Back to results
US 12,034,980 B2Granted
Video coding with embedded motion
AssigneeXylon LLC
InventorDonald Martin Monro
Life of this patent
Estimating the term…
Abstract
Implementations related to coding and/or decoding image data employing video coding with embedded motion are disclosed.
Claims18
A method, comprising:
- receiving image data comprising a sequence of one or more image frames, wherein the one or more image frames includes a first image frame;
- applying a motion data estimate to the first image frame to generate a second image frame, wherein the motion data estimate is based on a reconstruction image;
- subtracting the second image frame from the first image frame to generate a difference image frame;
- performing a wavelet transform on the first image frame and the difference image frame to generate a wavelet-transformed first image frame and a wavelet-transformed difference image frame;
- identifying a first basis function plus a residual that represents at least, in part, a first portion of the wavelet-transformed first image frame;
- identifying a second portion of the wavelet-transformed difference image frame that is represented at least by the first basis function;
- generating the reconstruction image using a sum of the second image frame and an inverse wavelength transform applied to the residual;
- and
- associating the motion data estimate with the first basis function;
- wherein the first portion includes two or more pixels of the first image frame,
- wherein the second portion includes two or more pixels of the wavelet-transformed difference image frame, and
- wherein the first basis function is selected from a dictionary of basis functions.
9 more elements
An apparatus, comprising:
- an encoder comprising a processor and a memory, the encoder adapted to:
- receive image data comprising a sequence of one or more image frames, wherein the one or more image frames includes a first image frame;
- apply a motion data estimate to the first image frame to generate a second image frame, wherein the motion data estimate is based on a reconstruction image;
- subtract the second image frame from the first image frame to generate a difference image frame;
- perform a wavelet transform on the first image frame and the difference image frame to generate a wavelet-transformed first image frame and a wavelet-transformed difference image frame;
- identify a first basis function plus a residual that represents at least, in part, a first portion of the wavelet-transformed first image frame;
- identify a second portion of the wavelet-transformed difference image frame that is represented at least by the first basis function;
- generate the reconstruction image using a sum of the second image frame and an inverse wavelength transform applied to the residual;
- and
- associate the motion data estimate with the first basis function;
- wherein the first portion includes two or more pixels of the first image frame,
- wherein the second portion includes two or more pixels of the wavelet-transformed difference image frame, and
- wherein the first basis function is selected from a dictionary of basis functions.
10 more elements
An apparatus, comprising:
- a decoder comprising a processor and a memory, the decoder adapted to:
- receive a bitstream, the bitstream representing a first image frame and a second image frame in a sequence of image frames, the bitstream including motion data estimate associated with a first basis function;
- extract the first basis function from the bitstream, the first basis function representing a first portion of the first image frame;
- extract the motion data estimate from the bitstream, wherein the motion data estimate indicates a displacement between the first portion of the first image frame and a second portion of the second image frame, wherein the first basis function represents the second portion of the second image frame;
- and
- generate the first image frame and the second image frame based on the first basis function and the displacement,
- wherein the first basis function represents two or more pixels of the first image frame, and wherein the first basis function is included in a dictionary of basis functions.
4 more elements
Dependencies are inferred from the claim text.