Method and device for generating an animation graph

Patent metadata

Field	Value
Publication number	US-12499599-B1
Application number	US-202217949403-A
Country	US
Kind code	B1
Filing date	Sep 21, 2022
Priority date	Sep 23, 2021
Publication date	Dec 16, 2025
Grant date	Dec 16, 2025

Abstract

In some implementations, the method includes: obtaining a plurality of animations; determining initial and end motion states for each of the plurality of animations; generating an animation graph including nodes for each of the plurality of animations by connecting, with a directional edge, a first node with an end motion state to a second node with an initial motion state that matches the end motion state of the first node; generating a transitional animation that is not included among the plurality of animations from an initial reference motion state to a target motion state that corresponds to a path that traverses the animation graph from a third node associated with the initial reference motion state to a fourth node associated with the target motion state; and updating the animation graph by removing one or more nodes from the animation graph based at least in part on the transitional animation.

First claim

What is claimed is:

1 . A method comprising:
at a computing system including non-transitory memory and one or more processors, wherein the computing system is communicatively coupled to a display device and one or more input devices: obtaining a plurality of pre-existing animations; determining an initial motion state and an end motion state for each of the plurality of pre-existing animations; generating an animation graph including nodes for each of the plurality of pre-existing animations by connecting, with a directional edge, a first node with an end motion state to a second node with an initial motion state that matches the end motion state of the first node; generating a transitional animation that is not included among the plurality of pre-existing animations from an initial reference motion state to a target motion state that corresponds to a path that traverses the animation graph from a third node associated with the initial reference motion state to a fourth node associated with the target motion state; updating the animation graph by:
inserting a machine learning (ML) sub-system into the animation graph to generate a non-transitional output motion or a transitional output motion; or
replacing a non-transitional node within the animation graph with a ML sub-system to generate a non-transitional output motion or a transitional output motion.

2 . The method of claim 1 , wherein the plurality of pre-existing animations is associated with a virtual agent (VA).

3 . The method of claim 2 , further comprising:
presenting, via the display device, the VA performing the transitional animation.

4 . The method of claim 3 , wherein the display device corresponds to a transparent lens assembly, and wherein the presentation of the VA performing the transitional animation is projected onto the transparent lens assembly.

5 . The method of claim 3 , wherein the display device corresponds to a near-eye system, and wherein presenting the VA performing the transitional animation includes compositing the presentation of the VA performing the transitional animation with one or more images of a physical environment captured by an exterior-facing image sensor.

6 . The method of claim 2 , further comprising:
obtaining an initial reference motion state and a target motion state; determining one or more candidate paths through the animation graph based on the initial reference motion state and the target motion state; selecting a respective candidate path from among the one or more candidate paths; generating a motion output based on the respective candidate path; and presenting, via the display device, the VA performing the motion output.

7 . The method of claim 6 , wherein the motion output corresponds to a sequence of two or more actions or motions states.

8 . The method of claim 2 , wherein the VA corresponds to one of a humanoid, an animal, a robot, or a vehicle.

9 . The method of claim 1 , further comprising:
adding one or more nodes to the animation graph to increase a number of connected nodes and/or to connect unconnected nodes.

10 . The method of claim 1 , further comprising:
generating a motion graph by encoding the plurality of pre-existing animations according to one or more embedding space parameters, wherein the animation graph is generated by connecting, with the directional edge, the first node with the end motion state to the second node with the initial motion state that matches the end motion state of the first node based on the motion graph.

11 . The method of claim 10 , wherein the one or more embedding space parameters correspond to at least one of: intended inputs associated with the motion graph, intended outputs associated with the motion graph, labels associated with the plurality of pre-existing animations, and metadata associated with the plurality of pre-existing animations.

12 . The method of claim 1 , wherein the ML sub-system corresponds to one of a neural network (NN), a convolutional neural network (CNN), a recurrent neural network (RNN), a deep neural network (DNN), a support-vector machine (SVM), a relevance vector machine (RVM), or a random forest algorithm.

13 . The method of claim 1 , further comprising:
training the ML sub-system by reducing an error value of random walks through the animation graph.

14 . A device comprising:
one or more processors; a non-transitory memory; an interface for communicating with a display device and one or more input devices; and one or more programs stored in the non-transitory memory, which, when executed by the one or more processors, cause the device to:
obtain a plurality of pre-existing animations;
determine an initial motion state and an end motion state for each of the plurality of pre-existing animations;
generate an animation graph including nodes for each of the plurality of pre-existing animations by connecting, with a directional edge, a first node with an end motion state to a second node with an initial motion state that matches the end motion state of the first node;
generate a transitional animation that is not included among the plurality of pre-existing animations from an initial reference motion state to a target motion state that corresponds to a path that traverses the animation graph from a third node associated with the initial reference motion state to a fourth node associated with the target motion state; and
update the animation graph by inserting a machine learning (ML) sub-system into the animation graph to generate a non-transitional output motion or a transitional output motion; or
replacing a non-transitional node within the animation graph with a ML sub-system to generate a non-transitional output motion or a transitional output motion.

15 . The device of claim 14 , wherein the plurality of pre-existing animations is associated with a virtual agent (VA).

16 . The device of claim 15 , wherein the one or more programs further cause the device to:
obtain an initial reference motion state and a target motion state; determine one or more candidate paths through the animation graph based on the initial reference motion state and the target motion state; select a respective candidate path from among the one or more candidate paths; generate a motion output based on the respective candidate path; and present, via the display device, the VA performing the motion output.

17 . The device of claim 16 , wherein the motion output corresponds to a sequence of two or more actions or motions states.

18 . The device of claim 14 , wherein the one or more programs further cause the device to:
generate a motion graph by encoding the plurality of pre-existing animations according to one or more embedding space parameters, wherein the animation graph is generated by connecting, with the directional edge, the first node with the end motion state to the second node with the initial motion state that matches the end motion state of the first node based on the motion graph.

19 . A non-transitory memory storing one or more programs, which, when executed by one or more processors of a device with an interface for communicating with a display device and one or more input devices, cause the device to:
obtain a plurality of pre-existing animations; determine an initial motion state and an end motion state for each of the plurality of pre-existing animations; generate an animation graph including nodes for each of the plurality of pre-existing animations by connecting, with a directional edge,

Assignees

• Apple Inc

Inventors

• Manglik Aashi
• Sivapurapu Siva Chandra Mouli
• Budhram Shaun D
• Ahn Edward
• Drummond Mark E
• Gamboa Noah
• Mahasseni Behrooz

Classifications

• G06T13/40 Primary

  of characters, e.g. humans, animals or virtual beings · CPC title

• G06T13/00 Primary

  Animation · CPC title