WO2015101251A1 - 非对称运动分割方式编码的方法和编码设备 - Google Patents

非对称运动分割方式编码的方法和编码设备 Download PDF

Info

Publication number
WO2015101251A1
WO2015101251A1 PCT/CN2014/095287 CN2014095287W WO2015101251A1 WO 2015101251 A1 WO2015101251 A1 WO 2015101251A1 CN 2014095287 W CN2014095287 W CN 2014095287W WO 2015101251 A1 WO2015101251 A1 WO 2015101251A1
Authority
WO
WIPO (PCT)
Prior art keywords
mode
asymmetric
encoding
coding
partitioning
Prior art date
Application number
PCT/CN2014/095287
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 WO2015101251A1 publication Critical patent/WO2015101251A1/zh

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
    • H04N19/50Methods 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
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/119Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
    • 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
    • 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
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction

Definitions

  • the present invention relates to the field of image processing, and in particular, to a method and an encoding device for encoding asymmetric motion segmentation methods in the field of image processing.
  • the AMP fast implementation algorithm mainly includes: determining which asymmetric motion segmentation modes use the normal Normal mode for coding detection or determining which asymmetric motion segmentation modes adopt the merged Merge mode for coding detection.
  • the AMP fast implementation algorithm is widely used in the encoding process of various viewpoints of texture maps and depth maps of 3D video.
  • the importance of the AMP fast implementation algorithm in each view's texture map and depth map is not consistent, which affects the coding efficiency of 3D video. .
  • the embodiment of the invention provides a method and an encoding device for encoding an asymmetric motion segmentation mode, and determines a coding detection mode adopted by a base view of a texture map, a depth map, and a coding unit CU of a non-base view, respectively, which can effectively improve coding. effectiveness.
  • an embodiment of the present invention provides a method for encoding an asymmetric motion segmentation method, the method comprising: determining a coding unit CU; a CU that is a non-base view of a texture map, and according to a size of a CU, a current CU
  • the optimal prediction unit PU partitioning mode, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the CU of the upper CU determine that the first asymmetric PU partitioning mode of the CU does not conform to the first encoding mode.
  • the first reference to the CU is determined according to the inter-view reference view information of the view to which the CU belongs A coded detection method for asymmetric PU partitioning.
  • determining an encoding detection manner of the first asymmetric PU split mode of the first CU according to the inter-view reference view information of the view to which the first CU belongs including: When the first asymmetric PU partitioning mode of the CU is in accordance with the third encoding mode determination condition, the first asymmetric PU partitioning mode of the CU uses the normal Normal mode for encoding detection, and the Normal mode refers to the inter-view reference point information of the CU.
  • the method adopts the merged Merge method for coding detection.
  • the inter-view reference view information includes: a CU of the inter-view reference view of the view to which the CU belongs A symmetric PU partitioning method of the CU of the coding mode and the inter-view reference view.
  • the method further includes:
  • the CU is a non-base view CU of the texture map and according to the size of the CU, the current optimal PU partition mode of the CU, the current optimal PU partitioning mode of the CU, and the optimal PU partition mode of the CU of the upper CU,
  • the first asymmetric PU segmentation mode of the CU is coded by using the Normal mode.
  • the method further includes:
  • the CU is the CU of the base view of the texture map and is determined according to the size of the CU, the current optimal PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the first asymmetric PU partitioning mode of the CU does not meet the first encoding mode determination condition and meets the second encoding mode determination condition, the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the method further includes:
  • the CU is the CU of the base view of the texture map and is determined according to the size of the CU, the current optimal PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the first asymmetric PU partitioning mode of the CU conforms to the first coding mode determination condition, the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the method further includes:
  • the CU is the CU of the view of the depth map and determines the CU according to the size of the CU, the current optimal PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the first asymmetric PU segmentation mode does not meet the first coding mode determination condition and meets the second coding mode determination condition, the first asymmetric PU segmentation mode of the CU is detected by using the Merge method.
  • the method further includes:
  • the CU is the CU of the view of the depth map and determines the CU according to the size of the CU, the current optimal PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the first asymmetric PU partitioning mode conforms to the first coding mode determination condition, the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • an embodiment of the present invention provides a method for encoding an asymmetric motion partitioning method, the method comprising: determining a coding unit CU; the coding unit CU is a CU of a view of the texture map and according to a size of the CU, a current CU.
  • the optimal prediction unit PU partitioning mode, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the CU of the upper CU determine that the first asymmetric PU partitioning mode of the CU does not conform to the first encoding mode.
  • the condition is the second encoding mode, or the first asymmetric PU partitioning mode of the CU is consistent with the first encoding mode
  • the first asymmetric PU partitioning mode of the CU is detected by using the normal Normal mode. Refers to the complete motion prediction compensation method.
  • the method further includes: at a CU where the CU is a view of the depth map, and according to a size of the CU, a current optimal PU segmentation mode of the CU, and a current optimal PU segmentation of the CU.
  • the coding mode of the mode and the optimal PU segmentation mode of the CU of the CU determine that the first asymmetric PU segmentation mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode determination condition, and is performed on the CU.
  • the first asymmetric PU partitioning method adopts the merged Merge method for encoding detection.
  • the method further includes: at a CU where the CU is a view of the depth map and according to a size of the CU Determining the first asymmetric CU, the current optimal PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU, determining the first asymmetric of the CU When the PU split mode conforms to the first coding mode determination condition, the first asymmetric PU partitioning mode of the CU is coded using the Normal mode.
  • an embodiment of the present invention provides an encoding device, where the encoding device includes: a first determining module, configured to determine a coding unit CU; and a second determining module, configured to use a CU that is a non-base view of the texture map. Determining the first asymmetric PU of the CU according to the size of the CU, the current optimal prediction unit PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the coding detection mode of the first asymmetric PU segmentation mode of the CU is determined according to the inter-view reference view information of the view to which the CU belongs.
  • the second determining module is specifically configured to: at a CU where the CU is a non-base view of the texture map, and according to a size of the CU, a current optimal prediction unit PU split mode of the CU, Determining the encoding mode of the current optimal PU partitioning mode of the CU and the optimal PU partitioning mode of the upper CU of the CU, determining that the first asymmetric PU partitioning mode of the CU does not meet the first encoding mode determination condition, and conforms to the second encoding mode Determining a condition, and determining, according to the inter-view reference view information of the view to which the CU belongs, that the first asymmetric PU segmentation mode of the CU conforms to the third coding mode determination condition, the first asymmetric PU segmentation mode of the CU is encoded by using the normal Normal mode.
  • the Normal mode refers to the complete motion prediction compensation mode; or the CU is the CU of the non-base view of the texture map and according to the size of the CU, the current optimal prediction unit PU partition mode of the CU, and the current optimal PU partition mode of the CU.
  • the coding mode and the optimal PU segmentation mode of the upper CU of the CU determine that the first asymmetric PU segmentation mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode.
  • Determining a condition and determining, according to the inter-view reference view information of the view to which the CU belongs, that the first asymmetric PU split mode of the CU does not meet the third coding mode determination condition, and adopting the merged Merge mode for the first asymmetric PU split mode of the CU Perform code detection.
  • the CU coding mode and the inter-view reference view of the inter-view reference view of the view to which the CU belongs Symmetric PU partitioning of the CU.
  • the second determining module is further The CU for the non-base view CU of the texture map and according to the size of the CU, the current optimal PU partition mode of the CU, the current optimal PU partition mode of the CU, and the optimal PU of the upper CU of the CU.
  • the segmentation mode determines that the first asymmetric PU segmentation mode of the CU conforms to the first coding
  • the first asymmetric PU partitioning mode of the CU is coded using the Normal mode.
  • the second determining module is further For the CU where the CU is the base view of the texture map and according to the size of the CU, the current optimal PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the optimal PU partition of the upper CU of the CU.
  • the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the second determining module is further For the CU where the CU is the base view of the texture map and according to the size of the CU, the current optimal PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the optimal PU partition of the upper CU of the CU In the manner that the first asymmetric PU partitioning mode of the CU is consistent with the first encoding mode determination condition, the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the second determining module is further The CU for the CU is the view of the depth map and according to the size of the CU, the current optimal PU partitioning mode of the CU, the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU. And determining that the first asymmetric PU segmentation mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode determination condition, and adopts the Merge mode for coding detection on the first asymmetric PU segmentation mode of the CU.
  • the second determining module is further The CU for the CU is the view of the depth map and according to the size of the CU, the current optimal PU partitioning mode of the CU, the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the first asymmetric PU partitioning mode of the CU conforms to the first coding mode determination condition
  • the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • an embodiment of the present invention provides an encoding device, where the encoding device includes: a third determining module, configured to determine a coding unit CU; and a fourth determining module, configured to be in the coding unit CU
  • the CU of the view of the texture map and determines the CU according to the size of the CU, the current optimal prediction unit PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the optimal PU partition mode of the upper CU of the CU.
  • the first asymmetric PU partitioning manner does not meet the first encoding mode determination condition and conforms to the second encoding mode determining condition, or determines that the first asymmetric PU partitioning mode of the CU conforms to the first encoding mode determining condition, and is first to the CU.
