Video encoding method and apparatus, electronic device and storage medium
AssigneesDouyin Vision Co., Ltd., Lemon Inc.
InventorsCheng Chen, Shu Shi
Life of this patent
Estimating the term…
Abstract
The embodiment of the disclosure provides a video encoding method and apparatus, an electronic device and a storage medium. The method includes the steps of determining a video segment included in a target video according to a duration corresponding to a first picture group parameter; determining a target feature corresponding to video encoding data of a previous video segment for a current video segment; inputting the target feature into a target decision model, and determining a picture group parameter corresponding to the current video segment by using the target decision model; and encoding the current video segment based on the picture group parameter corresponding to the current video segment.
Claims20
A video encoding method, comprising:
- determining a video segment comprised in a target video according to a duration corresponding to a first picture group parameter;
- determining a target feature corresponding to video encoding data of a previous video segment for a current video segment;
- inputting the target feature into a target decision model, and determining a picture group parameter corresponding to the current video segment by using the target decision model, wherein the target decision model represents a decision model associated with a picture group parameter corresponding to the previous video segment in at least two decision models, and the at least two decision models are trained based on a sample set generated by encoding a historical video based on at least two picture group parameters;
- encoding the current video segment based on the picture group parameter corresponding to the current video segment.
An electronic device, comprising:
- one or more processors;
- a storage device, configured to store one or more programs;
- the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the acts comprising:
- determining a video segment comprised in a target video according to a duration corresponding to a first picture group parameter;
- determining a target feature corresponding to video encoding data of a previous video segment for a current video segment;
- inputting the target feature into a target decision model, and determining a picture group parameter corresponding to the current video segment by using the target decision model, wherein the target decision model represents a decision model associated with a picture group parameter corresponding to the previous video segment in at least two decision models, and the at least two decision models are trained based on a sample set generated by encoding a historical video based on at least two picture group parameters;
- encoding the current video segment based on the picture group parameter corresponding to the current video segment.
A non-transitory storage medium comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a computer processor, are configured to implement the acts comprising:
- determining a video segment comprised in a target video according to a duration corresponding to a first picture group parameter;
- determining a target feature corresponding to video encoding data of a previous video segment for a current video segment;
- inputting the target feature into a target decision model, and determining a picture group parameter corresponding to the current video segment by using the target decision model, wherein the target decision model represents a decision model associated with a picture group parameter corresponding to the previous video segment in at least two decision models, and the at least two decision models are trained based on a sample set generated by encoding a historical video based on at least two picture group parameters;
- encoding the current video segment based on the picture group parameter corresponding to the current video segment.
Dependencies are inferred from the claim text.