Neural network based filtering process for multiple color components in video coding
AssigneeQUALCOMM Incorporated
InventorsHongtao Wang, Samuel James Eadie, Muhammed Zeyd Coban, Marta Karczewicz
Life of this patent
Estimating the term…
Abstract
A method of processing video data includes receiving a syntax element that defines a filtering mode for a neural network (NN) model for both a first color component and a second color component, applying an instance of the NN model, in the defined filtering mode, to a first block of the first color component to generate a first filtered block, and storing the first filtered block for a coding unit (CU).
Claims19
A method of processing video data, the method comprising:
- receiving a syntax element that defines a filtering mode for a neural network (NN) model for both a first color component and a second color component, wherein the first color component is a first chroma component, and the second color component is a second chroma component;
- applying a single instance of the NN model, in the defined filtering mode, to a first block of the first color component to generate a first filtered block and to a second block of the second color component to generate a second filtered block, wherein applying the single instance of the NN model includes applying a single filtering process jointly to the first block and the second block;
- outputting the first filtered block and the second filtered block after applying the single filtering process jointly to the first block and the second block;
- and
- storing the first filtered block and the second filtered block for a coding unit (CU).
A device for processing video data, the device comprising:
- memory configured to store the video data;
- and
- one or more processors implement in circuitry, coupled to the memory, and configured to:
- receive a syntax element that defines a filtering mode for a neural network (NN) model for both a first color component and a second color component, wherein the first color component is a first chroma component, and the second color component is a second chroma component;
- apply a single instance of the NN model, in the defined filtering mode, to a first block of the first color component to generate a first filtered block and to a second block of the second color component to generate a second filtered block, wherein to apply the single instance of the NN model, the one or more processors are configured to apply a single filtering process jointly to the first block and the second block;
- output the first filtered block and the second filtered block after applying the single filtering process jointly to the first block and the second block;
- and
- store, in the memory, the first filtered block and the second filtered block for a coding unit (CU).
A non-transitory computer-readable storage medium storing instructions thereon that when executed cause one or more processors to:
- receive a syntax element that defines a filtering mode for a neural network (NN) model for both a first color component and a second color component, wherein the first color component is a first chroma component, and the second color component is a second chroma component;
- apply a single instance of the NN model, in the defined filtering mode, to a first block of the first color component to generate a first filtered block and to a second block of the second color component to generate a second filtered block, wherein the instructions that cause the one or more processors to apply the single instance of the NN model comprise instructions that cause the one or more processors to apply a single filtering process jointly to the first block and the second block;
- output the first filtered block and the second filtered block after applying the single filtering process jointly to the first block and the second block;
- and
- store the first filtered block and the second filtered block for a coding unit (CU).
Dependencies are inferred from the claim text.