Method and apparatus for generating virtual avatar
AssigneeBeijing Baidu Netcom Science Technology Co., Ltd.
InventorsXiang Long, Xin Li, Henan Zhang, Hao Sun
Life of this patent
Estimating the term…
Abstract
A method and apparatus for generating a virtual avatar are provided. The method may include: acquiring a first avatar, and determining an expression parameter of the first avatar, where the expression parameter of the first avatar including an expression parameter of at least one of five sense organs; and determining, based on the expression parameter of at least one of the five sense organs, a target virtual avatar that is associated with an attribute of the first avatar and has an expression of the first avatar.
Claims16
A method for generating a virtual avatar, the method comprising:
- acquiring a first avatar, and determining an expression parameter of the first avatar, the expression parameter of the first avatar comprising an expression parameter of at least one of five sense organs;
- and
- determining, based on the expression parameter of at least one of the five sense organs, a target virtual avatar that is associated with an attribute of the first avatar and has an expression of the first avatar;
- and
- generating an expressionless reference virtual avatar associated with the attribute of the first avatar,
- wherein the determining, based on the expression parameter of at least one of the five sense organs, the target virtual avatar that is associated with the attribute of the first avatar and has the expression of the first avatar comprises:
- inputting the expressionless reference virtual avatar and the expression parameter of at least one of the five sense organs into a pre-trained deep neural network to obtain a first virtual avatar that is associated with the attribute of the first avatar and has the expression of the first avatar;
- and
- generating, based on the first virtual avatar, the target virtual avatar,
- wherein the expression parameter of the first avatar further comprises a head angle, and wherein the generating, based on the first virtual avatar, the target virtual avatar comprises:
- inputting the first virtual avatar and the head angle into a pre-trained generative adversarial network to obtain the target virtual avatar that is outputted from the pre-trained generative adversarial network, and is associated with the attribute of the first avatar and has the expression and head angle of the first avatar.
An electronic device, comprising:
- one or more processors;
- and
- a storage apparatus for storing one or more programs, the one or more programs, when executed by the one or more processors, causing the one or more processors to perform operations comprising:
- acquiring a first avatar, and determining an expression parameter of the first avatar, the expression parameter of the first avatar comprising an expression parameter of at least one of five sense organs;
- and
- determining, based on the expression parameter of at least one of the five sense organs, a target virtual avatar that is associated with an attribute of the first avatar and has an expression of the first avatar;
- and
- generating an expressionless reference virtual avatar associated with the attribute of the first avatar,
- wherein the determining, based on the expression parameter of at least one of the five sense organs, the target virtual avatar that is associated with the attribute of the first avatar and has the expression of the first avatar comprises:
- inputting the expressionless reference virtual avatar and the expression parameter of at least one of the five sense organs into a pre-trained deep neural network to obtain a first virtual avatar that is associated with the attribute of the first avatar and has the expression of the first avatar;
- and
- generating, based on the first virtual avatar, the target virtual avatar,
- wherein the expression parameter of the first avatar further comprises a head angle, and wherein the generating, based on the first virtual avatar, the target virtual avatar comprises:
- inputting the first virtual avatar and the head angle into a pre-trained generative adversarial network to obtain the target virtual avatar that is outputted from the pre-trained generative adversarial network, and is associated with the attribute of the first avatar and has the expression and head angle of the first avatar.
A non-transitory computer readable storage medium, storing a computer program thereon, the computer program, when executed by a processor, causing the processor to perform operations comprising:
- acquiring a first avatar, and determining an expression parameter of the first avatar, the expression parameter of the first avatar comprising an expression parameter of at least one of five sense organs;
- and
- determining, based on the expression parameter of at least one of the five sense organs, a target virtual avatar that is associated with an attribute of the first avatar and has an expression of the first avatar;
- and
- generating an expressionless reference virtual avatar associated with the attribute of the first avatar,
- wherein the determining, based on the expression parameter of at least one of the five sense organs, the target virtual avatar that is associated with the attribute of the first avatar and has the expression of the first avatar comprises:
- inputting the expressionless reference virtual avatar and the expression parameter of at least one of the five sense organs into a pre-trained deep neural network to obtain a first virtual avatar that is associated with the attribute of the first avatar and has the expression of the first avatar;
- and
- generating, based on the first virtual avatar, the target virtual avatar,
- wherein the expression parameter of the first avatar further comprises a head angle, and wherein the generating, based on the first virtual avatar, the target virtual avatar comprises:
- inputting the first virtual avatar and the head angle into a pre-trained generative adversarial network to obtain the target virtual avatar that is outputted from the pre-trained generative adversarial network, and is associated with the attribute of the first avatar and has the expression and head angle of the first avatar.
Dependencies are inferred from the claim text.