所述深度图像子块中各像素点的残差信息;
    其中,所述块预测值为按照预先定义的预测规则从与所述深度图像子块相邻的一个深度图像块得到的,所述块偏移值为所述深度图像块的块原始像素值与所述块预测值的差值,再按照预设的映射关系进行映射得到的,所述深度图像子块中各像素点的残差信息为所述深度图像子块中各像素点的原始像素值与所述块预测值和所述块偏移值的和的差值。
  19. 根据权利要求18所述的解码装置,其特征在于,
    所述解码单元,具体用于将所述同一个深度图像子块的块预测值和所述块偏移值相加,得到所述同一个深度图像子块的区域常数值,将所述同一个深度图像子块中各像素点的原始像素值与所述区域常数值做差,得到所述深度图像子块中各像素点的残差信息。
  20. 根据权利要求19所述的解码装置,其特征在于,
    所述解码单元,具体用于当至少一个所述深度图像子块中的块偏移值为0时,将所述块偏移值为0的深度图像子块的预测值作为所述块偏移值为0的深度图像子块的区域常数值。
PCT/CN2015/081078 2014-06-26 2015-06-09 一种帧内深度图像块编码、解码的方法及装置 WO2015196929A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US15/389,196 US10218985B2 (en) 2014-06-26 2016-12-22 Intra-frame depth map block encoding and decoding methods, and apparatus
US16/261,097 US10951901B2 (en) 2014-06-26 2019-01-29 Intra-frame depth map block encoding and decoding methods, and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410300985.X 2014-06-26
CN201410300985.XA CN104079943B (zh) 2014-06-26 2014-06-26 一种帧内深度图像块编码、解码的方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/389,196 Continuation US10218985B2 (en) 2014-06-26 2016-12-22 Intra-frame depth map block encoding and decoding methods, and apparatus

Publications (1)

Publication Number Publication Date
WO2015196929A1 true WO2015196929A1 (zh) 2015-12-30

Family

ID=51600951

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/081078 WO2015196929A1 (zh) 2014-06-26 2015-06-09 一种帧内深度图像块编码、解码的方法及装置

Country Status (3)

Country Link
US (2) US10218985B2 (zh)
CN (2) CN104079943B (zh)
WO (1) WO2015196929A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104079943B (zh) * 2014-06-26 2017-11-28 华为技术有限公司 一种帧内深度图像块编码、解码的方法及装置
EP3664333B1 (en) * 2018-12-06 2023-12-13 Institut Mines Telecom Devices and methods for parallelized recursive block decoding
CN110363212B (zh) * 2019-03-28 2022-02-15 西南石油大学 基于边界扫描的多wedgelet图像近似方法
CN116582688A (zh) * 2023-05-04 2023-08-11 光线云(杭州)科技有限公司 一种适配云绘制系统的深度图压缩方法和装置

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103237214A (zh) * 2013-04-12 2013-08-07 华为技术有限公司 深度图像的编解码方法和编解码装置
CN103237216A (zh) * 2013-04-12 2013-08-07 华为技术有限公司 深度图像的编解码方法和编解码装置
CN103281541A (zh) * 2013-05-10 2013-09-04 北方工业大学 基于深度图像的帧内预测楔形分块方法
WO2014005248A1 (en) * 2012-07-02 2014-01-09 Qualcomm Incorporated Intra-coding of depth maps for 3d video coding
CN103686165A (zh) * 2012-09-05 2014-03-26 乐金电子(中国)研究开发中心有限公司 深度图像帧内编解码方法及视频编解码器
WO2014043828A1 (en) * 2012-09-24 2014-03-27 Qualcomm Incorporated Depth map coding
CN104079943A (zh) * 2014-06-26 2014-10-01 华为技术有限公司 一种帧内深度图像块编码、解码的方法及装置

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110310976A1 (en) * 2010-06-17 2011-12-22 Qualcomm Incorporated Joint Coding of Partition Information in Video Coding
US8526495B2 (en) * 2010-11-22 2013-09-03 Mediatek Singapore Pte. Ltd. Apparatus and method of constrained partition size for high efficiency video coding
US8976861B2 (en) * 2010-12-03 2015-03-10 Qualcomm Incorporated Separately coding the position of a last significant coefficient of a video block in video coding
US20130188013A1 (en) 2011-07-22 2013-07-25 Qualcomm Incorporated Mvc based 3dvc codec supporting inside view motion prediction (ivmp) mode
CN103067715B (zh) 2013-01-10 2016-12-28 华为技术有限公司 深度图像的编解码方法和编解码装置
US9369708B2 (en) * 2013-03-27 2016-06-14 Qualcomm Incorporated Depth coding modes signaling of depth data for 3D-HEVC