  • the asymmetric PU partitioning mode uses the normal Normal mode for encoding detection, and the Normal mode refers to the complete motion prediction compensation mode.
  • the fourth determining module is further configured to: at the CU where the CU is the view of the depth map, and according to the size of the CU, the current optimal PU segmentation mode of the CU, and the current maximum of the CU.
  • the coding mode of the PU partition mode and the optimal PU partition mode of the CU of the upper CU determine that the first asymmetric PU partition mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode determination condition.
  • the first asymmetric PU partitioning mode of the CU adopts the merged Merge method for encoding detection.
  • the fourth determining module is further configured to be used by the CU that is the CU of the depth map
  • the size of the CU, the current optimal PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU determine that the first asymmetric PU partitioning mode of the CU conforms to the first
  • the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the method and the encoding device for encoding the asymmetric motion segmentation mode in the embodiment of the present invention respectively determine the coding detection manner adopted by the coding unit CU of the base view and the non-base view of the texture map and the depth map, and
  • the reference inter-view reference view information is used to determine the coding detection mode of the CU of the non-base view, which can effectively improve the coding efficiency.
  • FIG. 1 is a schematic diagram of an asymmetric motion segmentation mode AMP during encoding.
  • FIG. 2 is a schematic flowchart of a method for asymmetric motion segmentation mode coding according to an embodiment of the present invention.
  • FIG. 3 is a schematic flowchart of a method for asymmetric motion segmentation mode coding according to another embodiment of the present invention.
  • FIG. 4 is a schematic flowchart of a method for asymmetric motion segmentation mode coding according to another embodiment of the present invention.
  • FIG. 5 is a schematic flowchart of a method for asymmetric motion segmentation mode coding according to another embodiment of the present invention.
  • Figure 6 is a schematic block diagram of an encoding device in accordance with one embodiment of the present invention.
  • Figure 7 is a schematic block diagram of an encoding device in accordance with another embodiment of the present invention.
  • Figure 1 shows a schematic diagram of an asymmetric motion segmentation mode AMP during encoding.
  • the video codec structure includes three units: a coding unit (CU), a prediction unit (PU), and a transform unit (TU).
  • the PU includes content information (such as a reference index, a motion vector, and the like) of the prediction block and the prediction block used for the prediction process, and generally has four symmetric PU division modes: 2N ⁇ 2N, 2N ⁇ N, N ⁇ 2N, and N ⁇ N.
  • the asymmetric motion segmentation mode AMP is newly added. That is, as shown in FIG. 1, in the interframe coding process, the CU has four asymmetric PU partitioning modes: 2N ⁇ nU, 2N ⁇ nD, nR ⁇ 2N, and nL ⁇ 2N.
  • the depth map and the texture map need to be separately encoded.
  • one viewpoint is usually used as a base view and other viewpoints are regarded as non-base views.
  • the base view can be encoded differently from other viewpoints, such as the High Efficiency Video Coding (HEVC) method.
  • HEVC High Efficiency Video Coding
  • reference information of the base view can be referred to, and in addition to the traditional HEVC method, more coding tools, such as disparity compensation prediction technology, disparity vector acquisition technology, and inter-view motion Prediction, inter-view residual prediction, luminance compensation, view synthesis prediction, etc. can be applied to the encoding process of non-base view.
  • code detection by using the Normal mode in the embodiment of the present invention refers to performing code detection by using a complete motion prediction compensation mode.
  • FIG. 2 shows a schematic flow diagram of a method 100 of asymmetric motion segmentation mode encoding, which may be performed by an encoding device, in accordance with an embodiment of the present invention.
  • the CU is a CU of a non-base view of the texture map, and according to the size of the CU, the current optimal prediction unit PU partition mode of the CU, the current optimal PU partition mode of the CU, and the upper CU.
  • the optimal PU partitioning mode of the first-level CU determining that the first asymmetric PU partitioning mode of the CU does not meet the first encoding mode determination condition, and conforms to the second encoding mode determination condition, according to the inter-view reference of the viewpoint to which the CU belongs
  • the view information determines a code detection mode of the first asymmetric PU partition mode of the CU.
  • the method for encoding the asymmetric motion segmentation mode determines the coding detection mode adopted by the coding unit CU of the base view and the non-base view of the texture map and the depth map, respectively, and refers to the inter-view reference view.
  • the information determines the coding detection mode of the CU of the non-base view, which can effectively improve the coding efficiency.
  • the basic viewpoint may also be referred to as an independent viewpoint, which is a viewpoint that is first encoded.
  • the encoding device when encoding the CU, the encoding device first determines whether the encoding is in the inter encoding process. After determining that the encoding is in the inter-frame encoding process, the encoding device determines whether it is currently in the texture map encoding process or the depth map encoding process. In the embodiment of the present invention, optionally, the encoding device determines an encoding process currently in the texture map, and determines that the viewpoint to which the CU belongs is a non-base viewpoint.
  • determining, according to the size of the CU, the current optimal prediction unit PU partitioning manner of the CU, and the current optimal PU partitioning mode of the CU determining that the first asymmetric PU partitioning manner of the CU does not conform to the first encoding Way to judge the condition.
  • the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU partitioning mode of the CU is 2N ⁇ N, and the asymmetric PU partitioning mode 2N ⁇ nU of the CU may be determined.
  • 2N ⁇ nD is a first coding mode determination condition; or, the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU division mode of the CU is 2N ⁇ 2N, and If the encoding mode of the current optimal prediction unit PU partitioning mode of the CU is not the merge or skip mode, the asymmetric PU partitioning modes 2N ⁇ nU and 2N ⁇ nD of the CU may be determined to be in accordance with the first encoding mode determination condition.
  • the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU segmentation mode of the CU is N ⁇ 2N, and the asymmetric PU segmentation mode of the CU may be determined to be nL ⁇ 2N and nR ⁇ 2N are in accordance with the first coding mode judgment condition; or, the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU division mode of the CU is 2N ⁇ 2N, and the current optimality of the CU is If the coding mode of the prediction unit PU partitioning mode is not the merge or skip mode, the asymmetric PU partitioning manners nL ⁇ 2N and nR ⁇ 2N of the CU may be determined to be in accordance with the first coding mode determination condition. On the contrary, it is a condition that does not conform to the first coding mode.
  • the method conforms to the second coding mode judgment condition.
  • the current optimal PU partitioning mode of the CU is 2N ⁇ 2N, and the encoding mode of the current optimal PU partitioning mode of the CU is not the skip encoding mode; or the upper CU of the CU
  • the optimal PU partitioning mode is the AMP mode; or the current optimal PU partitioning mode of the CU is 2N ⁇ N, and the optimal PU partitioning mode of the upper CU of the CU is a non-PU partitioning mode, then the The asymmetric PU partitioning modes 2N ⁇ nU and 2N ⁇ nD of the CU are in accordance with the second encoding mode determination condition.
  • the current optimal PU partitioning mode of the CU is 2N ⁇ 2N, and the encoding mode of the current optimal PU partitioning mode of the CU is not the skip encoding mode; or the previous one of the CU
  • the optimal PU partitioning mode of the CU is AMP mode; or the current optimal PU partitioning mode of the CU is N ⁇ 2N, and the optimal PU partitioning mode of the upper CU of the CU is non-PU partitioning mode, It is determined that the asymmetric PU partitioning manners nL ⁇ 2N and nR ⁇ 2N of the CU are in accordance with the second encoding mode determination condition. On the contrary, it is a condition that does not conform to the second coding mode.
  • the first asymmetric PU segmentation mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode determination condition.
  • other asymmetric PU partitioning manners of the CU may conform to the first encoding mode determination condition; or may not meet the first encoding mode determination condition, and conform to the second encoding mode determination condition;
  • the first coding mode judgment condition is not met, and the second coding mode judgment condition is not met.
  • the code detection manner that is finally adopted may be different, and the embodiment of the present invention is different. This is not limited.
  • the inter-view reference view of the view to which the CU belongs may be a base view or a non-base view.
  • the inter-view reference view information may include an encoding mode of the CU of the inter-view reference view and a symmetric PU split mode of the CU.
  • the third encoding mode determination condition is a determination condition based on the above-described inter-view reference viewpoint information.
  • the optimal PU partitioning mode of the CU of the inter-view reference view is 2N ⁇ N; or the optimal PU partitioning mode of the CU of the inter-view reference view is 2N ⁇ 2N and the CU of the inter-view reference view If the encoding mode is not the merge or skip mode, it can be determined that the asymmetric PU partitioning modes 2N ⁇ nU and 2N ⁇ nD of the CU are in accordance with the third encoding mode determination condition.
  • the optimal PU partitioning mode of the CU of the inter-view reference view is N ⁇ 2N; or the optimal PU partitioning mode of the CU of the inter-view reference view is 2N ⁇ 2N and the inter-view reference view If the encoding mode of the CU is not the merge or skip mode, it may be determined that the asymmetric PU partitioning manners nL ⁇ 2N and nR ⁇ 2N of the CU are in accordance with the third encoding mode determination condition. On the contrary, it is a condition that does not meet the third coding mode.
  • the first asymmetric PU partitioning mode of the CU meets the third encoding mode determining condition, the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the first asymmetric PU partitioning manner of the CU when the first asymmetric PU partitioning manner of the CU does not meet the third encoding mode determining condition, the first asymmetric PU partitioning manner of the CU is performed by using a Merge manner for encoding detection.
  • the method 100 will be described below in conjunction with a specific example.
  • a 16 ⁇ 16 size CU of a non-base view of the texture map is encoded.
  • the 2N ⁇ nU and 2N ⁇ nD of the CU do not meet the first determination condition and conform to the second determination condition.
  • the method 100 can also include:
  • the CU is a non-base view CU of the texture map and according to the size of the CU, the current optimal PU partition mode of the CU, the current optimal PU partition mode of the CU, and the upper CU of the CU.
  • the optimal PU partitioning mode determines that the first asymmetric PU partitioning mode of the CU conforms to the first encoding mode determination condition, and performs the encoding detection by using the Normal mode for the first asymmetric PU partitioning mode of the CU.
  • the first asymmetric PU split mode of the CU is adopted.
  • the Normal mode performs code detection. For example, a PN of a 16 ⁇ 16 size of a non-base view of the texture map is encoded, and nL ⁇ 2N and nR ⁇ 2N of the CU comply with the first determination condition, and then the normal LL is used for nL ⁇ 2N and nR ⁇ 2N. The way to perform coding detection.
  • the texture map further includes a CU of the base view, and one of the asymmetric PU split modes may not meet the first coding mode determination condition, and conforms to the second coding mode determination condition. Accordingly, the method 100 can also include:
