Moving picture encoding method and decoding method

• US 20050152452A1
• Filed: 07/11/2003
• Published: 07/14/2005
• Est. Priority Date: 07/15/2002
• Status: Active Grant

• Current Assignee: Maxell, Ltd.
• Original Assignee: Hitachi Consumer Electronics Company Limited (Hitachi, Ltd.)
• Inventors: Suzuki, Yoshinori
• Granted Patent: US 7,970,058 B2
• Time in Patent Office: Days
• US Class Current: 375/240.160
• CPC Class Codes: H04N 19/109 among a plurality of temporal predictive coding modes
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/44 Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
• CPC Class Codes: H04N 19/46 Embedding additional information in the video signal during the compression process H04N19/517, H04N19/68, H04N19/70 take precedence
• CPC Class Codes: H04N 19/51 Motion estimation or motion compensation
• CPC Class Codes: H04N 19/513 Processing of motion vectors
• CPC Class Codes: H04N 19/517 by encoding
• CPC Class Codes: H04N 19/573 Motion compensation with multiple frame prediction using two or more reference frames in a given prediction direction
• CPC Class Codes: H04N 19/577 Motion compensation with bidirectional frame interpolation, i.e. using B-pictures
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/70 characterised by syntax aspects related to video coding, e.g. related to compression standards
• CPC Class Codes: H05K 999/99 dummy group

Abstract

Conventionally there has been a case that the direct mode cannot be applied effectively depending on the block. With such being the case, information indicating whether a backward reference frame set by default can be utilized in the direct mode is provided to a decoder. A switching procedure to switch to a compensation method applicable when a collocated block has no forward motion vector for effective use, and the compensation method are also provided to the decoder. Thus, it is possible to clearly determine whether the reference frame can be used in the direct mode. Further, when the frame number has no time information, it is possible to effectively send information indicating the relationship between the reference frame and the current frame. Furthermore, the alternative mode and its switching procedure of the present invention make it possible to improve the prediction performance when the direct mode cannot be applied.

Claims

1. A moving picture decoding method, which receives information on motion vectors, and performs motion compensation using recorded reference images and the information on the motion vectors to synthesize a predicted image,

the motion compensation having multiple block modes including a mode without motion vector decoding,

said method comprising:

a step of selecting a prediction mode representing the direction(s) of prediction;

a step of selecting a frame(s) to be referred to in each direction of prediction in the prediction mode from among multiple candidate reference frames; and

a step of selecting motion vector information used in the prediction mode.

2. A moving picture decoding method according to claim 1, wherein said step of selecting a reference frame(s) is performed after said step of selecting a prediction mode.

3. A moving picture decoding method according to claim 1, wherein said step of selecting a prediction mode is performed based on whether blocks adjacent to a current block have a motion vector.

4. A moving picture decoding method according to claim 3, wherein when the adjacent blocks have bi-directional motion vectors along the time axis or when they have no motion vector, a bi-directional prediction mode is selected in said step of selecting a prediction mode, and when the adjacent blocks have one-way motion vectors along the time axis, a forward or backward prediction mode is selected in said step of selecting a prediction mode.

5. A moving picture decoding method according to claim 1, wherein said step of selecting a frame(s) to be referred to is a step of selecting one reference frame from multiple reference frames identified by index numbers, and when prediction in the selected direction is applied to the multiple adjacent blocks, a reference frame used for any one of the adjacent blocks is selected, when prediction in the selected direction is applied to only one of the multiple adjacent blocks, a reference frame corresponding to the index number used for the adjacent block is selected, or when the selected prediction mode is not applied to any of the adjacent blocks, a reference frame corresponding to index number 0 is selected.

6. A moving picture decoding method according to claim 5, wherein when prediction in the selected direction is applied to the multiple adjacent blocks, a reference frame having the smallest index number is selected from the reference frames applied.

7. A moving picture decoding method according to claim 3, wherein the adjacent blocks are three candidate blocks.

8. A moving picture decoding method according to claim 4, wherein the adjacent blocks are three candidate blocks.

9. A moving picture decoding method according to claim 1, wherein said step of selecting a motion vector(s) and said step of selecting a frame(s) to be referred to are performed on blocks different in size from each other.

