Prediction apparatus, prediction method and program
AssigneeNippon Telegraph and Telephone Corporation
InventorsTakehito MIYAZAWA, Yukihiro BANDO, Atsushi SHIMIZU
Life of this patent
Estimating the term…
Abstract
A prediction device for predicting a motion vector of a coding target block in coding of a composite image formed by combining multiple viewpoint images includes: a storage unit that stores a first predicted motion vector indicating a predicted motion vector candidate; and an update unit that if a predetermined condition is met, generates a second predicted motion vector that is a vector indicating a pixel in an adjacent viewpoint, the pixel being located at a position that is same as a position of the coding target block, and stores the second predicted motion vector the storage unit.
Claims7
A prediction device for predicting a motion vector of a coding target block in coding of a composite image formed by combining multiple viewpoint images, the prediction device comprising:
- a storage unit that stores a first predicted motion vector indicating a predicted motion vector candidate;
- a processor;
- and
- a storage medium having computer program instructions stored thereon, when executed by the processor, perform to:
- if a predetermined condition is met, generates a second predicted motion vector that is a vector indicating a pixel in an adjacent viewpoint, the pixel being located at a position that is same as a position of the coding target block, and stores the second predicted motion vector in the storage unit.
A prediction device for predicting a motion vector of a coding target block in coding of a composite image formed by combining multiple viewpoint images, the prediction device comprising:
- a storage unit that stores a first predicted motion vector indicating a predicted motion vector candidate;
- a processor;
- and
- a storage medium having computer program instructions stored thereon, when executed by the processor, perform to:
- if there are a plurality of the first predicted motion vectors stored in the storage unit, generates a second predicted motion vector that is a vector indicating a pixel in an adjacent viewpoint, the pixel being located at a position that is same as a position of the coding target block, and overwrites one of the first predicted motion vectors, whichever is lower in priority, with the second predicted motion vector.
A prediction method for predicting a motion vector of a coding target block in coding of a composite image formed by combining multiple viewpoint images, the prediction method comprising:
- a storage step of storing a first predicted motion vector indicating a predicted motion vector candidate;
- and
- an update step of if a predetermined condition is met, generating a second predicted motion vector that is a vector indicating a pixel in an adjacent viewpoint, the pixel being located at a position that is same as a position of the coding target block, and storing the second predicted motion vector.
Dependencies are inferred from the claim text.