  • the CU is the CU of the base view of the texture map and is based on the size of the CU, the current optimal PU partition mode of the CU, the current optimal PU partitioning mode of the CU, and the highest CU of the CU. If the first asymmetric coding mode is not consistent with the first coding mode determination condition and the second coding mode determination condition is met, the first asymmetric PU segmentation mode of the CU is Normal. The way to perform coding detection.
  • the CU is The first asymmetric PU segmentation mode uses the Normal mode for encoding detection. For example, encoding a 32 ⁇ 32 size CU of a base view of the texture map, The 2N ⁇ nU and 2N ⁇ nD of the CU do not meet the first judgment condition, and the second judgment condition is met, and the 2N ⁇ nU and 2N ⁇ nD of the CU are coded and detected in the Normal mode.
  • the method 100 can also include:
  • the CU is the CU of the base view of the texture map and is based on the size of the CU, the current optimal PU partition mode of the CU, the current optimal PU partitioning mode of the CU, and the highest CU of the CU.
  • the PU partitioning mode when the first asymmetric PU partitioning mode of the CU is determined to be in accordance with the first encoding mode, the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the first asymmetric PU partitioning mode of the CU adopts the Normal mode.
  • Perform code detection For example, a 32 ⁇ 32 size CU of a base view of the texture map is encoded, and nL ⁇ 2N and nR ⁇ 2N of the CU meet the first judgment condition, and the normal manner is adopted for nL ⁇ 2N and nR ⁇ 2N of the CU. Perform code detection.
  • the symmetric PU segmentation method performs coding detection.
  • the method 100 can also include:
  • the CU is the CU of the view of the depth map, and according to the size of the CU, the current optimal PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the optimal CU of the CU.
  • the PU split mode when the first asymmetric PU split mode of the CU does not meet the first coding mode determination condition, and the second coding mode determination condition is met, the first asymmetric PU split mode of the CU adopts the Merge mode. Perform code detection.
  • the CU that determines the CU is the view point of the depth map, and determines that the first asymmetric PU partitioning mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode determination condition, where the CU is
  • An asymmetric PU segmentation method uses the Merge method for encoding detection. For example, encoding a 32 ⁇ 32 size CU of one view of the depth map, 2N ⁇ nU of the CU If 2N ⁇ nD does not meet the first judgment condition and meets the second judgment condition, the 2N ⁇ nU and 2N ⁇ nD of the CU are coded and detected by using the Merge method.
  • the view of the depth map to which the CU of the embodiment of the present invention belongs may be a base view or a non-base view, which is not limited by the embodiment of the present invention.
  • the method 100 can also include:
  • the CU is the CU of the view of the depth map, and according to the size of the CU, the current optimal PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the optimal CU of the CU.
  • the PU segmentation mode when the first asymmetric PU segmentation mode of the CU is determined to be in accordance with the first coding mode determination condition, the first asymmetric PU segmentation mode of the CU is coded by using the Normal mode.
  • the first asymmetric PU segmentation mode of the CU adopts the Normal mode.
  • Perform code detection For example, a 32 ⁇ 32 size CU of one view of the depth map is encoded, and nL ⁇ 2N and nR ⁇ 2N of the CU meet the first determination condition, and then nL ⁇ 2N and nR ⁇ 2N of the CU are performed in the Normal mode. Code detection.
  • the asymmetric PU partitioning mode of the CU of one view of the depth map it may not meet the first encoding mode determining condition or the second encoding mode determining condition, and at this time, the asymmetric of the CU is not
  • the PU split method performs code detection.
  • viewpoint of the depth map of the embodiment of the present invention may be a base view or a non-base view, which is not limited by the embodiment of the present invention.
  • FIG. 3 shows a schematic flow chart of a method 200 of asymmetric motion segmentation mode coding according to another embodiment of the present invention. The method is performed by an encoding device, and the method 200 illustrated in FIG. 3 includes:
  • the CU currently being coded is the CU of the view of the depth map
  • S205 Perform an encoding detection on the asymmetric PU partitioning method of the CU that does not conform to the first encoding mode determination condition and conform to the viewpoint of the depth map of the second encoding mode determination condition, and complete the encoding by using the Merge method.
  • S208 Perform encoding detection on the asymmetric PU partitioning mode of the CU of the base view of the texture map conforming to the first encoding mode determination condition, and complete the encoding after completion.
  • determining whether the asymmetric PU segmentation mode of the CU of the base view of the texture map conforms to the second coding mode determination condition If yes, execute S210; if no, end.
  • S210 Perform an encoding detection on the asymmetric PU partitioning method of the CU that does not conform to the first encoding mode determination condition and conform to the basic view of the texture map of the second encoding mode determination condition, and complete the encoding after completion.
  • the asymmetric PU partition mode of the CU conforms to the first coding mode determination condition. If yes, execute S212; if no, execute S213.
  • the asymmetric PU partitioning method of the CU of the non-base view of the texture map that meets the first coding mode determination condition is coded by using the Normal method, and is completed after completion.
  • the asymmetric PU partitioning manner of the CU of the non-base view of the texture map does not conform to the first coding mode determination condition, and the second PU coding mode is determined, and whether the asymmetric PU partition mode of the CU is consistent with the first
  • the three coding method judges the condition. If yes, execute S215; if no, execute S216.
  • S215 Performing code detection by using the Normal mode for the asymmetric PU segmentation mode of the CU that does not conform to the first coding mode determination condition and conforms to the second coding mode determination condition, and conforms to the non-base view of the texture map of the third coding mode determination condition, Finished when finished.
  • S216 Performing code detection by using the Merge method for the asymmetric PU segmentation mode of the CU that does not conform to the first coding mode determination condition and conforms to the second coding mode determination condition, and does not conform to the third coding mode determination condition of the non-base view of the texture map. , after the completion is over.
  • the method for encoding the asymmetric motion segmentation mode determines the coding detection manner adopted by the coding unit CU of the base view and the non-base view of the texture map and the depth map, respectively, and refers to the inter-view reference view information. Determining the coding detection mode of the CU of the non-base view can effectively improve the coding efficiency.
  • FIG. 4 shows a schematic flowchart of a method 300 for asymmetric motion segmentation mode coding, which may be performed by an encoding device, in accordance with an embodiment of the present invention.
  • the coding unit CU is the CU of the view of the texture map, and according to the size of the CU, the current optimal prediction unit PU partition mode of the CU, the current optimal PU partition mode of the CU, and the previous one of the CU.
  • the optimal PU partitioning mode of the CU determines that the first asymmetric PU partitioning mode of the CU does not meet the first encoding mode determination condition and conforms to the second encoding mode determination condition, or determines that the first asymmetric PU segmentation mode of the CU is consistent.
  • the first coding mode determines the condition
  • the first asymmetric PU segmentation mode of the CU is coded by using the normal Normal mode, and the Normal mode refers to the complete motion prediction compensation mode.
  • the method for encoding the asymmetric motion segmentation mode in the embodiment of the present invention can improve the coding efficiency by performing the code detection on the Asymmetric PU segmentation mode of the view by preferentially performing the Normal mode.
  • code detection by using the Normal mode in the embodiment of the present invention refers to performing code detection by using a complete motion prediction compensation mode.
  • the encoding device first determines whether the encoding is in an inter encoding process. After determining that the encoding is in the inter-frame encoding process, the encoding device determines whether it is currently in the texture map encoding process or the depth map encoding process. In the embodiment of the present invention, optionally, the encoding device determines that the CU is a CU of a viewpoint of the texture map.
  • the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU partitioning mode of the CU is 2N ⁇ N, and the asymmetric PU partitioning mode 2N ⁇ nU of the CU may be determined.
  • 2N ⁇ nD is a first coding mode determination condition; or, the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU division mode of the CU is 2N ⁇ 2N, and the current optimal prediction unit of the CU If the encoding mode of the PU partitioning mode is not the merge or skip mode, the asymmetric PU partitioning mode of the CU may be determined to be 2N ⁇ nU and 2N ⁇ nD. Code mode judgment condition.
  • the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU segmentation mode of the CU is N ⁇ 2N, and the asymmetric PU segmentation mode of the CU may be determined to be nL ⁇ 2N and nR ⁇ 2N are in accordance with the first coding mode judgment condition; or, the size of the CU is not 64 ⁇ 64, and the current optimal prediction unit PU division mode of the CU is 2N ⁇ 2N, and the current optimality of the CU is If the coding mode of the prediction unit PU partitioning mode is not the merge or skip mode, the asymmetric PU partitioning manners nL ⁇ 2N and nR ⁇ 2N of the CU may be determined to be in accordance with the first coding mode determination condition. On the contrary, it is a condition that does not conform to the first coding mode.
  • the first asymmetric of the CU is determined according to the current optimal prediction unit PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the PU split mode conforms to the second coding mode determination condition.
  • the current optimal PU partitioning mode of the CU is 2N ⁇ 2N, and the encoding mode of the current optimal PU partitioning mode of the CU is not the skip encoding mode; or the upper CU of the CU
  • the optimal PU partitioning mode is the AMP mode; or the current optimal PU partitioning mode of the CU is 2N ⁇ N, and the optimal PU partitioning mode of the upper CU of the CU is a non-PU partitioning mode, then the The asymmetric PU partitioning modes 2N ⁇ nU and 2N ⁇ nD of the CU are in accordance with the second encoding mode determination condition.
  • the current optimal PU partitioning mode of the CU is 2N ⁇ 2N, and the encoding mode of the current optimal PU partitioning mode of the CU is not the skip encoding mode; or the previous one of the CU
  • the optimal PU partitioning mode of the CU is AMP mode; or the current optimal PU partitioning mode of the CU is N ⁇ 2N, and the optimal PU partitioning mode of the upper CU of the CU is non-PU partitioning mode, It is determined that the asymmetric PU partitioning manners nL ⁇ 2N and nR ⁇ 2N of the CU are in accordance with the second encoding mode determination condition. On the contrary, it is a condition that does not conform to the second coding mode.
  • the first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the first asymmetric PU partitioning manner of the CU may not meet the first encoding mode determination condition or the second encoding mode determination condition. At this time, the asymmetric PU partitioning mode of the CU is not encoded and detected.
  • the viewpoint to which the CU belongs may be the basic of the texture map.
  • the viewpoint can also be a non-base view of the texture map.
  • the CU may also be a CU of a view of the depth map, and one of the asymmetric PU partitioning modes may not meet the first encoding mode determining condition, and when the second encoding mode determining condition is met, correspondingly