10. A moving picture decoding method according to claim 1, wherein information for defining a prediction procedure performed when the mode without motion vector decoding is selected as a block mode is included in a header attached on a block basis.

11. A moving picture decoding method according to claim 10, wherein the information for defining a prediction procedure is information for selecting a prediction method for each block from multiple prediction methods for the mode without motion vector decoding.

20. A moving picture decoding method according to claim 10, wherein when multiple methods are defined as the prediction procedure for the mode without motion vector encoding, the information for defining a prediction procedure is information for selecting one method for each block.

12. A moving picture encoding method for multiplexing information on a difference between an input image and a predicted image synthesized using motion compensation with information on a motion vector(s) determined by the motion compensation,

the motion compensation having multiple block modes including a mode without motion vector encoding,

said method comprising:

a step of selecting a prediction mode representing the direction(s) of prediction;

a step of selecting a frame(s) to be referred to in each direction of prediction in the prediction mode from among multiple candidate reference frames; and

a step of selecting motion vector information used in the prediction mode.

13. A moving picture encoding method according to claim 12, wherein said step of selecting a prediction mode is performed based on whether blocks adjacent to a current block have a motion vector.

14. A moving picture encoding method according to claim 13, wherein when the adjacent blocks have bi-directional motion vectors along the time axis or when they have no motion vector, a bi-directional prediction mode is selected in said step of selecting a prediction mode, and when the adjacent blocks have one-way motion vectors along the time axis, a forward or backward prediction mode is selected in said step of selecting a prediction mode.

15. A moving picture encoding method according to claim 12, wherein said step of selecting a frame(s) to be referred to is a step of selecting one reference frame from multiple reference frames identified by index numbers, and when prediction in the selected direction is applied to the multiple adjacent blocks, a reference frame used for any one of the adjacent blocks is selected, when prediction in the selected direction is applied to only one of the multiple adjacent blocks, a reference frame corresponding to the index number used for the adjacent block is selected, or when the selected prediction mode is not applied to any of the adjacent blocks, a reference frame corresponding to index number 0 is selected.

16. A moving picture encoding method according to claim 15, wherein when prediction in the selected direction is applied to the multiple adjacent blocks, a reference frame having the smallest index number is selected from the reference frames applied.

17. A moving picture encoding method according to claim 13, wherein the adjacent blocks are three candidate blocks.

18. A moving picture encoding method according to claim 12, wherein said step of selecting a motion vector(s) and said step of selecting a frame(s) to be referred to are performed on blocks different in size from each other.

19. A moving picture encoding method according to claim 12, wherein information for defining a prediction procedure performed when the mode without motion vector encoding is selected as a block mode is included and multiplexed in a header attached on a block basis.

21. A moving picture decoding method, which receives information on motion vectors, and performs motion compensation using recorded reference images and the information on the motion vectors to synthesize a predicted image, the motion compensation having multiple block modes including a mode without motion vector decoding, wherein

a prediction procedure in the mode without motion vector decoding comprising:

a step of selecting a prediction mode representing the direction(s) of prediction using decoded information;

a step of selecting, using the decoded information, a frame(s) to be referred to in each direction of prediction in the prediction mode from among multiple candidate reference frames; and

a step of selecting motion vector information used in the prediction mode using the decoded information.

22. A moving picture encoding method for multiplexing information on a difference between an input image and a predicted image synthesized using motion compensation with information on a motion vector(s) determined by the motion compensation, the motion compensation having multiple block modes including a mode without motion vector encoding, wherein

a prediction procedure in the mode without motion vector encoding comprising:

a step of selecting a prediction mode representing the direction(s) of prediction using coded information;

a step of selecting, using the coded information, a frame(s) to be referred to in each direction of prediction in the prediction mode from among multiple candidate reference frames; and

a step of selecting motion vector information used in the prediction mode using the coded information.