Facial gesture driven animation of non-facial features

Patent metadata

Field	Value
Publication number	US-9799133-B2
Application number	US-201414779535-A
Country	US
Kind code	B2
Filing date	Dec 23, 2014
Priority date	Dec 23, 2014
Publication date	Oct 24, 2017
Grant date	Oct 24, 2017

Abstract

Examples of systems and methods for non-facial animation in facial performance driven avatar system are generally described herein. A method for facial gesture driven body animation may include capturing a series of images of a face, and computing facial motion data for each of the images in the series of images. The method may include identifying an avatar body animation based on the facial motion data, and animating a body of an avatar using the avatar body animation.

First claim

What is claimed is:

1. A system for facial gesture driven body animation, the system comprising:
an image capture device to capture a series of images of a face;
a facial recognition circuit to compute facial motion data for each of the images in the series of images;
a processing circuit to identify an avatar non-facial body animation based on the facial motion data by determining movement of a landmark point on the face and identify the avatar non-facial body animation based on the movement of the landmark, wherein the movement of the landmark is a user defined movement of the landmark, the user defined movement selected from a list of available landmark movements; and
an animation circuit to animate a body of an avatar using the avatar non-facial body animation, wherein the body of the avatar includes a plurality of parts of the avatar connected to an avatar face.

2. The system of claim 1 , wherein to identify the avatar non-facial body animation, the processing circuit is to determine a lack of movement of the face in the series of images and identify the avatar non-facial body animation based on the lack of movement.

3. The system of claim 1 , wherein the avatar non-facial body animation is a user defined avatar body animation selected by the user and mapped to the facial motion data.

4. The system of claim 1 , wherein the processing circuit is to identify the avatar non-facial body animation after receiving a user indication to enter a body animation mode.

5. The system of claim 1 , wherein the animation circuit is to animate the avatar face using the facial motion data.

6. The system of claim 5 , wherein the processing circuit is to blend the avatar face and the body of the avatar into a blended avatar, and wherein the animation circuit is to animate the blended avatar.

7. The system of claim 5 , wherein to animate the body of the avatar, the animation circuit is to animate the body of the avatar using the avatar face animation.

8. The system of claim 5 , wherein the animation circuit is to animate the avatar face based on properties of the body animation.

9. The system of claim 5 , wherein the animation circuit is to animate the body of the avatar based on properties of the facial animation.

10. The system of claim 1 , wherein to animate the body of the avatar, the animation circuit is to animate a second part of the body relative to a first part of the body.

11. The system of claim 10 , wherein to animate the second part of the body relative to the first part of the body, the animation circuit is to use a first weighted value.

12. The system of claim 11 , wherein to animate the body of the avatar, the animation circuit is to animate a third part of the body relative to the second part of the body using a second weighted value, wherein the second weighted value is less than the first weighted value.

13. At least one non-transitory machine-readable medium including instructions for receiving information, which when executed by a machine, cause the machine to:
capture a series of images of a face;
compute facial motion data for each of the images in the series of images;
identify an avatar non-facial body animation based on the facial motion data by determining movement of a landmark point on the face and identify the avatar non-facial body animation based on the movement of the landmark wherein the movement of the landmark is a user defined movement of the landmark, the user defined movement selected from a list of available landmark movements; and
animate a body of an avatar using the avatar non-facial body animation, wherein the body of the avatar includes a plurality of parts of the avatar connected to an avatar face.

14. The at least one non-transitory machine-readable medium of claim 13 , further comprising operations to receive a user indication to enter a body animation mode, and wherein operations to identify the avatar non-facial body animation include operations to identify the avatar non-facial body animation after receiving the user indication.

15. The at least one non-transitory machine-readable medium of claim 13 , further comprising operations to animate the avatar face using the facial motion data.

16. The at least one non-transitory machine-readable medium of claim 15 , further comprising operations to blend the avatar face and the body of the avatar into a blended avatar, and animate the blended avatar.

17. The at least one non-transitory machine-readable medium of claim 13 , further comprising operations to animate a second part of the body relative to a first part of the body.

18. The at least one non-transitory machine-readable medium of claim 13 , wherein the operations to animate the second part of the body relative to the first part of the body include operations to use a first weighted value.

19. A method for facial gesture driven body animation, the method comprising:
capturing a series of images of a face;
computing facial motion data for each of the images in the series of images;
identifying an avatar non-facial body animation based on the facial motion data by determining movement of a landmark point on the face and identify the avatar non-facial body animation based on the movement of the landmark, wherein the movement of the landmark is a user defined movement of the landmark, the user defined movement selected from a list of available landmark movements; and
animating a body of an avatar using the avatar non-facial body animation, wherein the body of the avatar includes a plurality of parts of the avatar connected to an avatar face.

20. The method of claim 19 , wherein identifying the avatar non-facial body animation includes determining a lack of movement of the face in the series of images and identifying the avatar non-facial body animation based on the lack of movement.

21. The method of claim 19 , wherein the avatar non-facial body animation is a user defined avatar body animation.

Assignees

• Intel Corp

Inventors

• Tong Xiaofeng
• Li Qiang Eric
• Du Yangzhou
• Li Wenlong
• Yip Johnny C

Classifications

• G06T13/40 Primary

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

• G06F3/012

  Head tracking input arrangements · CPC title

• G06T1/0007

  Image acquisition · CPC title

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title

• G06T2207/10004

  Still image; Photographic image · CPC title