  • the method 300 can also include:
  • the CU is the CU of the view of the depth map, and according to the size of the CU, the current optimal PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the optimal CU of the CU.
  • the PU splitting mode when the first asymmetric PU partitioning mode of the CU does not meet the first encoding mode determination condition, and the second encoding mode determining condition is met, the first asymmetric PU partitioning mode of the CU is merged with the Merge. The way to perform coding detection.
  • the CU that determines the CU is the view point of the depth map, and determines that the first asymmetric PU partitioning mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode determination condition, where the CU is An asymmetric PU segmentation method uses the Merge method for encoding detection. For example, a 32 ⁇ 32 size CU of one view of the depth map is encoded, and 2N ⁇ nU and 2N ⁇ nD of the CU do not meet the first judgment condition, and the second judgment condition is met, and the 2N ⁇ nU of the CU is 2N ⁇ nD uses Merge to perform coding detection.
  • the view of the depth map to which the CU of the embodiment of the present invention belongs may be a base view or a non-base view, which is not limited by the embodiment of the present invention.
  • the method 300 can also include:
  • the CU is the CU of the view of the depth map, and according to the size of the CU, the current optimal PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the optimal CU of the CU.
  • the PU segmentation mode when the first asymmetric PU segmentation mode of the CU is determined to be in accordance with the first coding mode determination condition, the first asymmetric PU segmentation mode of the CU is coded by using the Normal mode.
  • the first asymmetric PU segmentation mode of the CU adopts the Normal mode.
  • Perform code detection For example, encoding a 32 ⁇ 32 size CU of one view of the depth map, where nL ⁇ 2N and nR ⁇ 2N of the CU meet the first judgment condition, Then, the nL ⁇ 2N and nR ⁇ 2N of the CU are coded and detected in the Normal mode.
  • the asymmetric PU partitioning mode of the CU of one view of the depth map it may not meet the first encoding mode determining condition or the second encoding mode determining condition, and at this time, the asymmetric of the CU is not
  • the PU split method performs code detection.
  • the view point of the depth map to which the CU of the embodiment of the present invention belongs may be a base view point or a non-base view point, which is not limited by the embodiment of the present invention.
  • the embodiments corresponding to FIG. 4 are performed in parallel, and the order of execution thereof is not limited.
  • FIG. 5 shows a schematic flow diagram of a method 400 of asymmetric motion segmentation mode encoding in accordance with another embodiment of the present invention.
  • the method is performed by an encoding device, and the method 400 illustrated in FIG. 5 includes:
  • S401 Determine a CU of a view point of whether the currently coded CU is a depth map.
  • S402. Determine whether the asymmetric PU split mode of the CU of the view of the depth map meets the first coding mode determination condition.
  • S403 Perform an encoding detection on the asymmetric PU partitioning method of the CU of the view of the depth map that satisfies the first encoding mode determination condition, and complete the encoding after completion.
  • S405 Perform coding detection by using the Merge method for the asymmetric PU partitioning method of the CU that does not conform to the first coding mode determination condition and conform to the view of the depth map of the second coding mode determination condition, and complete the completion.
  • the asymmetric PU partition of the CU that determines the viewpoint of the texture map currently being encoded Whether the formula meets the first coding mode judgment condition. If yes, execute S407; if no, execute S408.
  • S409 Perform an encoding detection on the asymmetric PU partitioning method of the CU that does not conform to the first encoding mode determination condition and conform to the viewpoint of the texture map of the second encoding mode determination condition, and complete the encoding after completion.
  • the method for encoding the asymmetric motion segmentation mode in the embodiment of the present invention can improve the coding efficiency by performing the code detection on the Asymmetric PU segmentation mode of the view by preferentially performing the Normal mode.
  • the viewpoint may be a base view or a non-base view.
  • Fig. 6 shows a schematic block diagram of an encoding device of one embodiment of the present invention.
  • the encoding device 500 includes:
  • a first determining module 510 configured to determine a coding unit CU
  • a second determining module 520 configured to: at a CU where the CU is a non-base view of the texture map, and according to a size of the CU, a current optimal prediction unit PU partition mode of the CU, and a current optimal PU partition mode of the CU
  • the mode and the optimal PU segmentation mode of the CU of the CU determine that the first asymmetric PU segmentation mode of the CU does not meet the first coding mode determination condition, and the second coding mode determination condition is met, according to the CU belongs to
  • the inter-view reference view information of the view determines the coding detection mode of the first asymmetric PU partition mode of the CU.
  • the encoding apparatus of the embodiment of the present invention determines the CU of the non-base view by introducing the code detection mode adopted by the coding unit CU of the base view and the non-base view of the texture map and the depth map, respectively, and introducing the inter-view reference view information.
  • the coding detection method can effectively improve the coding efficiency.
  • the second determining module 520 is specifically configured to: use the CU as a non-base view of the texture map.
  • the CU of the point is determined according to the size of the CU, the current optimal prediction unit PU partitioning mode of the CU, the encoding mode of the current optimal PU partitioning mode of the CU, and the optimal PU partitioning mode of the upper CU of the CU.
  • the first asymmetric PU segmentation mode of the CU does not meet the first coding mode determination condition, and conforms to the second coding mode determination condition, and determines the first asymmetric of the CU according to the inter-view reference view information of the view to which the CU belongs.
  • the first asymmetric PU partitioning mode of the CU adopts the normal Normal mode for encoding detection, and the Normal mode refers to a complete motion prediction compensation mode; or
  • the CU is a non-base view CU of the texture map and according to the size of the CU, the current optimal prediction unit PU partition mode of the CU, the coding mode of the current optimal PU partition mode of the CU, and the upper level of the CU
  • the optimal PU partitioning mode of the CU determining that the first asymmetric PU partitioning manner of the CU does not comply with the first encoding mode determination condition, and conforms to the second encoding mode determination condition, and according to the inter-view reference viewpoint information of the viewpoint to which the CU belongs
  • the first asymmetric PU partitioning mode of the CU is performed by using a merged Merge mode for encoding detection.
  • the CU of the view to which the CU belongs depends on the coding mode of the CU of the view and the symmetric PU partition of the CU of the inter-view reference view.
  • the second determining module 520 is further configured to: in the CU where the CU is a non-base view of the texture map, and according to the size of the CU, the current optimal PU partition mode of the CU, and the current optimal PU segmentation of the CU.
  • the code detection is performed by the Normal method.
  • the second determining module 520 is further configured to use, in the CU that the CU is the base view of the texture map, according to the size of the CU, the current optimal PU partitioning manner of the CU, and the current optimal PU partitioning manner of the CU. And the optimal PU partitioning mode of the CU of the CU, determining that the first asymmetric PU partitioning mode of the CU does not meet the first encoding mode determination condition, and conforms to the second encoding mode determining condition, The first asymmetric PU partitioning mode of the CU is coded by using the Normal mode.
  • the second determining module 520 is configured to: at the CU of the base view of the texture map, according to the size of the CU, the current optimal PU partition mode of the CU, and the current optimal PU partition mode of the CU. And determining, by the CU, the optimal PU partitioning mode of the CU, the first asymmetric PU partitioning manner of the CU is in accordance with the first encoding mode, and the first asymmetric PU partitioning mode of the CU adopts the Normal mode. Perform code detection.
  • the second determining module 520 is a CU that is a CU of the depth map according to the size of the CU, a current optimal PU partitioning manner of the CU, and a current optimal PU partitioning manner of the CU.
  • the first asymmetric PU partitioning method uses the Merge method for encoding detection.
  • the second determining module 520 is further configured to: in the CU where the CU is the view of the depth map, and according to the size of the CU, the current optimal PU split mode of the CU, and the current optimal PU split mode of the CU.
  • the Normal mode performs code detection.
  • the encoding device 500 can implement the various processes implemented by the encoding device in the embodiments of FIG. 2 and FIG. 3. To avoid repetition, details are not described herein again.
  • Fig. 7 shows a schematic block diagram of an encoding apparatus of another embodiment of the present invention.
  • the encoding device 600 includes:
  • a third determining module 610 configured to determine a coding unit CU
  • the fourth determining module 620 is configured to: in the coding unit CU, the CU of the view of the texture map, and according to the size of the CU, the current optimal prediction unit PU partition mode of the CU, and the current optimal PU partition mode of the CU.
  • the optimal PU partitioning manner of the CU determining that the first asymmetric PU partitioning manner of the CU does not meet the first encoding mode determination condition and conforms to the second encoding mode determining condition, or determining the first CU
  • the first asymmetric PU partitioning mode of the CU is coded by using the normal Normal mode, and the Normal mode refers to a complete motion prediction compensation mode.
  • the coding apparatus of the embodiment of the present invention can perform coding detection by preferentially performing the Normal mode on the asymmetric PU segmentation mode of the view, thereby improving coding efficiency.
  • the fourth determining module 620 is further configured to: in the CU where the CU is a view of the depth map, and according to a size of the CU, a current optimal PU split mode of the CU, and a current optimal PU split mode of the CU.
  • Encoding mode and an optimal PU segmentation mode of the CU of the CU determining that the first asymmetric PU segmentation mode of the CU does not meet the first coding mode determination condition, and when the second coding mode determination condition is met,
  • the first asymmetric PU partitioning mode of the CU adopts the merged Merge method for encoding detection.
  • the fourth determining module 620 is further configured to: in the CU where the CU is a view of the depth map, and according to a size of the CU, a current optimal PU split mode of the CU, and a current optimal PU split mode of the CU.
  • the Normal mode performs code detection.
  • the encoding device 600 can implement the various processes implemented by the encoding device in the embodiments of FIG. 4 and FIG. 5. To avoid repetition, details are not described herein again.
  • the disclosed systems, devices, and methods may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a standalone product. Based on such understanding, the present invention
  • the technical solution in essence or the part contributing to the prior art or part of the technical solution may be embodied in the form of a software product stored in a storage medium, including a plurality of instructions for making one
  • the computer device (which may be a personal computer, server, or network device, etc.) performs all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, 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

