Systems and methods of detecting body movements using globally generated multi-dimensional gesture data
First Claim
1. A method of identifying a movement of a subject based on data, the method comprising:
- receiving at a network interface, a stream of frames comprising gesture data, each frame of the stream of frames identifying positions of one or more body parts of a subject with respect to the waist of the subject'"'"'s body, the stream of frames being stored in storage memory;
extrapolating, by a processor, from the stream of frames comprising gesture data, one or more frames corresponding to a first movement;
assigning, by a classifier, the one or more frames to the first movement, the classifier applying a scale invariant feature transform is used to determine a descriptor of the first movement;
dividing, by the processor using a self-organizing map, the stream of frames into separate phases;
identifying, by the processor using a scalar vector machine, transition conditions within a feature space between phases;
receiving, by the network interface, a new gesture data identifying positions of one or more body parts of a new subject with respect to the waist of the new subject'"'"'s body;
extrapolating, by a processor, from the new gesture data one or more frames comprising gesture data identifying one or more features of the body movement of the new subject;
determining, by the processor, that movement of the new subject corresponds to the first movement responsive to comparing at least a portion of the new gesture data to at least a portion of the gesture data of the one or more frames corresponding to a first movement by applying at least both the descriptor provided by the scale invariant feature transform and the transition conditions identified by the scalar vector machine.
1 Assignment
0 Petitions
Accused Products
Abstract
The disclosure describes systems and methods of detecting body movements using gesture data. The gesture data may be self-referenced and may be comprised by frames which may identify locations or positions of body parts of a subject with respect to a particular reference point within the frame. A classifier may process frames to learn body movements and store the frames of gesture data in a database. Data comprising frames of self-referenced gesture data may be received by a recognizer which recognizes movements of the subject identified by the frames by matching gesture data of the incoming frames to the classified self-referenced gesture data stored in the database.
16 Citations
28 Claims
-
1. A method of identifying a movement of a subject based on data, the method comprising:
-
receiving at a network interface, a stream of frames comprising gesture data, each frame of the stream of frames identifying positions of one or more body parts of a subject with respect to the waist of the subject'"'"'s body, the stream of frames being stored in storage memory; extrapolating, by a processor, from the stream of frames comprising gesture data, one or more frames corresponding to a first movement; assigning, by a classifier, the one or more frames to the first movement, the classifier applying a scale invariant feature transform is used to determine a descriptor of the first movement; dividing, by the processor using a self-organizing map, the stream of frames into separate phases; identifying, by the processor using a scalar vector machine, transition conditions within a feature space between phases; receiving, by the network interface, a new gesture data identifying positions of one or more body parts of a new subject with respect to the waist of the new subject'"'"'s body; extrapolating, by a processor, from the new gesture data one or more frames comprising gesture data identifying one or more features of the body movement of the new subject; determining, by the processor, that movement of the new subject corresponds to the first movement responsive to comparing at least a portion of the new gesture data to at least a portion of the gesture data of the one or more frames corresponding to a first movement by applying at least both the descriptor provided by the scale invariant feature transform and the transition conditions identified by the scalar vector machine. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 26, 27, 28)
-
-
19. A system for identifying a movement of a subject based on data, the system comprising:
-
a database storing a stream of frames received at a network interface, each frame of the stream of frames comprising gesture data identifying positions of one or more body parts of a subject with respect to the waist of the subject'"'"'s body; a processor configured to extrapolate, from the the stream of frames comprising gesture data, one or more frames corresponding to a first movement, wherein the processor is configured to divide the stream of frames into separate phases using a self-organizing map, and wherein the processor is configured to identify, using a scalar vector machine, transition conditions within a feature space between phases; a classifier configured to assign the one or more frames to a first movement, the classifier configured to apply a scale invariant feature transform to determine a descriptor of the first movement; the network interface being configured to receive a new gesture data identifying positions of one or more body parts of a new subject with respect to the waist of the new subjects body; the processor configured to extrapolate from the new gesture data one or more frames comprising gesture data identifying one or more features of the body movement of the new subject; the processor further configured to determine that a movement of the new subject corresponds to the first movement responsive to comparing at least a portion of the new gesture data to the at least a portion of the new gesture data in the one or more frames of the stream of frames stored in the database. - View Dependent Claims (20, 21, 22, 23, 24, 25)
-
Specification