Pose tracking pipeline
View Patent ↗A method of tracking a target includes receiving from a source a depth image of a scene including the human subject. The depth image includes a depth for each of a plurality of pixels. The method further includes identifying pixels of the depth image that belong to the human subject and deriving from the identified pixels of the depth image one or more machine readable data structures representing the human subject as a body model including a plurality of shapes.
1. A method of computer modeling a human subject, the method comprising:
receiving from a source a depth image of a scene including the human subject, the depth image including a depth for each of a plurality of pixels;
identifying pixels of the depth image that belong to the human subject; and
deriving from the identified pixels of the depth image one or more machine readable data structures representing the human subject as a body model including a plurality of shapes.
2. The method of claim 1 , where the plurality of shapes include a plurality of spheres.
3. The method of claim 2 , where each of the plurality of spheres is represented by a position variable defining a centroid position of that sphere.
4. The method of claim 3 , where each of the plurality of spheres is represented by a size variable defining a size of that sphere.
5. The method of claim 1 , where all of the plurality of shapes are a same shape and a same size.
6. The method of claim 1 , further comprising deriving from the body model a skeleton including a plurality of skeletal points, each skeletal point having a three-dimensional location.
7. The method of claim 1 , where the source includes a structured light depth sensor.
8. The method of claim 1 , where the source includes a time-of-flight depth sensor.
9. A method of computer modeling a human subject, the method comprising:
receiving from a source a depth image of a scene including the human subject, the depth image including a depth for each of a plurality of pixels; and
modeling the human subject with a plurality of shapes, each of the plurality of shapes characterized by a machine readable position variable, the machine readable position variable for each shape defining a position of that shape within a volume occupied by the human subject.
10. The method of claim 9 , where the plurality of shapes include a plurality of spheres.
11. The method of claim 10 , where the machine readable position variable for each of the plurality of spheres defines a centroid position of that sphere.
12. The method of claim 11 , where each of the plurality of spheres is represented by a size variable defining a size of that sphere.
13. The method of claim 9 , where all of the plurality of shapes are a same shape and a same size.
14. The method of claim 9 , further comprising deriving from the plurality of shapes a skeleton including a plurality of skeletal points, each skeletal point having a three-dimensional location.
15. The method of claim 9 , where the source includes a structured light depth sensor.
16. The method of claim 9 , where the source includes a time-of-flight depth sensor.
17. A data-holding subsystem holding instructions executable by a logic subsystem to:
receive from a source a depth image of a scene including the human subject; and
model the human subject with a plurality of spheres, each of the plurality of spheres characterized by a machine readable position variable, the machine readable position variable for each sphere defining a centroid position of that sphere such that the plurality of spheres collectively occupy a volume of the human subject.
18. The data-holding of claim 17 , where all of the plurality of spheres are a same shape and a same size.
19. The data-holding subsystem of claim 17 , where the source includes a structured light depth sensor.
20. The data-holding subsystem of claim 17 , where the source includes a time-of-flight depth sensor.