本发明公开了一种非对称运动分割方式编码的方法和编码设备,该方法包括:确定编码单元CU;在该CU为纹理图的非基本视点的CU且根据该CU的大小、目前最优预测单元PU分割方式、目前最优PU分割方式的编码方式和上一级CU的最优PU分割方式,确定该第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,根据该CU所属的视点的视间参考视点信息确定该第一非对称PU分割方式的编码检测方式。本发明实施例的非对称运动分割方式编码的方法和编码设备,通过对基本视点和非基本视点的CU分别进行编码方式的确定,并参考视间参考视点信息确定非基本视点的CU的编码检测方式,可以有效提高编码效率。

Description

非对称运动分割方式编码的方法和编码设备
本申请要求于2014年1月3日提交中国专利局、申请号为201410003971.1、发明名称为“非对称运动分割方式编码的方法和编码设备”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
本发明涉及图像处理领域,尤其涉及图像处理领域中非对称运动分割方式编码的方法和编码设备。
背景技术
为了提高视频编码效率,在原有的四种对称的运动分割方式的基础上,增加了四种非对称运动分割方式(asymmetric motion partitions,AMP)。在采用AMP技术的同时亦应用了一些针对AMP的快速实现算法。AMP快速实现算法主要包括:确定哪些非对称运动分割方式采用正常Normal方式进行编码检测或确定哪些非对称运动分割方式采用合并Merge方式进行编码检测。
目前,该AMP快速实现算法广泛应用在3D视频的纹理图和深度图的各个视点的编码过程中。但由于3D视频编码增加的视间预测、纹理图和深度图间的预测参考关系,使得AMP快速实现算法在每个视点的纹理图和深度图的重要性并不一致,影响了3D视频的编码效率。
发明内容
本发明实施例提供了一种非对称运动分割方式编码的方法和编码设备,确定对纹理图、深度图的基本视点和非基本视点的编码单元CU分别所采取的编码检测方式,可以有效提高编码效率。
第一方面,本发明实施例提供了一种非对称运动分割方式编码的方法,该方法包括:确定编码单元CU;在CU为纹理图的非基本视点的CU且根据CU的大小、CU的目前最优预测单元PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,根据CU所属的视点的视间参考视点信息确定CU的第一 非对称PU分割方式的编码检测方式。
在第一方面的第一种可能的实现方式中,根据该第一CU所属的视点的视间参考视点信息确定该第一CU的第一非对称PU分割方式的编码检测方式,包括:在根据CU所属的视点的视间参考视点信息,确定CU的第一非对称PU分割方式符合第三编码方式判断条件时,CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,Normal方式指完整的运动预测补偿方式;或在根据CU所属的视点的视间参考视点信息,确定CU的第一非对称PU分割方式不符合第三编码方式判断条件时,对CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
结合第一方面或第一方面的第一种可能的实现方式,在第一方面的第二种可能的实现方式中,视间参考视点信息包括:CU所属的视点的视间参考视点的CU的编码方式和视间参考视点的CU的对称PU分割方式。
结合第一方面或第一方面的第一种至第二种可能的实现方式中的任一种可能的实现方式,在第一方面的第三种可能的实现方式中,该方法还包括:在CU为纹理图的非基本视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
结合第一方面或第一方面的第一种至第三种可能的实现方式中的任一种可能的实现方式,在第一方面的第四种可能的实现方式中,该方法还包括:在CU为纹理图的基本视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
结合第一方面或第一方面的第一种至第四种可能的实现方式中的任一种可能的实现方式,在第一方面的第五种可能的实现方式中,该方法还包括:在CU为纹理图的基本视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
结合第一方面或第一方面的第一种至第五种可能的实现方式中的任一种可能的实现方式,在第一方面的第六种可能的实现方式中,该方法还包括:在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,对CU的第一非对称PU分割方式采用Merge方式进行编码检测。
结合第一方面或第一方面的第一种至第六种可能的实现方式中的任一种可能的实现方式,在第一方面的第七种可能的实现方式中,该方法还包括:在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
第二方面,本发明实施例提供了一种非对称运动分割方式编码的方法,该方法包括:确定编码单元CU;在编码单元CU为纹理图的视点的CU且根据CU的大小、CU的目前最优预测单元PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件并且符合第二编码方式判断条件,或确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,Normal方式指完整的运动预测补偿方式。
在第二方面的第一种可能的实现方式中,该方法还包括:在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,对CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
结合第二方面或第二方面的第一种可能的实现方式,在第二方面的第二种可能的实现方式中,该方法还包括:在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称 PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
第三方面,本发明实施例提供了一种编码设备,该编码设备包括:第一确定模块,用于确定编码单元CU;第二确定模块,用于在CU为纹理图的非基本视点的CU且根据CU的大小、CU的目前最优预测单元PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,根据CU所属的视点的视间参考视点信息确定CU的第一非对称PU分割方式的编码检测方式。
在第三方面的第一种可能的实现方式中,第二确定模块具体用于:在CU为纹理图的非基本视点的CU且根据CU的大小、CU的目前最优预测单元PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件,并且根据CU所属的视点的视间参考视点信息,确定CU的第一非对称PU分割方式符合第三编码方式判断条件时,CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,Normal方式指完整的运动预测补偿方式;或在CU为纹理图的非基本视点的CU且根据CU的大小、CU的目前最优预测单元PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件,并且根据CU所属的视点的视间参考视点信息,确定CU的第一非对称PU分割方式不符合第三编码方式判断条件时,对CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
结合第三方面或第三方面的第一种可能的实现方式,在第三方面的第二种可能的实现方式中,CU所属的视点的视间参考视点的CU的编码方式和视间参考视点的CU的对称PU分割方式。
结合第三方面或第三方面的第一种至第二种可能的实现方式中的任一种可能的实现方式,在第一方面的第三种可能的实现方式中,第二确定模块,还用于在CU为纹理图的非基本视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式符合第一编码 方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
结合第三方面或第三方面的第一种至第三种可能的实现方式中的任一种可能的实现方式,在第三方面的第四种可能的实现方式中,第二确定模块,还用于在CU为纹理图的基本视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
结合第三方面或第三方面的第一种至第四种可能的实现方式中的任一种可能的实现方式,在第三方面的第五种可能的实现方式中,第二确定模块,还用于在CU为纹理图的基本视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
结合第三方面或第三方面的第一种至第五种可能的实现方式中的任一种可能的实现方式,在第三方面的第六种可能的实现方式中,第二确定模块,还用于在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,对CU的第一非对称PU分割方式采用Merge方式进行编码检测。
结合第三方面或第三方面的第一种至第六种可能的实现方式中的任一种可能的实现方式,在第三方面的第七种可能的实现方式中,第二确定模块,还用于在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
第四方面,本发明实施例提供了一种编码设备,该编码设备包括:第三确定模块,用于确定编码单元CU;第四确定模块,用于在编码单元CU为 纹理图的视点的CU且根据CU的大小、CU的目前最优预测单元PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件并且符合第二编码方式判断条件,或确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,Normal方式指完整的运动预测补偿方式。
在第四方面的第一种可能的实现方式中,第四确定模块,还用于在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,对CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
结合第四方面或第四方面的第一种可能的实现方式,在第四方面的第二种可能的实现方式中,第四确定模块,还用于在CU为深度图的视点的CU且根据CU的大小、CU的目前最优PU分割方式、CU的目前最优PU分割方式的编码方式和CU的上一级CU的最优PU分割方式,确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对CU的第一非对称PU分割方式采用Normal方式进行编码检测。
基于上述技术方案,本发明实施例的非对称运动分割方式编码的方法和编码设备,通过分别确定对纹理图、深度图的基本视点和非基本视点的编码单元CU所采取的编码检测方式,并参考视间参考视点信息确定非基本视点的CU的编码检测方式,可以有效提高编码效率。
附图说明
为了更清楚地说明本发明实施例的技术方案,下面将对本发明实施例中所需要使用的附图作简单地介绍,显而易见地,下面所描述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是编码过程中非对称运动分割方式AMP的示意图。
图2是根据本发明实施例的非对称运动分割方式编码的方法的示意性流程图。
图3是根据本发明另一实施例的非对称运动分割方式编码的方法的示意性流程图。
图4是根据本发明另一实施例的非对称运动分割方式编码的方法的示意性流程图。
图5是根据本发明另一实施例的非对称运动分割方式编码的方法的示意性流程图。
图6是本发明一个实施例的编码设备的示意性框图。
图7是本发明另一个实施例的编码设备的示意性框图。
具体实施方式
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,所描述的实施例是本发明的一部分实施例,而不是全部实施例。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的前提下所获得的所有其他实施例,都应属于本发明保护的范围。
图1示出了编码过程中非对称运动分割方式AMP的示意图。
视频编解码结构中包括三种单元:编码单元(Coding Unit,CU)、预测单元(Prediction Unit,PU)和变换单元(Transform Unit,TU)。其中,PU包括用于预测过程的预测块及预测块的内容信息(如参考索引,运动矢量等),一般有四种对称PU分割方式:2N×2N、2N×N、N×2N和N×N。为了进一步提高编码效率,在原有的四种对称PU分割方式的基础上,新增加了非对称运动分割方式AMP。即如图1所示,在帧间编码过程中,CU有2N×nU、2N×nD、nR×2N和nL×2N四种非对称PU分割方式。
对3D视频进行编码时,对于每一视点,需要对深度图和纹理图分别进行编码。在帧间编码过程中,通常将一个视点作为基本视点(Base View),将其他视点作为非基本视点。可以对该基本视点采用与其他视点不同的编码方式,例如采用传统高效视频编码(High Efficiency Video Coding,HEVC)方式。而对非基本视点编码时,为了提高编码效率,可以参考基本视点的编码信息,并且除了利用传统HEVC方式之外,更多的编码工具,例如视差补偿预测技术,视差矢量获取技术,视间运动预测,视间残差预测,亮度补偿,视点合成预测等,均可应用在非基本视点的编码过程中。
对每一视点的每一CU的四种非对称PU分割方式中的每一种,需要确 定其是采用正常Normal方式进行编码检测还是采用合并Merge方式进行编码检测。
应理解,本发明实施例的采用Normal方式进行编码检测,是指采用完整的运动预测补偿方式进行编码检测。
图2示出了根据本发明实施例的非对称运动分割方式编码的方法100的示意性流程图,该方法100可以由编码设备执行。
S101,确定编码单元CU;
S102,在该CU为纹理图的非基本视点的CU且根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,根据该CU所属的视点的视间参考视点信息确定该CU的第一非对称PU分割方式的编码检测方式。
,因此,本发明实施例的非对称运动分割方式编码的方法,通过分别确定对纹理图、深度图的基本视点和非基本视点的编码单元CU所采取的编码检测方式,并参考视间参考视点信息确定非基本视点的CU的编码检测方式,可以有效提高编码效率。
本发明实施例中,基本视点也可称之为独立视点,是首先进行编码的视点。
具体而言,在S101中,确定当前正在编码的编码单元CU。在S102中,在对该CU进行编码时,编码设备首先确定编码是否处于帧间编码过程。在确定编码处于帧间编码过程后,编码设备确定当前是处于纹理图编码过程还是深度图编码过程。本发明实施例中,可选地,编码设备确定当前处于纹理图的编码过程,并且确定该CU所属的视点是非基本视点。
进一步地,根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件。
具体地,作为一个实施例,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为2N×N,则可以确定该CU的非对称PU分割方式2N×nU和2N×nD为符合第一编码方式判断条件;或者,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为2N×2N,并且 该CU的目前最优预测单元PU分割方式的编码方式不为merge或skip模式,则也可以确定该CU的非对称PU分割方式2N×nU和2N×nD为符合第一编码方式判断条件。
