Decoupled mode inference and prediction
AssigneeInterDigital VC Holdings, Inc.
InventorsFranck Galpin, Antoine Robert, Fabrice Urban
Life of this patent
Estimating the term…
Abstract
A method of predicting or inferring a piece of information for encoding or decoding uses a predictor from a spatial neighbor, or from a reference picture, instead of predicting or inferring the information, such as a flag for a current block, from a motion predictor block. This procedure can be used in several ways, or used in various motion vector prediction modes, such as Frame Rate Up Conversion or various variations of Advanced Motion Prediction. Other embodiments provide other methods and combinations to derive or infer motion information to increase coding efficiency.
Claims12
A method of decoding video data, comprising:
- obtaining a first piece of information from a video bitstream, wherein the first piece of information is associated with a current video block and includes a combination of an actual value of a parameter and a prediction value associated with the first piece of information;
- determining the prediction value associated with the first piece of information based on one or more neighboring video blocks of the current video block or based on contextual information associated with the current video block;
- performing an inverse combining function on the first piece of information and the prediction value to derive a second piece of information, wherein the inverse combining function is inverse to a combining function used by a video encoding device to combine the actual value of the parameter and the prediction value;
- and
- decoding the current video block based on the second piece of information.
A video decoding device, comprising:
- a processor configured to:
- obtain a first piece of information from a video bitstream, wherein the first piece of information is associated with a current video block and includes a combination of an actual value of a parameter and a prediction value associated with the first piece of information;
- determine the prediction value associated with the first piece of information based on one or more neighboring video blocks of the current video block or based on contextual information associated with the current video block;
- perform an inverse combining function on the first piece of information and the prediction value to derive a second piece of information, wherein the inverse combining function is inverse to a combining function used by a video encoding device to combine the actual value of the parameter and the prediction value;
- and
- decode the current video block based on the second piece of information.
A method of encoding video data, comprising:
- obtaining an actual value of a parameter associated with a current video block;
- obtaining a prediction value associated with the parameter based on one or more neighboring video blocks of the current video block or based on contextual information associated with the current video block;
- combining the actual value of the parameter and the prediction value associated with the parameter into a piece of information that indicates the value of an illumination flag or a motion vector associated with the current video block, wherein the combination is derived by at least applying an XOR operator to the actual value of the parameter and the prediction value;
- and
- encoding the piece of information.
A video encoding device, comprising:
- a processor configured to:
- obtain an actual value of a parameter associated with a current video block;
- obtain a prediction value associated with the parameter based on one or more neighboring video blocks of the current video block or based on contextual information associated with the current video block;
- combine the actual value of the parameter and the prediction value associated with the parameter into a piece of information that indicates the value of an illumination flag or a motion vector associated with the current video block, wherein the combination is derived by at least applying an XOR operator to the actual value of the parameter and the prediction value;
- and
- encode the piece of information.
Dependencies are inferred from the claim text.