Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
  • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
  • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
  • H04N19/51—Motion estimation or motion compensation
  • H04N19/513—Processing of motion vectors
  • H04N19/517—Processing of motion vectors by encoding
  • H04N19/52—Processing of motion vectors by encoding by predictive encoding
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
  • H04N19/597—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding specially adapted for multi-view video sequence encoding

Definitions

• the invention relates generally to video processing.
• the present invention relates to methods and apparatuses of motion prediction in 3D video coding.
• Three dimensional video coding is a popular topic in recent year.
• ISO/IEC moving picture experts group MPEG now is working on a new 3DVC standard, which aims to significantly improve the coding performance for 3D video.
• There are several tools which can exploit the correlations between neighboring views such as inter-view texture prediction and inter-view motion prediction.
• the current 3DVC system is based on both H.264/AVC, i.e. ATM, and high efficiency video coding (HEVC), i.e. HTM.
• HEVC high efficiency video coding
• the inter- view motion prediction in the current 3DVC system can be further improved.
• a close loop derivation concept of the inter-view motion vector is described, which uses both the forward disparity from the current view to the reference view and the backward disparity from the reference view to the current view to derive the inter- view motion information.
• an inter-view motion derivation which uses the coded disparity vectors (DV) from spatial neighbors is described.
• Fig. 1 is a diagram illustrating the process of inter-view motion derivation in HTM
• Fig. 2 is a diagram illustrating an example of the close loop inter-view motion derivation
• Fig. 3 is a diagram illustrating the disparity derivation method for the current block by using its spatial neighbors.
• a 3D sequence consists of several sequences with the same scene but slightly different view angles.
• one of the views is chosen as base view which is coded independently without referencing any other views.
• the views other than base view call dependent views which are coded dependently with referencing other views.
• the view being coded as current view and the view it references as reference view.
• the base view is used as reference view.
• both texture and depth map are transmitted.
• the motion information from the reference views can be used as prediction of the current view.
• the block type, reference index, prediction direction and motion vector of the current block can be predicted by that from the corresponding block in the reference view.
• the corresponding block in the reference view is generally found by using the disparity between the two blocks.
• Fig. 1 shows an example which is in the current HTM version 0.3.
• the disparity of the central point in the current block is used to derive the corresponding block in the reference view.
• the motion vector of the block in the reference view is used for the current block.
• the current method of inter-view motion vector prediction can be further improved.
• a close loop derivation concept is used to further improve the derivation of the inter-view motion derivation. This concept is to use more information from in the reference view to help derive the motion information for the current view. In specific, both the forward disparity from the current view to the reference view and the backward disparity from the reference view to the current view are used to derive the inter-view motion information.
• Fig. 2 shows an example of the close loop method.
• the base view is reference view.
• the dependent is the current view.
• the central point of the current block (time n) in dependent view and its forward disparity, disaprityO, is used to find its inter-view corresponding point in the base view.
• the motion vector and reference index of the block which contains this point in the based view (time n) is used to find the corresponding position of this block in the temporal reference picture (time n-1).
• the temporal corresponding block of the current block in dependent view can be found by using forward disparity, disaprityl.
• the mvl can be derived by using mvO, disparityO and disparity 1 as
• mvl mvO + disparityO + disparity 1.
• mvO and mvl can be used individually or together. When used together, the two motion vectors are adaptively switched explicitly, e.g. sending signal flag as indication in block level or region level, or implicitly, e.g. using the difference between disparityO and disparityl to judge the precision of mvl.
• the mvO and mvl can also be switched according the existence of the picture n-1 in base view. In specific, if the picture is in the decoded frame buffer when coding the current picture (time n) in the dependent view, mvl is used. Otherwise, mvO is used.
• DisaprityO can certainly be derived using the disparity of the central point or the average disparity of any subset of the pixels in the current block.
• the disparity value can be derived using the corresponding depth map (coded depth or estimated depth) and camera parameters, as well as the coded disparity from the spatial or temporal neighbors of the current block. So does disparityl.
• the derived motion information can both be directly used as the motion information of the current block and be used as the prediction for the motion information of the current block.
• the backward disparity is not limited to the disparityl shown in Fig. 2. It can also be the disparity of the corresponding block in picture n in reference view.
• the forward disparity and the backward disparity can be used together to decide if the block found by forward disparity is good enough, e.g. judging by the difference of the two disparities.
• the disparity vectors from the spatial neighbors are used for the current block together with the disparity derived by depth to improve the current inter- view motion derivation as shown in Fig. 3.
• the central point of the current block in the dependent view and its disparity are used to find the corresponding point in the base view. After that, the motion of the block which includes the corresponding point in the base view is used as the inter-view candidate of the current block.
• the disparity can be derived from both the neighboring blocks and the depth value of the central point. In specific, if one of the neighboring blocks has disparity vector (DV), e.g. DV A in Figure 2, the DV is used as the disparity.
• DV disparity vector
• the depth-based disparity is used, which is derived using the depth value of the central point and camera parameters.
• DVs from spatial neighbours can partially avoid error propagation in case that the depth value of the central point is not available, e.g. the depth image is lost.
• the depth-based disparity can be totally replaced by the coded DVs from spatial neighbours to reduce the computational complexity and avoid the error propagation. In this case, if there is no DV in the spatial neighbours, no inter- view candidate is used for the current block.
• an embodiment of the present invention can be a circuit integrated into a video compression chip or program codes integrated into video compression software to perform the processing described herein.
• An embodiment of the present invention may also be program codes to be executed on a Digital Signal Processor (DSP) to perform the processing described herein.
• DSP Digital Signal Processor
• the invention may also involve a number of functions to be performed by a computer processor, a digital signal processor, a microprocessor, or field programmable gate array (FPGA).
• processors can be configured to perform particular tasks according to the invention, by executing machine-readable software code or firmware code that defines the particular methods embodied by the invention.
• the software code or firmware codes may be developed in different programming languages and different format or style.
• the software code may also be compiled for different target platform.
• different code formats, styles and languages of software codes and other means of configuring code to perform the tasks in accordance with the invention will not depart from the spirit and scope of the invention.

Landscapes

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

Priority Applications (1)

Applications Claiming Priority (1)

Publications (1)

Family

ID=49482150

Family Applications (1)

Country Status (1)

Cited By (3)

Citations (3)

• 2012
  • 2012-04-27 WO PCT/CN2012/074807 patent/WO2013159326A1/fr not_active Ceased

Patent Citations (3)

Similar Documents

Legal Events