可选地,作为另一实施例,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为N×2N,则可以确定该CU的非对称PU分割方式nL×2N和nR×2N为符合第一编码方式判断条件;或者,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为2N×2N,并且该CU的目前最优预测单元PU分割方式的编码方式不为merge或skip模式,则也可以确定该CU的非对称PU分割方式nL×2N和nR×2N为符合第一编码方式判断条件。反之,则为不符合第一编码方式判断条件。
根据该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合第二编码方式判断条件。
具体地,作为一个实施例,该CU的目前最优PU分割方式为2N×2N,并且该CU的目前最优PU分割方式的编码方式不为skip编码方式;或者,该CU的上一级CU的最优PU分割方式为AMP方式;或者,该CU的目前最优PU分割方式为2N×N,并且该CU的上一级CU的最优PU分割方式为非PU分割方式,则可以确定该CU的非对称PU分割方式2N×nU和2N×nD为符合第二编码方式判断条件。
可选地,作为另一实施例,该CU的目前最优PU分割方式为2N×2N,并且该CU的目前最优PU分割方式的编码方式不为skip编码方式;或者,该CU的上一级CU的最优PU分割方式为AMP方式;或者,该CU的目前最优PU分割方式为N×2N,并且该CU的上一级CU的最优PU分割方式为非PU分割方式,则可以确定该CU的非对称PU分割方式nL×2N和nR×2N为符合第二编码方式判断条件。反之,则为不符合第二编码方式判断条件。
应理解,在本发明实施例中,该CU的第一非对称PU分割方式不符合第一编码方式判断条件,符合第二编码方式判断条件。除第一非对称PU分割方式以外,该CU的其它非对称PU分割方式可以符合第一编码方式判断条件;也可以不符合第一编码方式判断条件,符合第二编码方式判断条件;还可以既不符合第一编码方式判断条件,也不符合第二编码方式判断条件。根据判断结果不同,其最终采取的编码检测方式可能不同,本发明实施例对 此不作限定。
在确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,符合第二编码方式判断条件之后,根据该CU所属的视点的视间参考视点信息确定该CU的第一非对称PU分割方式的编码检测方式。其中,该CU所属的视点的视间参考视点可以是基本视点,也可以是非基本视点。视间参考视点信息可以包括视间参考视点的CU的编码方式和该CU的对称PU分割方式。第三编码方式判断条件为基于上述视间参考视点信息的判断条件。
具体地,作为一个实施例,视间参考视点的CU的最优PU分割方式为2N×N;或者,视间参考视点的CU的最优PU分割方式为2N×2N并且视间参考视点的CU的编码方式不为merge或skip模式,则可以确定该CU的非对称PU分割方式2N×nU和2N×nD为符合第三编码方式判断条件。
可选地,作为另一实施例,视间参考视点的CU的最优PU分割方式为N×2N;或者,视间参考视点的CU的最优PU分割方式为2N×2N并且视间参考视点的CU的编码方式不为merge或skip模式,则可以确定该CU的非对称PU分割方式nL×2N和nR×2N为符合第三编码方式判断条件。反之,则为不符合第三编码方式判断条件。
可选地,作为一个实施例,该CU的第一非对称PU分割方式符合第三编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
可选地,作为另一个实施例,该CU的第一非对称PU分割方式不符合第三编码方式判断条件时,对该CU的第一非对称PU分割方式采用Merge方式进行编码检测。
下面将结合具体的例子,对方法100进行说明。
在视频编码过程中,对纹理图的一个非基本视点的16×16大小的CU进行编码。该CU的2N×nU和2N×nD不符合第一判断条件,符合第二判断条件。在此基础上,继而判断该CU的2N×nU和2N×nD不符合第三判断条件,则对该CU的2N×nU和2N×nD采用Merge方式进行编码检测。如果该CU的2N×nU和2N×nD在不符合第一判断条件,符合第二判断条件地基础上,符合第三判断条件,则对该非基本视点的2N×nU和2N×nD采用Normal方式进行编码检测。
图2是CU为纹理图的非基本视点的CU,并且该CU的第一非对称PU 分割方式不符合第一判断条件,符合第二判断条件的情况的说明。在本发明实施例中,对于一个纹理图的非基本视点的CU,其四种非对称PU分割方式中的一种可能符合第一编码方式判断条件。相应地,该方法100还可以包括:
在该CU为纹理图的非基本视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
具体而言,确定该CU为纹理图的非基本视点的CU,并且确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。例如,对纹理图的一个非基本视点的16×16大小的CU进行编码,该CU的nL×2N和nR×2N符合第一判断条件,则对该CU的nL×2N和nR×2N采用Normal方式进行编码检测。
应理解,对于纹理图的一个非基本视点的CU的非对称PU分割方式,其可能既不符合第一编码方式判断条件,也不符合第二编码方式判断条件,此时,不对该CU的该非对称PU分割方式进行编码检测。
以上对纹理图的非基本视点的CU进行了说明。在本发明实施例中,纹理图还包括基本视点的CU,其非对称PU分割方式中的一种可能不符合第一编码方式判断条件,符合第二编码方式判断条件。相应地,该方法100还可以包括:
在该CU为纹理图的基本视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合该第一编码方式判断条件,并且符合该第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
具体而言,确定该CU为纹理图的基本视点的CU,并且确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,符合第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。例如,对纹理图的一个基本视点的32×32大小的CU进行编码,该 CU的2N×nU和2N×nD不符合第一判断条件,符合第二判断条件,则对该CU的2N×nU和2N×nD采用Normal方式进行编码检测。
以上是CU为纹理图的基本视点,并且该CU的非对称PU分割方式不符合第一判断条件,符合第二判断条件的情况的说明。在本发明实施例中,对于一个纹理图的基本视点的CU,其四种非对称PU分割方式中的一种可能符合第一编码方式判断条件。相应地,该方法100还可以包括:
在该CU为纹理图的基本视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合该第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
具体而言,确定该CU为纹理图的基本视点的CU,并且确定CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。例如,对纹理图的一个基本视点的32×32大小的CU进行编码,该CU的nL×2N和nR×2N符合第一判断条件,则对该CU的nL×2N和nR×2N采用Normal方式进行编码检测。
应理解,对于纹理图的一个基本视点的CU的非对称PU分割方式,其可能既不符合第一编码方式判断条件,也不符合第二编码方式判断条件,此时,不对该CU的该非对称PU分割方式进行编码检测。
以上对纹理图的视点的CU进行了说明。在本发明实施例中,对于深度图的视点,其CU的非对称PU分割方式中的一种可能不符合第一编码方式判断条件,符合第二编码方式判断条件。相应地,该方法100还可以包括:
在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合该第一编码方式判断条件,并且符合该第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用Merge方式进行编码检测。
具体而言,确定该CU为深度图的视点的CU,并且确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,符合第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用Merge方式进行编码检测。例如,对深度图的一个视点的32×32大小的CU进行编码,该CU的2N×nU 和2N×nD不符合第一判断条件,符合第二判断条件,则对该CU的2N×nU和2N×nD采用Merge方式进行编码检测。
应理解,本发明实施例的CU所属的深度图的视点可以是基本视点,也可以是非基本视点,本发明实施例对此不作限定。
以上是CU为深度图的视点的CU,并且该CU的非对称PU分割方式不符合第一判断条件,符合第二判断条件的情况的说明。在本发明实施例中,对于一个深度图的视点的CU,其四种非对称PU分割方式中的一种可能符合第一编码方式判断条件。相应地,该方法100还可以包括:
在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合该第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
具体而言,确定该CU为深度图的视点的CU,并且确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。例如,对深度图的一个视点的32×32大小的CU进行编码,该CU的nL×2N和nR×2N符合第一判断条件,则对该CU的nL×2N和nR×2N采用Normal方式进行编码检测。
应理解,对于深度图的一个视点的CU的非对称PU分割方式,其可能既不符合第一编码方式判断条件,也不符合第二编码方式判断条件,此时,不对该CU的该非对称PU分割方式进行编码检测。
还应理解,本发明实施例的深度图的视点可以是基本视点,也可以是非基本视点,本发明实施例对此不作限定。
应理解,在本发明实施例中,优选的,图2对应的各实施例是并列执行的,对其执行的先后顺序不作限定。
图3示出了根据本发明另一实施例的非对称运动分割方式编码的方法200的示意性流程图。该方法由编码设备执行,图3所示的方法200包括:
S201,确定当前进行编码的CU是否深度图的视点的CU。
具体地,在帧间编码过程中,判断当前正在编码的视点的CU是否深度图的视点的CU。如果是,执行S202;如果否,即为纹理图的视点的CU,执行S206。
S202,确定深度图的视点的CU的非对称PU分割方式是否符合第一编码方式判断条件。
具体地,当前正在编码的CU为深度图的视点的CU时,确定该CU的非对称PU分割方式是否符合第一编码方式判断条件。如果是,执行S203;如果否,执行S204。
S203,对符合第一编码方式判断条件的深度图的视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
S204,确定深度图的视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。
具体地,在深度图的视点的CU的非对称PU分割方式不符合第一编码方式判断条件的基础上,确定深度图的视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。如果是,执行S205;如果否,结束。
S205,对不符合第一编码方式判断条件,符合第二编码方式判断条件的深度图的视点的CU的非对称PU分割方式采用Merge方式进行编码检测,完成后结束。
S206,确定当前进行编码的CU所属的视点是否基本视点。
具体地,在帧间编码对纹理图编码的过程中,判断当前正在编码的CU所属的视点是否基本视点。如果是,执行S207;如果否,执行S211。
S207,确定纹理图的基本视点的CU的非对称PU分割方式是否符合第一编码方式判断条件。
具体地,当确定当前正在编码的CU为纹理图的基本视点的CU时,确定该CU的非对称PU分割方式是否符合第一编码方式判断条件。如果是,执行S208;如果否,执行S209。
S208,对符合第一编码方式判断条件的纹理图的基本视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
S209,确定纹理图的基本视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。
具体地,在纹理图的基本视点的CU的非对称PU分割方式不符合第一编码方式判断条件的基础上,确定纹理图的基本视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。如果是,执行S210;如果否,结束。
S210,对不符合第一编码方式判断条件,符合第二编码方式判断条件的纹理图的基本视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
S211,确定纹理图的非基本视点的CU的非对称PU分割方式是否符合第一编码方式判断条件。
具体地,当确定当前正在编码的CU为纹理图的非基本视点的CU时,确定该CU的非对称PU分割方式是否符合第一编码方式判断条件。如果是,执行S212;如果否,执行S213。
S212,对符合第一编码方式判断条件的纹理图的非基本视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
S213,确定纹理图的非基本视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。
具体地,在纹理图的非基本视点的CU的非对称PU分割方式不符合第一编码方式判断条件的基础上,确定该CU的非对称PU分割方式是否符合第二编码方式判断条件。如果是,执行S214;如果否,结束。
S214,确定纹理图的非基本视点的CU的非对称PU分割方式是否符合第三编码方式判断条件。
具体地,在纹理图的非基本视点的CU的非对称PU分割方式不符合第一编码方式判断条件,符合第二编码方式判断条件的基础上,确定该CU的非对称PU分割方式是否符合第三编码方式判断条件。如果是,执行S215;如果否,执行S216。
S215,对不符合第一编码方式判断条件,符合第二编码方式判断条件,并且符合第三编码方式判断条件的纹理图的非基本视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
S216,对不符合第一编码方式判断条件,符合第二编码方式判断条件,并且不符合第三编码方式判断条件的纹理图的非基本视点的CU的非对称PU分割方式采用Merge方式进行编码检测,完成后结束。
因此,本发明实施例的非对称运动分割方式编码的方法,通过分别确定对纹理图、深度图的基本视点和非基本视点的编码单元CU所采取的编码检测方式,并参考视间参考视点信息确定非基本视点的CU的编码检测方式,可以有效提高编码效率。
应理解,本发明实施例的各过程的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本发明实施例的实施过程构成任何限定。
图4示出了根据本发明实施例的非对称运动分割方式编码的方法300的示意性流程图,该方法300可以由编码设备执行,包括。
S301,确定编码单元CU;
