Method and apparatus for effectively compressing motion vectors in multi-layer structure
AssigneeSamsung Electronics Co., Ltd.
InventorsHo-Jin Ha, Sang-Chang Cha, Woo-jin Han
Life of this patent
Estimating the term…
Abstract
A motion vector compression apparatus includes: a down-sampling module for down-sampling an original frame to have a size of a frame in each layer; a motion vector search module for obtaining a motion vector in which an error or a cost function is minimized with respect to the down-sampled frame; a reference vector generation module for generating a reference motion vector in a predetermined enhanced layer by means of a block of a lower layer corresponding to a predetermined block in the predetermined enhanced layer, and motion vectors in blocks around the block; and a motion difference module for calculating a difference between the obtained motion vector and the reference motion vector.
Claims2
A multi-layer video decoding method comprising:
- a) analyzing an inputted bit stream to extract texture information and motion information;
- b) analyzing the extracted motion information to calculate a reference motion vector with respect to a predetermined enhanced layer, adding a motion difference contained in the motion information and the calculated reference motion vector, thereby restoring a motion vector;
- c) performing an inverse-quantization for the texture information to output a transform coefficient;
- d) performing an inverse spatial transform to convert the transform coefficient into a transform coefficient of a spatial domain;
- and
- e) performing an inverse temporal filtering for the transform coefficient by means of the restored motion vector, thereby restoring frames constituting a video sequence,
- wherein step b) comprises:
- generating the reference motion vector in the predetermined enhanced layer by means of a block of a base layer corresponding to a predetermined block in the predetermined enhanced layer and motion vectors in blocks around the block;
- and adding the obtained reference motion vector and the motion vector difference, thereby generating a motion vector.
A video decoder supporting a motion vector of a multi-layer structure, the video decoder comprising:
- an entropy decoding module for analyzing an inputted bit stream to extract texture information and motion information;
- a motion vector restoration module for analyzing the extracted motion information to calculate a reference motion vector with respect to a predetermined enhanced layer, adding a motion difference contained in the motion information and the calculated reference motion vector, and thus restoring a motion vector;
- an inverse quantization module for performing an inverse-quantization for the texture information to output a transform coefficient;
- an inverse spatial transform module for performing an inverse spatial transform to convert the transform coefficient into a transform coefficient of a spatial domain;
- and
- an inverse temporal filtering module for performing an inverse temporal filtering for the transform coefficient by means of the restored motion vector, thereby restoring frames constituting a video sequence,
- wherein the motion vector restoration module comprises:
- a reference vector calculation module for generating the reference motion vector in the predetermined enhanced layer by means of a block of a lower layer corresponding to a predetermined block in the predetermined enhanced layer and motion vectors in blocks around the block:
- a filter module for providing a predetermined filter applied to an interpolation process for generating the reference motion vector:
- and
- a motion add module for adding the obtained reference motion vector and the motion vector difference, and thus generating a motion vector.
Dependencies are inferred from the claim text.