System and method for creating avatars or animated sequences using human body features extracted from a still image
OwnerAyla Mandel, David Mandel +1inventors
InventorsAyla Mandel, David Mandel, Stephen D. Williams
Life of this patent
Estimating the term…
Abstract
A user may create an avatar and/or animated sequence illustrating a particular object or living being performing a certain activity, using images of portions of the object or living being extracted from a still image or set of still images of the object or living being. A mathematical model used to represent the avatar may be animated according to user-selected motion information and may be modified according to various parameters including explicit end-user adjustments and information representative of a human emotion, mood, or feeling that may be derived from an image of the user or information from a news source or social network.
Claims24
A method of operating a system for creating an animated sequence of images from one or more still images of a particular object from a plurality of objects or a particular living being from a plurality of living beings, the system comprising at least one processor for communicatively coupling to a display device for displaying the animated sequence of images and to an image capture device for capturing a plurality of images of a user of the system, the method comprising:
- recognizing one or more portions of the particular object or the particular living being in corresponding image portions of the one or more still images, wherein recognizing each of the one or more portions of the particular object or the particular living being in the one or more still images associates with each corresponding image portion, reference points and parameters for underlying structural features of the recognized portion of the particular object or the particular living being that define location, dimensions, and constraints upon motion of the recognized portion relative to other recognized portions of the particular object or particular living being;
- extracting, by the at least one processor, the corresponding image portions of the particular object or the particular living being from the one or more still images;
- applying, by the at least one processor, the corresponding image portions to corresponding surface portions of a movable mathematical model created using physical characteristics of the plurality of objects or the plurality of living beings, according to the reference points and parameters associated with the corresponding image portions and corresponding reference points of the movable mathematical model;
- animating, by the at least one processor, the movable mathematical model according to motion template information selected by the user of the system, wherein the motion template information comprises a time ordered series of magnitude and direction information for the respective reference points of underlying structural features of each of the recognized portions of the particular object or the particular living being;
- rendering, by the at least one processor, one or more images from the movable mathematical model at corresponding positions according to the time ordered series of the motion template information selected by the user for the animation;
- and
- transmitting, to the display device by the at least one processor, the one or more rendered images for display.
A system for creating an animated sequence of images from one or more still images of a particular object from a plurality of objects or a particular living being from a plurality of living beings, the system comprising:
- at least one processor communicatively coupled to a display device, the at least one processor being operable to, at least:
- recognize one or more portions of the particular object or the particular living being in corresponding image portions of the one or more still images, wherein recognizing each of the one or more portions of the particular object or the particular living being in the one or more still images associates with each corresponding image portion, reference points and parameters for underlying structural features of the recognized portion of the particular object or the particular living being that define location, dimensions, and constraints upon motion of the recognized portion relative to other recognized portions of the particular object or particular living being;
- extract the corresponding image portions of the particular object or the particular living being from the one or more still images;
- apply the corresponding image portions to corresponding surface portions of a movable mathematical model created using physical characteristics of the plurality of objects or the plurality of living beings, according to the reference points and parameters associated with corresponding image portions and corresponding reference points of the movable mathematical model;
- animate the movable mathematical model according to motion template information selected by the user of the system, wherein the motion template information comprises a time ordered series of magnitude and direction information for the respective reference points of underlying structural features of each of the recognized portions of the particular object or the particular living being;
- render one or more images from the movable mathematical model at corresponding positions according to the time ordered series of the motion template information selected by the user for the animation;
- and
- transmit the one or more rendered images to the display device.
A non-transitory computer-readable medium having stored thereon a plurality of code sections, each code section comprising executable instructions, the plurality of code sections for creating an animated sequence of images from one or more still images of a particular object from a plurality of objects or a particular living being from a plurality of living beings by, at least:
- recognizing one or more portions of the particular object or the particular living being in corresponding image portions of the one or more still images, wherein recognizing each of the one or more portions of the particular object or the particular living being in the one or more still images associates with each corresponding image portion, reference points and parameters for underlying structural features of the recognized portion of the particular object or the particular living being that define location, dimensions, and constraints upon motion of the recognized portion relative to other recognized portions of the particular object or particular living being;
- extracting the corresponding image portions of the particular object or the particular living being from the one or more still images;
- applying the corresponding image portions to corresponding surface portions of a movable mathematical model created using physical characteristics of the plurality of objects or the plurality of living beings, according to the reference points and parameters associated with the corresponding image portions and corresponding reference points of the movable mathematical model;
- animating the movable mathematical model according to motion template information selected by the user, wherein the motion template information comprises a time ordered series of magnitude and direction information for the respective reference points of underlying structural features of each of the recognized portions of the particular object or the particular living being;
- rendering one or more images from the movable mathematical model at corresponding positions according to the time ordered series of the motion template information selected by the user for the animation;
- and
- transmitting the one or more rendered images to a display device.
Dependencies are inferred from the claim text.