S302,在编码单元CU为纹理图的视点的CU且根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件并且符合第二编码方式判断条件,或确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,该Normal方式指完整的运动预测补偿方式。
因此,本发明实施例的非对称运动分割方式编码的方法,通过对视点的非对称PU分割方式优先执行Normal方式进行编码检测,可以提高编码效率。
应理解,本发明实施例的采用Normal方式进行编码检测,是指采用完整的运动预测补偿方式进行编码检测。
具体而言,在S301中,确定当前正在编码的编码单元CU。在S302中,编码设备首先确定编码是否处于帧间编码过程。在确定编码处于帧间编码过程后,编码设备确定当前是处于纹理图编码过程还是深度图编码过程。本发明实施例中,可选地,编码设备确定该CU是纹理图的视点的CU。
根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件。
具体地,作为一个实施例,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为2N×N,则可以确定该CU的非对称PU分割方式2N×nU和2N×nD为符合第一编码方式判断条件;或者,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为2N×2N,并且该CU的目前最优预测单元PU分割方式的编码方式不为merge或skip模式,则也可以确定该CU的非对称PU分割方式2N×nU和2N×nD为符合第一编 码方式判断条件。
可选地,作为另一实施例,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为N×2N,则可以确定该CU的非对称PU分割方式nL×2N和nR×2N为符合第一编码方式判断条件;或者,该CU的大小不为64×64,并且该CU的目前最优预测单元PU分割方式为2N×2N,并且该CU的目前最优预测单元PU分割方式的编码方式不为merge或skip模式,则也可以确定该CU的非对称PU分割方式nL×2N和nR×2N为符合第一编码方式判断条件。反之,则为不符合第一编码方式判断条件。
进一步,根据该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合第二编码方式判断条件。
具体地,作为一个实施例,该CU的目前最优PU分割方式为2N×2N,并且该CU的目前最优PU分割方式的编码方式不为skip编码方式;或者,该CU的上一级CU的最优PU分割方式为AMP方式;或者,该CU的目前最优PU分割方式为2N×N,并且该CU的上一级CU的最优PU分割方式为非PU分割方式,则可以确定该CU的非对称PU分割方式2N×nU和2N×nD为符合第二编码方式判断条件。
可选地,作为另一实施例,该CU的目前最优PU分割方式为2N×2N,并且该CU的目前最优PU分割方式的编码方式不为skip编码方式;或者,该CU的上一级CU的最优PU分割方式为AMP方式;或者,该CU的目前最优PU分割方式为N×2N,并且该CU的上一级CU的最优PU分割方式为非PU分割方式,则可以确定该CU的非对称PU分割方式nL×2N和nR×2N为符合第二编码方式判断条件。反之,则为不符合第二编码方式判断条件。
在确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,符合第二编码方式判断条件时,或者确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
应理解,在本发明实施例中,该CU的第一非对称PU分割方式可以既不符合第一编码方式判断条件,也不符合第二编码方式判断条件。此时,不对该CU的该非对称PU分割方式进行编码检测。
还应理解,在本发明实施例中,该CU所属的视点可以是纹理图的基本 视点也可以是纹理图的非基本视点。
以上是对纹理图的视点的CU的说明。在本发明实施例中,该CU还可能是深度图的视点的CU,在其非对称PU分割方式中的一种可能不符合第一编码方式判断条件,符合第二编码方式判断条件时,相应地,该方法300还可以包括:
在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合该第一编码方式判断条件,并且符合该第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
具体而言,确定该CU为深度图的视点的CU,并且确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,符合第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用Merge方式进行编码检测。例如,对深度图的一个视点的32×32大小的CU进行编码,该CU的2N×nU和2N×nD不符合第一判断条件,符合第二判断条件,则对该CU的2N×nU和2N×nD采用Merge方式进行编码检测。
应理解,本发明实施例的CU所属的深度图的视点可以是基本视点,也可以是非基本视点,本发明实施例对此不作限定。
以上是对CU为深度图的视点的CU,并且该CU的非对称PU分割方式不符合第一判断条件,符合第二判断条件的情况的说明。在本发明实施例中,对于一个深度图的视点的CU,其四种非对称PU分割方式中的一种可能符合第一编码方式判断条件。相应地,该方法300还可以包括:
在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合该第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
具体而言,确定该CU为深度图的视点的CU,并且确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。例如,对深度图的一个视点的32×32大小的CU进行编码,该CU的nL×2N和nR×2N符合第一判断条件, 则对该CU的nL×2N和nR×2N采用Normal方式进行编码检测。
应理解,对于深度图的一个视点的CU的非对称PU分割方式,其可能既不符合第一编码方式判断条件,也不符合第二编码方式判断条件,此时,不对该CU的该非对称PU分割方式进行编码检测。
还应理解,本发明实施例的CU所属的深度图的视点可以是基本视点,也可以是非基本视点,本发明实施例对此不作限定。在本发明实施例中,图4对应的各实施例是并列执行的,对其执行的先后顺序不作限定。
图5示出了根据本发明另一实施例的非对称运动分割方式编码的方法400的示意性流程图。该方法由编码设备执行,图5所示的方法400包括:
S401,确定当前进行编码的CU是否深度图的视点的CU。
具体地,在帧间编码过程中,判断当前正在编码的视点的CU是否深度图的视点的CU。如果是,执行S402;如果否,即为纹理图的视点的CU,执行S406。
S402,确定深度图的视点的CU的非对称PU分割方式是否符合第一编码方式判断条件。
具体地,当前正在编码的CU为深度图的视点的CU时,确定该CU的非对称PU分割方式是否符合第一编码方式判断条件。如果是,执行S403;如果否,执行S404。
S403,对符合第一编码方式判断条件的深度图的视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
S404,确定深度图的视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。
具体地,在深度图的视点的CU的非对称PU分割方式不符合第一编码方式判断条件的基础上,确定深度图的视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。如果是,执行S405;如果否,结束。
S405,对不符合第一编码方式判断条件,符合第二编码方式判断条件的深度图的视点的CU的非对称PU分割方式采用Merge方式进行编码检测,完成后结束。
S406,确定纹理图的视点的CU的非对称PU分割方式是否符合第一编码方式判断条件。
具体地,确定当前正在编码的纹理图的视点的CU的非对称PU分割方 式是否符合第一编码方式判断条件。如果是,执行S407;如果否,执行S408。
S407,对符合第一编码方式判断条件的纹理图的视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
S408,确定纹理图的视点的CU的非对称PU分割方式是否符合第二编码方式判断条件。
具体地,在纹理图的视点的CU的非对称PU分割方式不符合第一编码方式判断条件的基础上,确定该CU的非对称PU分割方式是否符合第二编码方式判断条件。如果是,执行S409;如果否,结束。
S409,对不符合第一编码方式判断条件,符合第二编码方式判断条件的纹理图的视点的CU的非对称PU分割方式采用Normal方式进行编码检测,完成后结束。
因此,本发明实施例的非对称运动分割方式编码的方法,通过对视点的非对称PU分割方式优先执行Normal方式进行编码检测,可以提高编码效率。
应理解,在本发明实施例中,视点可以是基本视点也可以是非基本视点。
还应理解,本发明实施例的各过程的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本发明实施例的实施过程构成任何限定。
图6示出了本发明一个实施例的编码设备的示意性框图。如图6所示,编码设备500包括:
第一确定模块510,用于确定编码单元CU;
第二确定模块520,用于在该CU为纹理图的非基本视点的CU且根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,根据该CU所属的视点的视间参考视点信息确定该CU的第一非对称PU分割方式的编码检测方式。
因此,本发明实施例的编码设备,通过分别确定对纹理图、深度图的基本视点和非基本视点的编码单元CU所采取的编码检测方式,并引入视间参考视点信息确定非基本视点的CU的编码检测方式,可以有效提高编码效率。
可选地,第二确定模块520可具体用于:在该CU为纹理图的非基本视 点的CU且根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件,并且根据该CU所属的视点的视间参考视点信息,确定该CU的第一非对称PU分割方式符合第三编码方式判断条件时,该CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,该Normal方式指完整的运动预测补偿方式;或
在该CU为纹理图的非基本视点的CU且根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件,并且根据该CU所属的视点的视间参考视点信息,确定该CU的第一非对称PU分割方式不符合该第三编码方式判断条件时,对该CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。其中,该CU所属的视点的视间参考视点的CU的编码方式和视间参考视点的CU的对称PU分割方式。
可选地,第二确定模块520,还可用于在该CU为纹理图的非基本视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
可选地,第二确定模块520,还可用于在该CU为纹理图的基本视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合该第一编码方式判断条件,并且符合该第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
可选地,第二确定模块520,在该CU为纹理图的基本视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合该第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
可选地,第二确定模块520,在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合该第一编码方式判断条件,并且符合该第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用Merge方式进行编码检测。
可选地,第二确定模块520,还可用于在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合该第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
编码设备500能够实现图2和图3的实施例中由编码设备实现的各个过程,为避免重复,这里不再赘述。
图7示出了本发明另一个实施例的编码设备的示意性框图。如图7所示,编码设备600包括:
第三确定模块610,用于确定编码单元CU;
第四确定模块620,用于在编码单元CU为纹理图的视点的CU且根据该CU的大小、该CU的目前最优预测单元PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合第一编码方式判断条件并且符合第二编码方式判断条件,或确定该CU的第一非对称PU分割方式符合第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,该Normal方式指完整的运动预测补偿方式。
因此,本发明实施例的编码设备,通过对视点的非对称PU分割方式优先执行Normal方式进行编码检测,可以提高编码效率。
可选地,第四确定模块620,还可用于在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式不符合该第一编码方式判断条件,并且符合该第二编码方式判断条件时,对该CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
可选地,第四确定模块620,还可用于在该CU为深度图的视点的CU且根据该CU的大小、该CU的目前最优PU分割方式、该CU的目前最优PU分割方式的编码方式和该CU的上一级CU的最优PU分割方式,确定该CU的第一非对称PU分割方式符合该第一编码方式判断条件时,对该CU的第一非对称PU分割方式采用Normal方式进行编码检测。
编码设备600能够实现图4和图5的实施例中由编码设备实现的各个过程,为避免重复,这里不再赘述。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本发明的范围。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的***、装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中,应该理解到,所揭露的***、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个***,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本发明 的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本发明的具体实施方式,但本发明的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本发明揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本发明的保护范围之内。因此,本发明的保护范围应以权利要求的保护范围为准。

