Video Encoding and Decoding with Improved Error Resilience

Patent metadata

Field	Value
Publication number	US-2018242001-A1
Application number	US-201815962940-A
Country	US
Kind code	A1
Filing date	Apr 25, 2018
Priority date	Jan 12, 2011
Publication date	Aug 23, 2018
Grant date	—

Abstract

An encoding method comprises obtaining a target number of motion information predictors to be used for an image portion to encode and generating a set of motion information predictors using the obtained target number. The set is generated by: obtaining a first set of motion information predictors each associated with an image portion having a predetermined spatial and/or temporal relationship with the image portion being encoded; modifying the first set of motion information predictors by removing duplicate motion information predictors to obtain a reduced set of motion information predictors containing a first number of motion information predictors, each motion information predictor of the reduced set being different from any other motion information predictor of the reduced set; comparing the first number of motion information predictors to the target number obtained, and if the first number is lower than the target number, obtaining an additional motion information predictor and adding it to the reduced set of motion information predictors.

First claim

1 . A method of decoding a bitstream comprising a sequence of digital images encoded using inter prediction, the method comprising the steps of:
obtaining one or more potential motion information predictors including a motion information predictor from an image block situated above an image block being decoded, if available; and generating a set of motion information predictors including at least one motion information predictor from among the obtained one or more potential motion information predictors, each potential motion information predictor being associated with an image block having a predetermined spatial relationship with the image block being decoded, wherein if the obtained one or more potential motion information predictors include one or more duplicates, the set of motion information predictors is generated without at least one of the duplicates and one or more additional motion information predictors are added until a number of motion information predictors in the set to be generated is a target number.

2 . A device for decoding a bitstream comprising a sequence of digital images, encoded using inter prediction, the device comprising:
an obtainer configured to obtain one or more potential motion information predictors including a motion information predictor from an image block situated above an image block being decoded, if available; and a generator configured to generate a set of motion information predictors including at least one motion information predictor from among the obtained one or more potential motion information predictors, each potential motion information predictor being associated with an image block having a predetermined spatial relationship with the image block being decoded, the generator being further configured so that if the obtained one or more potential motion information predictors include one or more duplicates, the set of motion information predictors is generated without at least one of the duplicates and one or more additional motion information predictor are added until a number of motion information predictors in the set to be generated is a target number.

3 . A non-transitory computer readable carrier medium comprising processor executable code for performing a method of decoding a bitstream comprising a sequence of digital images encoded using inter prediction, wherein execution of the processor executable code by one or more processors causes the one or more processors to obtain one or more potential motion information predictors including a motion information predictor from an image block situated above an image block being decoded, if available; and generate a set of motion information predictors including at least one motion information predictor from among the obtained one or more potential motion information predictors, each potential motion information predictor being associated with an image block having a predetermined spatial relationship with the image block being decoded,
wherein if the obtained one or more potential motion information predictors include one or more duplicates, the set of motion information predictors is generated without at least one of the duplicates and one or more additional motion information predictors are added until a number of motion information predictors in the set to be generated is a target number.

Assignees

• Canon Kk

Inventors

• Gisquet Christophe
• Laroche Guillaume

Classifications

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title

• H04N19/103

  Selection of coding mode or of prediction mode · CPC title

• H04N19/65

  using error resilience · CPC title

• H04N19/147

  according to rate distortion criteria (rate-distortion as a criterion for motion estimation H04N19/567) · CPC title

• H04N19/192

  the adaptation method, adaptation tool or adaptation type being iterative or recursive · CPC title