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014005248A1 (en) * 2012-07-02 2014-01-09 Qualcomm Incorporated Intra-coding of depth maps for 3d video coding
CN103686165A (zh) * 2012-09-05 2014-03-26 乐金电子(中国)研究开发中心有限公司 深度图像帧内编解码方法及视频编解码器
WO2014043828A1 (en) * 2012-09-24 2014-03-27 Qualcomm Incorporated Depth map coding
CN103237214A (zh) * 2013-04-12 2013-08-07 华为技术有限公司 深度图像的编解码方法和编解码装置
CN103237216A (zh) * 2013-04-12 2013-08-07 华为技术有限公司 深度图像的编解码方法和编解码装置
CN103281541A (zh) * 2013-05-10 2013-09-04 北方工业大学 基于深度图像的帧内预测楔形分块方法
CN104079943A (zh) * 2014-06-26 2014-10-01 华为技术有限公司 一种帧内深度图像块编码、解码的方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Test Model under Consideration for HEVC Based 3D Video Coding", ISO/IEC MPEG2011, 31 January 2011 (2011-01-31) *

Also Published As

Publication number Publication date
CN104079943B (zh) 2017-11-28
CN107659823B (zh) 2020-07-14
US10951901B2 (en) 2021-03-16
US20170105008A1 (en) 2017-04-13
CN107659823A (zh) 2018-02-02
US20190158852A1 (en) 2019-05-23
CN104079943A (zh) 2014-10-01
US10218985B2 (en) 2019-02-26

Similar Documents

Publication Publication Date Title
JP6036964B2 (ja) Hevcビデオ符号化における論理的イントラモードネーミング
US10951901B2 (en) Intra-frame depth map block encoding and decoding methods, and apparatus
JP6067563B2 (ja) ブロックに基づいたミクスドレゾリューションのデータプルーニングを使用した映像符号化
RU2017130321A (ru) Инициализатор предсказателя палитры при кодировании или декодировании самостоятельных кодируемых структур
JP2018509032A5 (zh)
RU2013152159A (ru) Способ декодирования изображений, способ кодирования изображений, устройство декодирования изображений, устройство кодирования изображений и устройство кодирования и декодирования изображений
JP6075468B2 (ja) Hevcにおける係数符号化の調和
KR102216585B1 (ko) 깊이 영상의 부호화 장치 및 복호화 장치, 부호화 방법 및 복호화 방법
RU2013146943A (ru) Способ декодирования изображения, способ кодирования изображения, устройство декодирования изображения, устройство кодирования изображения и устройство кодирования и декодирования изображения
RU2013152627A (ru) Способ декодирования изображения, способ кодирования изображения, устройство декодирования изображения, устройство кодирования изображения и устройство кодирования и декодирования изображения
KR102267770B1 (ko) 픽쳐들의 그룹에서 변경가능 엘리먼트들의 세트를 결정하기 위한 방법 및 디바이스
WO2016197334A1 (zh) 图像编解码方法和设备
JP2018533273A5 (zh)
WO2014166433A1 (zh) 深度图像的编解码方法和编解码装置
JP7067655B2 (ja) 画像コーディング装置、画像デコーディング装置、及び画像処理機器
JP2013162395A (ja) 画像符号化方法、画像符号化装置及び画像符号化プログラム
WO2019128443A1 (zh) 一种图像解码方法及解码器
KR20140007283A (ko) 깊이 영상의 부호화 장치 및 방법, 복호화 장치 및 방법
US11889055B2 (en) Methods and systems for combined lossless and lossy coding
WO2015192340A1 (zh) 基于调色板的图像编码方法、装置以及图像处理设备
EP4136578A1 (en) Methods and systems for combined lossless and lossy coding
CN115209162A (zh) 视频编码划分方法、视频处理方法及计算机设备

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: 15811051

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15811051

Country of ref document: EP

Kind code of ref document: A1