Claims (22)

  1. 一种非对称运动分割方式编码的方法,其特征在于,所述方法包括:
    确定编码单元CU;
    在所述CU为纹理图的非基本视点的CU且根据所述CU的大小、所述CU的目前最优预测单元PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,根据所述CU所属的视点的视间参考视点信息确定所述CU的第一非对称PU分割方式的编码检测方式。
  2. 根据权利要求1所述的方法,其特征在于,所述根据所述CU所属的视点的视间参考视点信息确定所述CU的第一非对称PU分割方式的编码检测方式,包括:
    在根据所述CU所属的视点的视间参考视点信息,确定所述CU的第一非对称PU分割方式符合第三编码方式判断条件时,所述CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,所述Normal方式指完整的运动预测补偿方式;或
    在根据所述CU所属的视点的视间参考视点信息,确定所述CU的第一非对称PU分割方式不符合所述第三编码方式判断条件时,对所述CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
  3. 根据权利要求1或2所述的方法,其特征在于,所述视间参考视点信息包括:
    所述CU所属的视点的视间参考视点的CU的编码方式和所述视间参考视点的CU的对称PU分割方式。
  4. 根据权利要求1至3中任一项所述的方法,其特征在于,所述方法还包括:
    在所述CU为纹理图的非基本视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  5. 根据权利要求1至4中任一项所述的方法,其特征在于,所述方法 还包括:
    在所述CU为纹理图的基本视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合所述第一编码方式判断条件,并且符合所述第二编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  6. 根据权利要求1至5中任一项所述的方法,其特征在于,所述方法还包括:
    在所述CU为纹理图的基本视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合所述第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  7. 根据权利要求1至6中任一项所述的方法,其特征在于,所述方法还包括:
    在所述CU为深度图的视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合所述第一编码方式判断条件,并且符合所述第二编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Merge方式进行编码检测。
  8. 根据权利要求1至7中任一项所述的方法,其特征在于,所述方法还包括:
    在所述CU为深度图的视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合所述第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  9. 一种非对称运动分割方式编码的方法,其特征在于,所述方法包括:
    确定编码单元CU;
    在编码单元CU为纹理图的视点的CU且根据所述CU的大小、所述CU的目前最优预测单元PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合第一编码方式判断条件并且符合第二编码方式判断条件,或确定所述CU的第一非对称PU分割方式符合第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,所述Normal方式指完整的运动预测补偿方式。
  10. 根据权利要求9所述的方法,其特征在于,所述方法还包括:
    在所述CU为深度图的视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合所述第一编码方式判断条件,并且符合所述第二编码方式判断条件时,对所述CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
  11. 根据权利要求1至10中任一项所述的方法,其特征在于,所述方法还包括:
    在所述CU为深度图的视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合所述第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  12. 一种编码设备,其特征在于,所述编码设备包括:
    第一确定模块,用于确定编码单元CU;
    第二确定模块,用于在所述CU为纹理图的非基本视点的CU且根据所述CU的大小、所述CU的目前最优预测单元PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件时,根据所述CU所属的视点的视间参考视点信息确定所述CU的第一非对称PU分割方式的编码检测方式。
  13. 根据权利要求12所述的编码设备,其特征在于,所述第二确定模块具体用于:
    在所述CU为纹理图的非基本视点的CU且根据所述CU的大小、所述CU的目前最优预测单元PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件,并且根据所述CU所属的视点的视间参考视点信息,确定所述CU的第一非对称PU分割方式符合第三编码方式判断条件时,所述CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,所述Normal方式指完整的运动预测补偿方式;或
    在所述CU为纹理图的非基本视点的CU且根据所述CU的大小、所述CU的目前最优预测单元PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合第一编码方式判断条件,并且符合第二编码方式判断条件,并且根据所述CU所属的视点的视间参考视点信息,确定所述CU的第一非对称PU分割方式不符合所述第三编码方式判断条件时,对所述CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
  14. 根据权利要求12或13所述的编码设备,其特征在于,所述视间参考视点信息包括:
    所述CU所属的视点的视间参考视点的CU的编码方式和所述视间参考视点的CU的对称PU分割方式。
  15. 根据权利要求12至14中任一项所述的编码设备,其特征在于,
    所述第二确定模块,还用于在所述CU为纹理图的非基本视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  16. 根据权利要求12至15中任一项所述的编码设备,其特征在于,
    所述第二确定模块,还用于在所述CU为纹理图的基本视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合所述第一编码方式判断条件,并且符合所述第二编码方式判断条件时,对所述CU的第一非对称PU分割 方式采用Normal方式进行编码检测。
  17. 根据权利要求12至16中任一项所述的编码设备,其特征在于,
    所述第二确定模块,还用于在所述CU为纹理图的基本视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合所述第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  18. 根据权利要求12至17中任一项所述的编码设备,其特征在于,
    所述第二确定模块,还用于在所述CU为深度图的视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合所述第一编码方式判断条件,并且符合所述第二编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Merge方式进行编码检测。
  19. 根据权利要求12至18中任一项所述的编码设备,其特征在于,
    所述第二确定模块,还用于在所述CU为深度图的视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合所述第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
  20. 一种编码设备,其特征在于,所述编码设备包括:
    第三确定模块,用于确定编码单元CU;
    第四确定模块,用于在编码单元CU为纹理图的视点的CU且根据所述CU的大小、所述CU的目前最优预测单元PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合第一编码方式判断条件并且符合第二编码方式判断条件,或确定所述CU的第一非对称PU分割方式符合第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用正常Normal方式进行编码检测,所述Normal方式指完整的运动预测补偿方式。
  21. 根据权利要求20所述的编码设备,其特征在于,
    所述第四确定模块,还用于在所述CU为深度图的视点的CU且根据所 述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式不符合所述第一编码方式判断条件,并且符合所述第二编码方式判断条件时,对所述CU的第一非对称PU分割方式采用合并Merge方式进行编码检测。
  22. 根据权利要求20或21所述的编码设备,其特征在于,
    所述第四确定模块,还用于在所述CU为深度图的视点的CU且根据所述CU的大小、所述CU的目前最优PU分割方式、所述CU的目前最优PU分割方式的编码方式和所述CU的上一级CU的最优PU分割方式,确定所述CU的第一非对称PU分割方式符合所述第一编码方式判断条件时,对所述CU的第一非对称PU分割方式采用Normal方式进行编码检测。
PCT/CN2014/095287 2014-01-03 2014-12-29 非对称运动分割方式编码的方法和编码设备 WO2015101251A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410003971.1A CN104768012B (zh) 2014-01-03 2014-01-03 非对称运动分割方式编码的方法和编码设备
CN201410003971.1 2014-01-03

Publications (1)

Publication Number Publication Date
WO2015101251A1 true WO2015101251A1 (zh) 2015-07-09

Family

ID=53493227

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/095287 WO2015101251A1 (zh) 2014-01-03 2014-12-29 非对称运动分割方式编码的方法和编码设备

Country Status (2)

Country Link
CN (1) CN104768012B (zh)
WO (1) WO2015101251A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3701721A1 (en) * 2017-10-27 2020-09-02 Huawei Technologies Co., Ltd. Apparatus and method for picture coding with asymmetric partitioning

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110134998A1 (en) * 2009-12-08 2011-06-09 Samsung Electronics Co., Ltd. Method and apparatus for encoding video by motion prediction using arbitrary partition, and method and apparatus for decoding video by motion prediction using arbitrary partition
CN102984521A (zh) * 2012-12-12 2013-03-20 四川大学 基于时域相关性的高性能视频编码帧间模式判决方法
CN103096073A (zh) * 2011-11-07 2013-05-08 吴秀美 构造合并列表的方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107959852A (zh) * 2011-11-08 2018-04-24 株式会社Kt 对视频信号进行解码的方法
US10805617B2 (en) * 2012-01-19 2020-10-13 Texas Instruments Incorporated Scalable prediction type coding

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110134998A1 (en) * 2009-12-08 2011-06-09 Samsung Electronics Co., Ltd. Method and apparatus for encoding video by motion prediction using arbitrary partition, and method and apparatus for decoding video by motion prediction using arbitrary partition
CN103096073A (zh) * 2011-11-07 2013-05-08 吴秀美 构造合并列表的方法
CN102984521A (zh) * 2012-12-12 2013-03-20 四川大学 基于时域相关性的高性能视频编码帧间模式判决方法

Also Published As

Publication number Publication date
CN104768012A (zh) 2015-07-08
CN104768012B (zh) 2018-04-20

Similar Documents

Publication Publication Date Title
WO2016000538A1 (zh) 视频编码的运动矢量列表构建方法和***
KR101846762B1 (ko) 3d 비디오 코딩에서의 서브-pu 움직임 정보 상속 방법
JP2015515227A5 (zh)
JP2016517682A5 (zh)
CN105453561B (zh) 三维以及多视图视频编码中导出默认视差向量的方法
JP2016537841A5 (zh)
JP2016504864A5 (zh)
JP2015519835A5 (zh)
JP2016507969A5 (zh)
WO2014036848A1 (en) Depth picture intra coding /decoding method and video coder/decoder
JP2016530784A5 (zh)
JP2016511604A5 (zh)
KR20120106510A (ko) 3차원 메쉬 부호화 방법 및 장치
WO2016054975A1 (zh) 预测块的划分方法、编码设备和解码设备
JP2015511786A5 (zh)
CN103686139A (zh) 帧图像转换方法、帧视频转换方法及装置
CN104918032A (zh) 简化基于深度的块分割的方法
WO2015101251A1 (zh) 非对称运动分割方式编码的方法和编码设备
US9900620B2 (en) Apparatus and method for coding/decoding multi-view image
CN106464882B (zh) 用于通过视频编码器处理视频帧的技术
WO2016050120A1 (en) Method of lookup table size reduction for depth modelling mode in depth coding
US20220284633A1 (en) Method and apparatus for constructing a 3d geometry
WO2022126326A1 (zh) 点云编解码方法、编码器、解码器以及计算机存储介质
CN103533361B (zh) 多视点视差矢量的确定方法、编码设备及解码设备
CN104768014B (zh) 编解码方法

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: 14877333

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: 14877333

Country of ref document: EP

Kind code of ref document: A1